From df29d741b9bb78dd2a4c94219bb18a76a3630c94 Mon Sep 17 00:00:00 2001
From: Saurav Panda <sgp65@cornell.edu>
Date: Sun, 18 Aug 2024 11:15:18 -0700
Subject: [PATCH 01/19] feat: gracefully hanndling when max_token is none

---
 config.json             | 4 ++--
 kaizen/llms/provider.py | 9 +++++++--
 2 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/config.json b/config.json
index da96147e..e1e639da 100644
--- a/config.json
+++ b/config.json
@@ -50,8 +50,8 @@
                 "model_name": "best",
                 "litellm_params": {
                     "model": "azure/gpt-4o",
-                    "input_cost_per_token": 0.000000015,
-                    "output_cost_per_token": 0.0000006,
+                    "input_cost_per_token": 0.000005,
+                    "output_cost_per_token": 0.000015,
                     "api_key": "os.environ/AZURE_API_KEY",
                     "api_base": "os.environ/AZURE_API_BASE",
                     "base_model": "azure/gpt-4o"
diff --git a/kaizen/llms/provider.py b/kaizen/llms/provider.py
index 2e32af15..64fe4bb8 100644
--- a/kaizen/llms/provider.py
+++ b/kaizen/llms/provider.py
@@ -8,7 +8,7 @@
 from litellm import Router
 import logging
 from collections import defaultdict
-
+DEFAULT_MAX_TOKENS = 8000
 
 def set_all_loggers_to_ERROR():
     print("All Loggers and their levels:")
@@ -205,6 +205,8 @@ def is_inside_token_limit(self, PROMPT: str, percentage: float = 0.8) -> bool:
         ]
         token_count = litellm.token_counter(model=self.model, messages=messages)
         max_tokens = litellm.get_max_tokens(self.model)
+        if not max_tokens:
+            max_tokens = DEFAULT_MAX_TOKENS
         return token_count <= max_tokens * percentage
 
     def available_tokens(
@@ -214,7 +216,10 @@ def available_tokens(
             model = self.model
         max_tokens = litellm.get_max_tokens(model)
         used_tokens = litellm.token_counter(model=model, text=message)
-        return int(max_tokens * percentage) - used_tokens
+        if max_tokens:
+            return int(max_tokens * percentage) - used_tokens
+        else:
+            return DEFAULT_MAX_TOKENS - used_tokens
 
     def get_token_count(self, message: str, model: str = None) -> int:
         if not model:

From 5a562d5d306d0a6df2d1b4e0bda89883539da632 Mon Sep 17 00:00:00 2001
From: Saurav Panda <sgp65@cornell.edu>
Date: Sun, 18 Aug 2024 11:15:35 -0700
Subject: [PATCH 02/19] feat: updated poetry version

---
 pyproject.toml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pyproject.toml b/pyproject.toml
index 3d7dd138..d4a742e7 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "kaizen-cloudcode"
-version = "0.4.1"
+version = "0.4.2"
 description = "An intelligent coding companion that accelerates your development workflow by providing efficient assistance, enabling you to craft high-quality code more rapidly."
 authors = ["Saurav Panda <saurav.panda@cloudcode.ai>"]
 license = "Apache2.0"

From 65b39858991e99365f51d32bea7967fe86434fb8 Mon Sep 17 00:00:00 2001
From: Saurav Panda <sgp65@cornell.edu>
Date: Sun, 18 Aug 2024 11:17:22 -0700
Subject: [PATCH 03/19] feat: linting fixes

---
 .experiments/code_review/main.py           |  4 ++--
 kaizen/formatters/code_review_formatter.py |  6 ++----
 kaizen/llms/provider.py                    |  2 ++
 kaizen/retriever/custom_vector_store.py    | 21 ++++++++++++++-------
 kaizen/retriever/llama_index_retriever.py  |  9 +++++++--
 5 files changed, 27 insertions(+), 15 deletions(-)

diff --git a/.experiments/code_review/main.py b/.experiments/code_review/main.py
index ef17342a..28375d34 100644
--- a/.experiments/code_review/main.py
+++ b/.experiments/code_review/main.py
@@ -43,7 +43,7 @@ def process_pr(pr_url, reeval_response=False):
         pull_request_files=pr_files,
         user="kaizen/example",
         reeval_response=reeval_response,
-        model="best"
+        model="best",
     )
 
     # topics = clean_keys(review_data.topics, "important")
@@ -104,7 +104,7 @@ def main(pr_urls):
     pr_urls = [
         "https://github.com/Cloud-Code-AI/kaizen/pull/335",
         "https://github.com/Cloud-Code-AI/kaizen/pull/440",
-        "https://github.com/Cloud-Code-AI/kaizen/pull/222"
+        "https://github.com/Cloud-Code-AI/kaizen/pull/222",
         # Add more PR URLs here
     ]
     main(pr_urls)
diff --git a/kaizen/formatters/code_review_formatter.py b/kaizen/formatters/code_review_formatter.py
index bbe1cfb4..e6a37aaa 100644
--- a/kaizen/formatters/code_review_formatter.py
+++ b/kaizen/formatters/code_review_formatter.py
@@ -95,9 +95,7 @@ def create_stats_section(reviews: List[Dict]) -> str:
     important_issues = sum(
         1 for review in reviews if review["confidence"] == "important"
     )
-    minor_issues = sum(
-        1 for review in reviews if review["confidence"] in ["moderate"]
-    )
+    minor_issues = sum(1 for review in reviews if review["confidence"] in ["moderate"])
     files_affected = len(set(review["file_name"] for review in reviews))
 
     output = "## 📊 Stats\n"
@@ -124,7 +122,7 @@ def create_issue_section(issue: Dict, index: int) -> str:
     output += f"⚖️ **Severity:** {issue['severity_level']}/10\n"
     output += f"🔍 **Description:** {issue['reason']}\n"
     output += f"💡 **Solution:** {issue['solution']}\n\n"
-    if issue.get('actual_code', None) or issue.get('fixed_code', ''):
+    if issue.get("actual_code", None) or issue.get("fixed_code", ""):
         output += "**Current Code:**\n"
         output += f"```python\n{issue.get('actual_code', '')}\n```\n\n"
         output += "**Suggested Code:**\n"
diff --git a/kaizen/llms/provider.py b/kaizen/llms/provider.py
index 64fe4bb8..cd16355d 100644
--- a/kaizen/llms/provider.py
+++ b/kaizen/llms/provider.py
@@ -8,8 +8,10 @@
 from litellm import Router
 import logging
 from collections import defaultdict
+
 DEFAULT_MAX_TOKENS = 8000
 
+
 def set_all_loggers_to_ERROR():
     print("All Loggers and their levels:")
     for name, logger in logging.Logger.manager.loggerDict.items():
diff --git a/kaizen/retriever/custom_vector_store.py b/kaizen/retriever/custom_vector_store.py
index f8f6ebc7..dfb2f061 100644
--- a/kaizen/retriever/custom_vector_store.py
+++ b/kaizen/retriever/custom_vector_store.py
@@ -8,13 +8,17 @@ class CustomPGVectorStore(PGVectorStore):
     def __init__(self, *args, **kwargs):
         super().__init__(*args, **kwargs)
         # Store the table name in a new attribute
-        self.table_name = kwargs.get('table_name', 'embeddings')
+        self.table_name = kwargs.get("table_name", "embeddings")
 
-    def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:
+    def custom_query(
+        self, query_embedding: List[float], repo_id: int, similarity_top_k: int
+    ) -> List[dict]:
         # Normalize the query embedding
         query_embedding_np = np.array(query_embedding)
-        query_embedding_normalized = query_embedding_np / np.linalg.norm(query_embedding_np)
-        
+        query_embedding_normalized = query_embedding_np / np.linalg.norm(
+            query_embedding_np
+        )
+
         # SQL query with repo_id filter and cosine similarity
         query = f"""
         SELECT 
@@ -35,10 +39,13 @@ def custom_query(self, query_embedding: List[float], repo_id: int, similarity_to
         LIMIT 
             %s
         """
-        
+
         with self.get_client() as client:
             with client.cursor() as cur:
-                cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))
+                cur.execute(
+                    query,
+                    (query_embedding_normalized.tolist(), repo_id, similarity_top_k),
+                )
                 results = cur.fetchall()
 
         return [
@@ -46,7 +53,7 @@ def custom_query(self, query_embedding: List[float], repo_id: int, similarity_to
                 "id": row[0],
                 "text": row[1],
                 "metadata": row[2] if isinstance(row[2], dict) else Json(row[2]),
-                "similarity": row[3]
+                "similarity": row[3],
             }
             for row in results
         ]
diff --git a/kaizen/retriever/llama_index_retriever.py b/kaizen/retriever/llama_index_retriever.py
index 5811f353..46e020dd 100644
--- a/kaizen/retriever/llama_index_retriever.py
+++ b/kaizen/retriever/llama_index_retriever.py
@@ -174,7 +174,12 @@ def store_abstraction_and_embedding(self, function_id: int, abstraction: str):
         # Create a TextNode for the vector store
         # Include repo_id in the metadata
         metadata = {"repo_id": self.repo_id}
-        node = TextNode(text=abstraction, id_=str(function_id), embedding=embedding, metadata=metadata)
+        node = TextNode(
+            text=abstraction,
+            id_=str(function_id),
+            embedding=embedding,
+            metadata=metadata,
+        )
 
         # Add the node to the vector store
         self.vector_store.add(nodes=[node])
@@ -350,7 +355,7 @@ def query(self, query_text: str, num_results: int = 5) -> List[Dict[str, Any]]:
 
         query_bundle = QueryBundle(query_str=query_text, embedding=embedding)
         retriever = index.as_retriever(similarity_top_k=num_results)
-        
+
         # Apply the filter during retrieval
         nodes = retriever.retrieve(query_bundle)  # Add potential filtering
 

From c00a5ff9ae8e2847ae554db70b0637aa453c54e7 Mon Sep 17 00:00:00 2001
From: Saurav Panda <sgp65@cornell.edu>
Date: Sun, 18 Aug 2024 11:22:06 -0700
Subject: [PATCH 04/19] feat: added pre commit

---
 .pre-commit-config.yaml | 12 ++++++
 poetry.lock             | 87 ++++++++++++++++++++++++++++++++++++++++-
 pyproject.toml          |  1 +
 3 files changed, 99 insertions(+), 1 deletion(-)
 create mode 100644 .pre-commit-config.yaml

diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
new file mode 100644
index 00000000..748c8157
--- /dev/null
+++ b/.pre-commit-config.yaml
@@ -0,0 +1,12 @@
+repos:
+  - repo: https://github.com/PyCQA/flake8
+    rev: 6.0.0
+    hooks:
+      - id: flake8
+        args: [--config=.flake8]
+
+  - repo: https://github.com/psf/black
+    rev: 23.3.0
+    hooks:
+      - id: black
+        args: [--line-length=88]
\ No newline at end of file
diff --git a/poetry.lock b/poetry.lock
index 91059500..1b3c4d2b 100644
--- a/poetry.lock
+++ b/poetry.lock
@@ -398,6 +398,17 @@ files = [
 [package.dependencies]
 pycparser = "*"
 
+[[package]]
+name = "cfgv"
+version = "3.4.0"
+description = "Validate configuration and produce human readable error messages."
+optional = false
+python-versions = ">=3.8"
+files = [
+    {file = "cfgv-3.4.0-py2.py3-none-any.whl", hash = "sha256:b7265b1f29fd3316bfcd2b330d63d024f2bfd8bcb8b0272f8e19a504856c48f9"},
+    {file = "cfgv-3.4.0.tar.gz", hash = "sha256:e52591d4c5f5dead8e0f673fb16db7949d2cfb3f7da4582893288f0ded8fe560"},
+]
+
 [[package]]
 name = "charset-normalizer"
 version = "3.3.2"
@@ -619,6 +630,17 @@ files = [
     {file = "dirtyjson-1.0.8.tar.gz", hash = "sha256:90ca4a18f3ff30ce849d100dcf4a003953c79d3a2348ef056f1d9c22231a25fd"},
 ]
 
+[[package]]
+name = "distlib"
+version = "0.3.8"
+description = "Distribution utilities"
+optional = false
+python-versions = "*"
+files = [
+    {file = "distlib-0.3.8-py2.py3-none-any.whl", hash = "sha256:034db59a0b96f8ca18035f36290806a9a6e6bd9d1ff91e45a7f172eb17e51784"},
+    {file = "distlib-0.3.8.tar.gz", hash = "sha256:1530ea13e350031b6312d8580ddb6b27a104275a31106523b8f123787f494f64"},
+]
+
 [[package]]
 name = "distro"
 version = "1.9.0"
@@ -1028,6 +1050,20 @@ testing = ["InquirerPy (==0.3.4)", "Jinja2", "Pillow", "aiohttp", "fastapi", "gr
 torch = ["safetensors[torch]", "torch"]
 typing = ["types-PyYAML", "types-requests", "types-simplejson", "types-toml", "types-tqdm", "types-urllib3", "typing-extensions (>=4.8.0)"]
 
+[[package]]
+name = "identify"
+version = "2.6.0"
+description = "File identification library for Python"
+optional = false
+python-versions = ">=3.8"
+files = [
+    {file = "identify-2.6.0-py2.py3-none-any.whl", hash = "sha256:e79ae4406387a9d300332b5fd366d8994f1525e8414984e1a59e058b2eda2dd0"},
+    {file = "identify-2.6.0.tar.gz", hash = "sha256:cb171c685bdc31bcc4c1734698736a7d5b6c8bf2e0c15117f4d469c8640ae5cf"},
+]
+
+[package.extras]
+license = ["ukkonen"]
+
 [[package]]
 name = "idna"
 version = "3.7"
@@ -1827,6 +1863,17 @@ plot = ["matplotlib"]
 tgrep = ["pyparsing"]
 twitter = ["twython"]
 
+[[package]]
+name = "nodeenv"
+version = "1.9.1"
+description = "Node.js virtual environment builder"
+optional = false
+python-versions = "!=3.0.*,!=3.1.*,!=3.2.*,!=3.3.*,!=3.4.*,!=3.5.*,!=3.6.*,>=2.7"
+files = [
+    {file = "nodeenv-1.9.1-py2.py3-none-any.whl", hash = "sha256:ba11c9782d29c27c70ffbdda2d7415098754709be8a7056d79a737cd901155c9"},
+    {file = "nodeenv-1.9.1.tar.gz", hash = "sha256:6ec12890a2dab7946721edbfbcd91f3319c6ccc9aec47be7c7e6b7011ee6645f"},
+]
+
 [[package]]
 name = "numpy"
 version = "1.26.4"
@@ -2152,6 +2199,24 @@ files = [
 dev = ["pre-commit", "tox"]
 testing = ["pytest", "pytest-benchmark"]
 
+[[package]]
+name = "pre-commit"
+version = "3.8.0"
+description = "A framework for managing and maintaining multi-language pre-commit hooks."
+optional = false
+python-versions = ">=3.9"
+files = [
+    {file = "pre_commit-3.8.0-py2.py3-none-any.whl", hash = "sha256:9a90a53bf82fdd8778d58085faf8d83df56e40dfe18f45b19446e26bf1b3a63f"},
+    {file = "pre_commit-3.8.0.tar.gz", hash = "sha256:8bb6494d4a20423842e198980c9ecf9f96607a07ea29549e180eef9ae80fe7af"},
+]
+
+[package.dependencies]
+cfgv = ">=2.0.0"
+identify = ">=1.0.0"
+nodeenv = ">=0.11.1"
+pyyaml = ">=5.1"
+virtualenv = ">=20.10.0"
+
 [[package]]
 name = "psycopg2-binary"
 version = "2.9.9"
@@ -3439,6 +3504,26 @@ typing-extensions = {version = ">=4.0", markers = "python_version < \"3.11\""}
 [package.extras]
 standard = ["colorama (>=0.4)", "httptools (>=0.5.0)", "python-dotenv (>=0.13)", "pyyaml (>=5.1)", "uvloop (>=0.14.0,!=0.15.0,!=0.15.1)", "watchfiles (>=0.13)", "websockets (>=10.4)"]
 
+[[package]]
+name = "virtualenv"
+version = "20.26.3"
+description = "Virtual Python Environment builder"
+optional = false
+python-versions = ">=3.7"
+files = [
+    {file = "virtualenv-20.26.3-py3-none-any.whl", hash = "sha256:8cc4a31139e796e9a7de2cd5cf2489de1217193116a8fd42328f1bd65f434589"},
+    {file = "virtualenv-20.26.3.tar.gz", hash = "sha256:4c43a2a236279d9ea36a0d76f98d84bd6ca94ac4e0f4a3b9d46d05e10fea542a"},
+]
+
+[package.dependencies]
+distlib = ">=0.3.7,<1"
+filelock = ">=3.12.2,<4"
+platformdirs = ">=3.9.1,<5"
+
+[package.extras]
+docs = ["furo (>=2023.7.26)", "proselint (>=0.13)", "sphinx (>=7.1.2,!=7.3)", "sphinx-argparse (>=0.4)", "sphinxcontrib-towncrier (>=0.2.1a0)", "towncrier (>=23.6)"]
+test = ["covdefaults (>=2.3)", "coverage (>=7.2.7)", "coverage-enable-subprocess (>=1)", "flaky (>=3.7)", "packaging (>=23.1)", "pytest (>=7.4)", "pytest-env (>=0.8.2)", "pytest-freezer (>=0.4.8)", "pytest-mock (>=3.11.1)", "pytest-randomly (>=3.12)", "pytest-timeout (>=2.1)", "setuptools (>=68)", "time-machine (>=2.10)"]
+
 [[package]]
 name = "wrapt"
 version = "1.16.0"
@@ -3646,4 +3731,4 @@ typescript = ["tree-sitter-typescript"]
 [metadata]
 lock-version = "2.0"
 python-versions = "^3.9.0"
-content-hash = "d5729e5e9a46bb7b0a7a3bf2f098dcb9d047f35a51aa418f4cef1e4bda8aff40"
+content-hash = "19b210560f02e7e1b4210a4cca0e50df1f6669e3b9586e24ca20a39265170490"
diff --git a/pyproject.toml b/pyproject.toml
index d4a742e7..8649fc15 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -40,6 +40,7 @@ tree-sitter-typescript = "^0.21.2"
 tree-sitter-rust = "^0.21.2"
 llama-index-llms-litellm = "^0.1.4"
 llama-index-embeddings-litellm = "^0.1.1"
+pre-commit = "^3.8.0"
 
 [tool.poetry.extras]
 python = ["tree-sitter-python"]

From e6e6335ab2c6011416702a8ec95977504f785895 Mon Sep 17 00:00:00 2001
From: Saurav Panda <sgp65@cornell.edu>
Date: Mon, 19 Aug 2024 14:55:33 -0700
Subject: [PATCH 05/19] feat: updated file generator script

---
 kaizen/reviewer/code_review.py | 6 +++---
 pyproject.toml                 | 2 +-
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/kaizen/reviewer/code_review.py b/kaizen/reviewer/code_review.py
index 9cae8732..9e48fa61 100644
--- a/kaizen/reviewer/code_review.py
+++ b/kaizen/reviewer/code_review.py
@@ -196,13 +196,14 @@ def _process_files(
         self.logger.debug("Processing based on files")
         reviews = []
         code_quality = None
-        for file_review, quality in self._process_files_generator(
+        file_chunks_generator = self._process_files_generator(
             pull_request_files,
             pull_request_title,
             pull_request_desc,
             user,
             reeval_response,
-        ):
+        )
+        for file_review, quality in file_chunks_generator:
             reviews.extend(file_review)
             if quality:
                 if code_quality and code_quality > quality:
@@ -221,7 +222,6 @@ def _process_files_generator(
     ) -> Generator[List[Dict], None, None]:
         combined_diff_data = ""
         available_tokens = self.provider.available_tokens(FILE_CODE_REVIEW_PROMPT)
-
         for file in pull_request_files:
             patch_details = file.get("patch")
             filename = file.get("filename", "").replace(" ", "")
diff --git a/pyproject.toml b/pyproject.toml
index 8649fc15..591323ec 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "kaizen-cloudcode"
-version = "0.4.2"
+version = "0.4.3"
 description = "An intelligent coding companion that accelerates your development workflow by providing efficient assistance, enabling you to craft high-quality code more rapidly."
 authors = ["Saurav Panda <saurav.panda@cloudcode.ai>"]
 license = "Apache2.0"

From 7967b74442c897c55122d5103dfd6d0354fb8fb5 Mon Sep 17 00:00:00 2001
From: Saurav Panda <sgp65@cornell.edu>
Date: Mon, 19 Aug 2024 15:26:08 -0700
Subject: [PATCH 06/19] bug: understanding of removal data

---
 kaizen/helpers/parser.py       |  5 +++--
 kaizen/reviewer/code_review.py | 15 +++++++++------
 pyproject.toml                 |  2 +-
 3 files changed, 13 insertions(+), 9 deletions(-)

diff --git a/kaizen/helpers/parser.py b/kaizen/helpers/parser.py
index 7cde39e0..b2dcd170 100644
--- a/kaizen/helpers/parser.py
+++ b/kaizen/helpers/parser.py
@@ -185,7 +185,7 @@ def format_change(old_num, new_num, change_type, content):
     return f"{old_num_str} {new_num_str} {change_type} {content}"
 
 
-def patch_to_combined_chunks(patch_text):
+def patch_to_combined_chunks(patch_text, ignore_deletions=False):
     lines = patch_text.split("\n")
     changes = []
     metadata = []
@@ -234,7 +234,8 @@ def patch_to_combined_chunks(patch_text):
             current_file_name = line
         elif line.startswith("-"):
             content = line[1:]
-            changes.append(format_change(removal_line_num, None, "-1:[-]", content))
+            if not ignore_deletions:
+                changes.append(format_change(removal_line_num, None, "-1:[-]", content))
             removal_line_num += 1
             unedited_removal_num = removal_line_num
         elif line.startswith("+"):
diff --git a/kaizen/reviewer/code_review.py b/kaizen/reviewer/code_review.py
index 9e48fa61..47dea566 100644
--- a/kaizen/reviewer/code_review.py
+++ b/kaizen/reviewer/code_review.py
@@ -105,6 +105,7 @@ def __init__(self, llm_provider: LLMProvider, default_model="default"):
             "completion_tokens": 0,
             "total_tokens": 0,
         }
+        self.ignore_deletions = False
 
     def is_code_review_prompt_within_limit(
         self,
@@ -115,7 +116,9 @@ def is_code_review_prompt_within_limit(
         prompt = CODE_REVIEW_PROMPT.format(
             PULL_REQUEST_TITLE=pull_request_title,
             PULL_REQUEST_DESC=pull_request_desc,
-            CODE_DIFF=parser.patch_to_combined_chunks(diff_text),
+            CODE_DIFF=parser.patch_to_combined_chunks(
+                diff_text, ignore_deletions=self.ignore_deletions
+            ),
         )
         return self.provider.is_inside_token_limit(PROMPT=prompt)
 
@@ -128,9 +131,11 @@ def review_pull_request(
         user: Optional[str] = None,
         reeval_response: bool = False,
         model="default",
+        ignore_deletions=False,
     ) -> ReviewOutput:
+        self.ignore_deletions = ignore_deletions
         prompt = CODE_REVIEW_PROMPT.format(
-            CODE_DIFF=parser.patch_to_combined_chunks(diff_text),
+            CODE_DIFF=parser.patch_to_combined_chunks(diff_text, self.ignore_deletions),
         )
         self.total_usage = {
             "prompt_tokens": 0,
@@ -232,7 +237,7 @@ def _process_files_generator(
             ):
                 temp_prompt = (
                     combined_diff_data
-                    + f"\n---->\nFile Name: {filename}\nPatch Details: {parser.patch_to_combined_chunks(patch_details)}"
+                    + f"\n---->\nFile Name: {filename}\nPatch Details: {parser.patch_to_combined_chunks(patch_details, self.ignore_deletions)}"
                 )
 
                 if available_tokens - self.provider.get_token_count(temp_prompt) > 0:
@@ -246,9 +251,7 @@ def _process_files_generator(
                     user,
                     reeval_response,
                 )
-                combined_diff_data = (
-                    f"\n---->\nFile Name: {filename}\nPatch Details: {patch_details}"
-                )
+                combined_diff_data = f"\n---->\nFile Name: {filename}\nPatch Details: {parser.patch_to_combined_chunks(patch_details,  self.ignore_deletions)}"
 
         yield self._process_file_chunk(
             combined_diff_data,
diff --git a/pyproject.toml b/pyproject.toml
index 591323ec..40efde48 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "kaizen-cloudcode"
-version = "0.4.3"
+version = "0.4.4"
 description = "An intelligent coding companion that accelerates your development workflow by providing efficient assistance, enabling you to craft high-quality code more rapidly."
 authors = ["Saurav Panda <saurav.panda@cloudcode.ai>"]
 license = "Apache2.0"

From cf48112a6bc7fd518174ffd30be24e803629afb4 Mon Sep 17 00:00:00 2001
From: Swapnil <swapnilchand51@gmail.com>
Date: Tue, 20 Aug 2024 14:24:05 +0530
Subject: [PATCH 07/19] Added raw_chat_completions() in llm.provider

---
 kaizen/llms/provider.py                       | 26 +++++++++++++++++++
 {tests => kaizen/tests}/__init__.py           |  0
 .../tests}/actions/diff_pr_test.py            |  0
 .../tests}/actions/test_review.py             |  0
 .../tests}/data/actions/valid_review.json     |  0
 .../tests}/helpers/test_output.py             |  0
 .../tests}/helpers/test_patch_parser.py       |  0
 {tests => kaizen/tests}/llms/test_provider.py |  0
 .../tests}/retriever/test_chunker.py          |  0
 9 files changed, 26 insertions(+)
 rename {tests => kaizen/tests}/__init__.py (100%)
 rename {tests => kaizen/tests}/actions/diff_pr_test.py (100%)
 rename {tests => kaizen/tests}/actions/test_review.py (100%)
 rename {tests => kaizen/tests}/data/actions/valid_review.json (100%)
 rename {tests => kaizen/tests}/helpers/test_output.py (100%)
 rename {tests => kaizen/tests}/helpers/test_patch_parser.py (100%)
 mode change 100755 => 100644
 rename {tests => kaizen/tests}/llms/test_provider.py (100%)
 rename {tests => kaizen/tests}/retriever/test_chunker.py (100%)

diff --git a/kaizen/llms/provider.py b/kaizen/llms/provider.py
index cd16355d..4567d4ef 100644
--- a/kaizen/llms/provider.py
+++ b/kaizen/llms/provider.py
@@ -162,6 +162,32 @@ def chat_completion(
         self.model = response["model"]
         return response["choices"][0]["message"]["content"], response["usage"]
 
+    def raw_chat_completion(
+        self,
+        prompt,
+        user: str = None,
+        model="default",
+        custom_model=None,
+        messages=None,
+        n_choices=1
+    ):
+        custom_model["n"] = n_choices
+        if not messages:
+            messages = [
+                {"role": "system", "content": self.system_prompt},
+                {"role": "user", "content": prompt},
+            ]
+        if not custom_model:
+            custom_model = {"model": model}
+        if "temperature" not in custom_model:
+            custom_model["temperature"] = self.default_temperature
+
+        response = self.provider.completion(
+            messages=messages, user=user, **custom_model
+        )
+        self.model = response["model"]
+        return response, response["usage"]
+
     @retry(max_attempts=3, delay=1)
     def chat_completion_with_json(
         self,
diff --git a/tests/__init__.py b/kaizen/tests/__init__.py
similarity index 100%
rename from tests/__init__.py
rename to kaizen/tests/__init__.py
diff --git a/tests/actions/diff_pr_test.py b/kaizen/tests/actions/diff_pr_test.py
similarity index 100%
rename from tests/actions/diff_pr_test.py
rename to kaizen/tests/actions/diff_pr_test.py
diff --git a/tests/actions/test_review.py b/kaizen/tests/actions/test_review.py
similarity index 100%
rename from tests/actions/test_review.py
rename to kaizen/tests/actions/test_review.py
diff --git a/tests/data/actions/valid_review.json b/kaizen/tests/data/actions/valid_review.json
similarity index 100%
rename from tests/data/actions/valid_review.json
rename to kaizen/tests/data/actions/valid_review.json
diff --git a/tests/helpers/test_output.py b/kaizen/tests/helpers/test_output.py
similarity index 100%
rename from tests/helpers/test_output.py
rename to kaizen/tests/helpers/test_output.py
diff --git a/tests/helpers/test_patch_parser.py b/kaizen/tests/helpers/test_patch_parser.py
old mode 100755
new mode 100644
similarity index 100%
rename from tests/helpers/test_patch_parser.py
rename to kaizen/tests/helpers/test_patch_parser.py
diff --git a/tests/llms/test_provider.py b/kaizen/tests/llms/test_provider.py
similarity index 100%
rename from tests/llms/test_provider.py
rename to kaizen/tests/llms/test_provider.py
diff --git a/tests/retriever/test_chunker.py b/kaizen/tests/retriever/test_chunker.py
similarity index 100%
rename from tests/retriever/test_chunker.py
rename to kaizen/tests/retriever/test_chunker.py

From f562e750537d06a26e23d799559d76ef97385cb5 Mon Sep 17 00:00:00 2001
From: Saurav Panda <saurav.panda@cloudcode.ai>
Date: Wed, 21 Aug 2024 00:14:28 -0400
Subject: [PATCH 08/19] Update README.md

---
 README.md | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/README.md b/README.md
index 624eba2e..e9038740 100644
--- a/README.md
+++ b/README.md
@@ -2,7 +2,7 @@
   <img src="/assets/logo.png" alt="Kaizen Logo" width="200"/>
 </p>
 
-<h1 align="center">Kaizen: The Ultimate Code Quality Guardian</h1>
+<h1 align="center">Accelarating Bug Detection</h1>
 
 <p align="center">
   <strong>Unleash the power of AI to find and squash bugs before they reach your customers.</strong>
@@ -153,4 +153,4 @@ Need help or have questions? Reach out to us at support@cloudcode.ai.
 
 <p align="center">
   Made with ❤️ by the Kaizen team
-</p>
\ No newline at end of file
+</p>

From 8ade2693659ca9f9be0152bb60bf7d4c274f4592 Mon Sep 17 00:00:00 2001
From: Saurav Panda <sgp65@cornell.edu>
Date: Thu, 22 Aug 2024 19:10:03 -0700
Subject: [PATCH 09/19] feat: added support for running single file tests

---
 examples/unittest/main.py         | 4 ++++
 kaizen/actors/unit_test_runner.py | 5 ++++-
 kaizen/generator/unit_test.py     | 4 ++--
 kaizen/llms/provider.py           | 2 +-
 4 files changed, 11 insertions(+), 4 deletions(-)

diff --git a/examples/unittest/main.py b/examples/unittest/main.py
index 7e586356..23b14af6 100644
--- a/examples/unittest/main.py
+++ b/examples/unittest/main.py
@@ -17,8 +17,12 @@
 )
 print(result)
 
+# Run all tests
 test_results = generator.run_tests()
 
+# Run a single test file:
+# test_results = generator.run_tests(file_path="test_create_folder.py")
+
 for file_path, result in test_results.items():
     print(f"Results for {file_path}:")
     if "error" in result:
diff --git a/kaizen/actors/unit_test_runner.py b/kaizen/actors/unit_test_runner.py
index 5c0b50aa..3e6ace9d 100644
--- a/kaizen/actors/unit_test_runner.py
+++ b/kaizen/actors/unit_test_runner.py
@@ -70,7 +70,7 @@ def find_project_root(self, file_path):
         self.logger.warning("Project root not found")
         return None
 
-    def discover_and_run_tests(self):
+    def discover_and_run_tests(self, test_file=None):
         self.logger.info("Starting test discovery and execution")
         results = {}
         for root, dirs, files in os.walk(self.test_directory):
@@ -80,6 +80,9 @@ def discover_and_run_tests(self):
                     extension = file.split(".")[-1]
                     self.logger.debug(f"Found test file: {file_path}")
                     if extension in self.supported_extensions:
+                        if file_path and file not in test_file:
+                            self.logger.debug("Skipping file test")
+                            continue
                         self.logger.info(f"Running tests for: {file_path}")
                         result = self.supported_extensions[extension](file_path)
                         results[str(file_path)] = result
diff --git a/kaizen/generator/unit_test.py b/kaizen/generator/unit_test.py
index 250de3d6..9e414d12 100644
--- a/kaizen/generator/unit_test.py
+++ b/kaizen/generator/unit_test.py
@@ -277,9 +277,9 @@ def generate_tests_with_feedback(self, test_code, feedback):
     def _create_output_folder(self, folder_name):
         os.makedirs(folder_name, exist_ok=True)
 
-    def run_tests(self):
+    def run_tests(self, test_file=None):
         runner = UnitTestRunner(self.output_folder)
-        return runner.discover_and_run_tests()
+        return runner.discover_and_run_tests(test_file=test_file)
 
     def format_test_scenarios(self, scenarios):
         formatted_scenarios = ""
diff --git a/kaizen/llms/provider.py b/kaizen/llms/provider.py
index 4567d4ef..e10c8a0f 100644
--- a/kaizen/llms/provider.py
+++ b/kaizen/llms/provider.py
@@ -169,7 +169,7 @@ def raw_chat_completion(
         model="default",
         custom_model=None,
         messages=None,
-        n_choices=1
+        n_choices=1,
     ):
         custom_model["n"] = n_choices
         if not messages:

From bc9bef45e8850ae390b239a942c28715283b14a4 Mon Sep 17 00:00:00 2001
From: Saurav Panda <sgp65@cornell.edu>
Date: Thu, 22 Aug 2024 19:15:10 -0700
Subject: [PATCH 10/19] fix: kaizen suggested bug fix

---
 kaizen/actors/unit_test_runner.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kaizen/actors/unit_test_runner.py b/kaizen/actors/unit_test_runner.py
index 3e6ace9d..7864baf8 100644
--- a/kaizen/actors/unit_test_runner.py
+++ b/kaizen/actors/unit_test_runner.py
@@ -80,7 +80,7 @@ def discover_and_run_tests(self, test_file=None):
                     extension = file.split(".")[-1]
                     self.logger.debug(f"Found test file: {file_path}")
                     if extension in self.supported_extensions:
-                        if file_path and file not in test_file:
+                        if test_file and file not in test_file:
                             self.logger.debug("Skipping file test")
                             continue
                         self.logger.info(f"Running tests for: {file_path}")

From db3dab8ff7dcd213872648ad6f6f5907fed4576e Mon Sep 17 00:00:00 2001
From: Saurav Panda <saurav.panda@cloudcode.ai>
Date: Thu, 22 Aug 2024 22:20:37 -0400
Subject: [PATCH 11/19] Update kaizen/actors/unit_test_runner.py

Co-authored-by: kaizen-bot[bot] <150987473+kaizen-bot[bot]@users.noreply.github.com>
---
 kaizen/actors/unit_test_runner.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/kaizen/actors/unit_test_runner.py b/kaizen/actors/unit_test_runner.py
index 7864baf8..982fa493 100644
--- a/kaizen/actors/unit_test_runner.py
+++ b/kaizen/actors/unit_test_runner.py
@@ -71,6 +71,8 @@ def find_project_root(self, file_path):
         return None
 
     def discover_and_run_tests(self, test_file=None):
+        if test_file is None:
+            self.logger.warning('No test file specified. Running all tests.')
         self.logger.info("Starting test discovery and execution")
         results = {}
         for root, dirs, files in os.walk(self.test_directory):

From 2a33d37d345a6a727104f997454bd46d8bb7e569 Mon Sep 17 00:00:00 2001
From: Saurav Panda <sgp65@cornell.edu>
Date: Thu, 22 Aug 2024 19:36:17 -0700
Subject: [PATCH 12/19] bugfix: code review issue when files are skipped

---
 examples/code_review/main.py   |  4 +++-
 kaizen/reviewer/code_review.py | 41 ++++++++++++++++++----------------
 pyproject.toml                 |  2 +-
 3 files changed, 26 insertions(+), 21 deletions(-)

diff --git a/examples/code_review/main.py b/examples/code_review/main.py
index 32794105..5c5e8903 100644
--- a/examples/code_review/main.py
+++ b/examples/code_review/main.py
@@ -33,7 +33,9 @@
 )
 
 topics = clean_keys(review_data.topics, "important")
-review_desc = create_pr_review_text(topics)
+review_desc = create_pr_review_text(
+    review_data.issues, code_quality=review_data.code_quality
+)
 comments, topics = create_review_comments(topics)
 
 print(f"Raw Topics: \n {json.dumps(topics, indent=2)}\n")
diff --git a/kaizen/reviewer/code_review.py b/kaizen/reviewer/code_review.py
index 47dea566..176bbbd7 100644
--- a/kaizen/reviewer/code_review.py
+++ b/kaizen/reviewer/code_review.py
@@ -1,4 +1,4 @@
-from typing import Optional, List, Dict, Generator
+from typing import Optional, List, Dict, Generator, Tuple
 from dataclasses import dataclass
 import logging
 from kaizen.helpers import parser
@@ -197,7 +197,7 @@ def _process_files(
         pull_request_desc: str,
         user: Optional[str],
         reeval_response: bool,
-    ) -> List[Dict]:
+    ) -> Tuple[List[Dict], Optional[float]]:
         self.logger.debug("Processing based on files")
         reviews = []
         code_quality = None
@@ -208,13 +208,13 @@ def _process_files(
             user,
             reeval_response,
         )
-        for file_review, quality in file_chunks_generator:
-            reviews.extend(file_review)
-            if quality:
-                if code_quality and code_quality > quality:
-                    code_quality = quality
-                else:
-                    code_quality = quality
+        for result in file_chunks_generator:
+            if result:  # Check if the result is not None
+                file_review, quality = result
+                reviews.extend(file_review)
+                if quality:
+                    if code_quality is None or quality < code_quality:
+                        code_quality = quality
         return reviews, code_quality
 
     def _process_files_generator(
@@ -224,7 +224,7 @@ def _process_files_generator(
         pull_request_desc: str,
         user: Optional[str],
         reeval_response: bool,
-    ) -> Generator[List[Dict], None, None]:
+    ) -> Generator[Optional[Tuple[List[Dict], Optional[float]]], None, None]:
         combined_diff_data = ""
         available_tokens = self.provider.available_tokens(FILE_CODE_REVIEW_PROMPT)
         for file in pull_request_files:
@@ -253,13 +253,16 @@ def _process_files_generator(
                 )
                 combined_diff_data = f"\n---->\nFile Name: {filename}\nPatch Details: {parser.patch_to_combined_chunks(patch_details,  self.ignore_deletions)}"
 
-        yield self._process_file_chunk(
-            combined_diff_data,
-            pull_request_title,
-            pull_request_desc,
-            user,
-            reeval_response,
-        )
+        if combined_diff_data:
+            yield self._process_file_chunk(
+                combined_diff_data,
+                pull_request_title,
+                pull_request_desc,
+                user,
+                reeval_response,
+            )
+        else:
+            yield None  # Yield None if there's no data to process
 
     def _process_file_chunk(
         self,
@@ -268,9 +271,9 @@ def _process_file_chunk(
         pull_request_desc: str,
         user: Optional[str],
         reeval_response: bool,
-    ) -> List[Dict]:
+    ) -> Optional[Tuple[List[Dict], Optional[float]]]:
         if not diff_data:
-            return []
+            return None
         prompt = FILE_CODE_REVIEW_PROMPT.format(
             FILE_PATCH=diff_data,
         )
diff --git a/pyproject.toml b/pyproject.toml
index 40efde48..273795bd 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "kaizen-cloudcode"
-version = "0.4.4"
+version = "0.4.5"
 description = "An intelligent coding companion that accelerates your development workflow by providing efficient assistance, enabling you to craft high-quality code more rapidly."
 authors = ["Saurav Panda <saurav.panda@cloudcode.ai>"]
 license = "Apache2.0"

From 2ab2f486d3f708a19282afd061a172a29223f85c Mon Sep 17 00:00:00 2001
From: Saurav Panda <sgp65@cornell.edu>
Date: Thu, 22 Aug 2024 21:46:39 -0700
Subject: [PATCH 13/19] feat: remove temp file path

---
 kaizen/actors/unit_test_runner.py | 2 +-
 kaizen/generator/unit_test.py     | 3 +++
 2 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/kaizen/actors/unit_test_runner.py b/kaizen/actors/unit_test_runner.py
index 982fa493..0a73454c 100644
--- a/kaizen/actors/unit_test_runner.py
+++ b/kaizen/actors/unit_test_runner.py
@@ -72,7 +72,7 @@ def find_project_root(self, file_path):
 
     def discover_and_run_tests(self, test_file=None):
         if test_file is None:
-            self.logger.warning('No test file specified. Running all tests.')
+            self.logger.warning("No test file specified. Running all tests.")
         self.logger.info("Starting test discovery and execution")
         results = {}
         for root, dirs, files in os.walk(self.test_directory):
diff --git a/kaizen/generator/unit_test.py b/kaizen/generator/unit_test.py
index 9e414d12..df024a63 100644
--- a/kaizen/generator/unit_test.py
+++ b/kaizen/generator/unit_test.py
@@ -119,11 +119,13 @@ def generate_tests(
         output_path: str = None,
         verbose: bool = False,
         enable_critique: bool = False,
+        temp_dir: str = "",
     ):
         self.max_critique = max_critique
         self.enable_critique = enable_critique
         self.verbose = verbose if verbose else self.verbose
         self.output_folder = output_path if output_path else self.output_folder
+        self.temp_dir = temp_dir
 
         file_extension = file_path.split(".")[-1]
         if file_extension not in self.SUPPORTED_LANGUAGES or file_extension == "pyc":
@@ -172,6 +174,7 @@ def _process_item(self, item, file_extension, file_path, folder_path):
         item["full_path"] = file_path
 
         test_code = self.generate_ai_tests(item, item["source"], file_extension)
+        test_code = test_code.replace(self.temp_dir, "")
 
         self._write_test_file(test_file_path, test_code)
 

From 00baf6979f8d199ea0b0382bdec01004a7a5b3d4 Mon Sep 17 00:00:00 2001
From: Saurav Panda <sgp65@cornell.edu>
Date: Sat, 24 Aug 2024 01:59:35 -0700
Subject: [PATCH 14/19] feat: moved code from pg vector to qdrant

---
 config.json                               |   5 +-
 docker-compose-dev.yml                    |  24 +-
 docker-compose.yml                        |  23 +-
 examples/ragify_codebase/main.py          |  18 +-
 kaizen/llms/provider.py                   |   1 +
 kaizen/retriever/custom_vector_store.py   |  59 ----
 kaizen/retriever/llama_index_retriever.py | 149 +++++----
 kaizen/retriever/qdrant_vector_store.py   |  26 ++
 poetry.lock                               | 351 ++++++++++++++++------
 pyproject.toml                            |   3 +-
 10 files changed, 436 insertions(+), 223 deletions(-)
 delete mode 100644 kaizen/retriever/custom_vector_store.py
 create mode 100644 kaizen/retriever/qdrant_vector_store.py

diff --git a/config.json b/config.json
index e1e639da..8aefb0fe 100644
--- a/config.json
+++ b/config.json
@@ -7,12 +7,11 @@
             {
                 "model_name": "embedding",
                 "litellm_params": {
-                    "model": "azure/text-embedding-small",
+                    "model": "azure/text-embedding-3-small",
                     "input_cost_per_token": 0.000000015,
                     "output_cost_per_token": 0.0000006,
                     "api_key": "os.environ/AZURE_API_KEY",
-                    "api_base": "os.environ/AZURE_API_BASE",
-                    "base_model": "text-embedding-3-small"
+                    "api_base": "os.environ/AZURE_API_BASE"
                 },
                 "model_info": {
                     "max_tokens": 8191,
diff --git a/docker-compose-dev.yml b/docker-compose-dev.yml
index 014bd59a..1f2a5bbc 100644
--- a/docker-compose-dev.yml
+++ b/docker-compose-dev.yml
@@ -14,6 +14,11 @@ services:
       - github_app_pem
     networks:
       - app-network
+    depends_on:
+      - redis
+      - postgres
+      - qdrant
+
   
   redis:
     image: "redis:alpine"
@@ -39,10 +44,27 @@ services:
     networks:
       - app-network
 
+  qdrant:
+    image: qdrant/qdrant:latest
+    ports:
+      - "6333:6333"
+      - "6334:6334"
+    volumes:
+      - qdrant_data:/qdrant/storage
+    environment:
+      - QDRANT__SERVICE__GRPC_PORT=6334
+    restart: always
+    networks:
+      - app-network
+
+volumes:
+  qdrant_data:
+    driver: local
+
 secrets:
   github_app_pem:
     file: ./GITHUB_APP_NIGHTLY.pem
   
 networks:
   app-network:
-    driver: bridge
\ No newline at end of file
+    driver: bridge
diff --git a/docker-compose.yml b/docker-compose.yml
index c9af20f7..4b249b8f 100644
--- a/docker-compose.yml
+++ b/docker-compose.yml
@@ -14,7 +14,9 @@ services:
       - github_app_pem
     depends_on:
       - redis
-  
+      - postgres
+      - qdrant
+
   postgres:
     image: postgres:16-bullseye
     env_file:
@@ -23,7 +25,7 @@ services:
       - ./init.sql:/docker-entrypoint-initdb.d/init.sql
     ports:
       - "5432:5432"
-  
+
   redis:
     image: "redis:alpine"
     environment:
@@ -31,8 +33,21 @@ services:
     ports:
       - "6379:6379"
 
+  qdrant:
+    image: qdrant/qdrant:latest
+    ports:
+      - "6333:6333"
+      - "6334:6334"
+    volumes:
+      - qdrant_data:/qdrant/storage
+    environment:
+      - QDRANT__SERVICE__GRPC_PORT=6334
+    restart: always
+
+volumes:
+  qdrant_data:
+    driver: local
 
 secrets:
   github_app_pem:
-    file: ./GITHUB_APP_NIGHTLY.pem
-
+    file: ./GITHUB_APP_NIGHTLY.pem
\ No newline at end of file
diff --git a/examples/ragify_codebase/main.py b/examples/ragify_codebase/main.py
index a0d6057d..dcff7207 100644
--- a/examples/ragify_codebase/main.py
+++ b/examples/ragify_codebase/main.py
@@ -7,11 +7,21 @@
 analyzer.setup_repository("./github_app/")
 
 # Perform queries (you can do this as many times as you want without calling setup_repository again)
-results = analyzer.query("Find functions that handle authentication")
+results = analyzer.query("jwt token generation")
 for result in results:
     print(f"File: {result['file_path']}")
-    print(f"Abstraction: {result['abstraction']}")
-    print(f"result:\n{result}")
+    # print(f"Abstraction: {result['abstraction']}")
+    # print(f"result:\n{result}")
+    print(f"Relevance Score: {result['relevance_score']}")
+    print("---")
+
+print("....... \n\n")
+
+results = analyzer.query("How do you filter the results?")
+for result in results:
+    print(f"File: {result['file_path']}")
+    # print(f"Abstraction: {result['abstraction']}")
+    # print(f"result:\n{result}")
     print(f"Relevance Score: {result['relevance_score']}")
     print("---")
 
@@ -19,4 +29,4 @@
 # analyzer.setup_repository("/path/to/your/repo")
 
 # Then you can query again with the updated data
-results = analyzer.query("authentication")
+# results = analyzer.query("authentication")
diff --git a/kaizen/llms/provider.py b/kaizen/llms/provider.py
index e10c8a0f..4274d736 100644
--- a/kaizen/llms/provider.py
+++ b/kaizen/llms/provider.py
@@ -83,6 +83,7 @@ def _setup_provider(self) -> None:
             "model_list": self.models,
             "allowed_fails": 1,
             "enable_pre_call_checks": True,
+            "routing_strategy": "simple-shuffle",
         }
 
         if self.config["language_model"].get("redis_enabled", False):
diff --git a/kaizen/retriever/custom_vector_store.py b/kaizen/retriever/custom_vector_store.py
deleted file mode 100644
index dfb2f061..00000000
--- a/kaizen/retriever/custom_vector_store.py
+++ /dev/null
@@ -1,59 +0,0 @@
-from llama_index.vector_stores.postgres import PGVectorStore
-from typing import List
-import numpy as np
-from psycopg2.extras import Json
-
-
-class CustomPGVectorStore(PGVectorStore):
-    def __init__(self, *args, **kwargs):
-        super().__init__(*args, **kwargs)
-        # Store the table name in a new attribute
-        self.table_name = kwargs.get("table_name", "embeddings")
-
-    def custom_query(
-        self, query_embedding: List[float], repo_id: int, similarity_top_k: int
-    ) -> List[dict]:
-        # Normalize the query embedding
-        query_embedding_np = np.array(query_embedding)
-        query_embedding_normalized = query_embedding_np / np.linalg.norm(
-            query_embedding_np
-        )
-
-        # SQL query with repo_id filter and cosine similarity
-        query = f"""
-        SELECT 
-            e.node_id,
-            e.text,
-            e.metadata,
-            1 - (e.embedding <=> %s::vector) as similarity
-        FROM 
-            {self.table_name} e
-        JOIN 
-            function_abstractions fa ON e.node_id = fa.function_id::text
-        JOIN 
-            files f ON fa.file_id = f.file_id
-        WHERE 
-            f.repo_id = %s
-        ORDER BY 
-            similarity DESC
-        LIMIT 
-            %s
-        """
-
-        with self.get_client() as client:
-            with client.cursor() as cur:
-                cur.execute(
-                    query,
-                    (query_embedding_normalized.tolist(), repo_id, similarity_top_k),
-                )
-                results = cur.fetchall()
-
-        return [
-            {
-                "id": row[0],
-                "text": row[1],
-                "metadata": row[2] if isinstance(row[2], dict) else Json(row[2]),
-                "similarity": row[3],
-            }
-            for row in results
-        ]
diff --git a/kaizen/retriever/llama_index_retriever.py b/kaizen/retriever/llama_index_retriever.py
index 46e020dd..bd345782 100644
--- a/kaizen/retriever/llama_index_retriever.py
+++ b/kaizen/retriever/llama_index_retriever.py
@@ -2,10 +2,11 @@
 import logging
 from llama_index.core import (
     StorageContext,
-    VectorStoreIndex,
 )
+from uuid import uuid4
+
 from llama_index.core.schema import TextNode
-from kaizen.retriever.custom_vector_store import CustomPGVectorStore
+from kaizen.retriever.qdrant_vector_store import QdrantVectorStore
 from sqlalchemy import create_engine, text
 from llama_index.llms.litellm import LiteLLM
 import networkx as nx
@@ -16,7 +17,6 @@
 from kaizen.retriever.code_chunker import chunk_code
 import traceback
 from llama_index.embeddings.litellm import LiteLLMEmbedding
-from llama_index.core import QueryBundle
 
 
 # Set up logging
@@ -39,15 +39,7 @@ def __init__(self, repo_id=1):
         )
         self.repo_id = repo_id
         self.graph = nx.DiGraph()
-        self.vector_store = CustomPGVectorStore.from_params(
-            database=os.environ["POSTGRES_DB"],
-            host=os.environ["POSTGRES_HOST"],
-            password=os.environ["POSTGRES_PASSWORD"],
-            port=os.environ["POSTGRES_PORT"],
-            user=os.environ["POSTGRES_USER"],
-            table_name="embeddings",
-            embed_dim=1536,
-        )
+        self.vector_store = QdrantVectorStore("embeddings", vector_size=1536)
         self.llm_provider = LLMProvider()
         self.llm = LiteLLM(model_name="small", router=self.llm_provider.provider)
         # embed_llm = LiteLLM(model_name="embedding", router=self.llm_provider.provider)
@@ -173,10 +165,11 @@ def store_abstraction_and_embedding(self, function_id: int, abstraction: str):
 
         # Create a TextNode for the vector store
         # Include repo_id in the metadata
-        metadata = {"repo_id": self.repo_id}
+        metadata = {"repo_id": self.repo_id, "function_id": function_id}
+        node_id = str(uuid4())
         node = TextNode(
             text=abstraction,
-            id_=str(function_id),
+            id_=node_id,
             embedding=embedding,
             metadata=metadata,
         )
@@ -189,19 +182,61 @@ def store_abstraction_and_embedding(self, function_id: int, abstraction: str):
     def generate_abstraction(
         self, code_block: str, language: str, max_tokens: int = 300
     ) -> str:
-        prompt = f"""Generate a concise yet comprehensive abstract description of the following {language} code block. 
-        Include information about:
-        1. The purpose or functionality of the code
-        2. Input parameters and return values (if applicable)
-        3. Any important algorithms or data structures used
-        4. Key dependencies or external libraries used
-        5. Any notable design patterns or architectural choices
-        6. Potential edge cases or error handling
-
-        Code:
-        ```{language}
-        {code_block}
-        ```
+        prompt = f"""Analyze the following {language} code block and generate a structured abstraction. 
+Your response should be in YAML format and include the following sections:
+
+summary: A concise one-sentence summary of the function's primary purpose.
+
+functionality: |
+  A detailed explanation of what the function does, including its main steps and logic.
+  Use multiple lines if needed for clarity.
+
+inputs:
+  - name: The parameter name
+    type: The parameter type
+    description: A brief description of the parameter's purpose
+    default_value: The default value, if any (or null if not applicable)
+
+output:
+  type: The return type of the function
+  description: |
+    A description of what is returned and under what conditions.
+    Use multiple lines if needed.
+
+dependencies:
+  - name: Name of the external library or module
+    purpose: Brief explanation of its use in this function
+
+algorithms:
+  - name: Name of the algorithm or data structure
+    description: Brief explanation of its use and importance
+
+edge_cases:
+  - A list of potential edge cases or special conditions the function handles or should handle
+
+error_handling: |
+  A description of how errors are handled or propagated.
+  Include specific error types if applicable.
+
+usage_context: |
+  A brief explanation of how this function might be used by parent functions or in a larger system.
+  Include typical scenarios and any important considerations for its use.
+
+complexity:
+  time: Estimated time complexity (e.g., O(n))
+  space: Estimated space complexity (e.g., O(1))
+
+code_snippet: |
+  ```{language}
+  {code_block}
+  ```
+
+Provide your analysis in this clear, structured YAML format. If any section is not applicable, use an empty list [] or null value as appropriate. Ensure that multi-line descriptions are properly indented under their respective keys.
+
+Code to analyze:
+```{language}
+{code_block}
+```
         """
 
         estimated_prompt_tokens = len(tokenizer.encode(prompt))
@@ -340,57 +375,47 @@ def store_function_relationships(self):
     #     logger.info(f"Query completed. Found {len(processed_results)} results.")
     #     return processed_results
 
-    def query(self, query_text: str, num_results: int = 5) -> List[Dict[str, Any]]:
-        logger.info(f"Performing query: '{query_text}' for repo_id: {self.repo_id}")
-
-        index = VectorStoreIndex.from_vector_store(
-            self.vector_store, embed_model=self.embed_model, llm=self.llm
-        )
-
+    def query(
+        self, query_text: str, num_results: int = 5, repo_id=None
+    ) -> List[Dict[str, Any]]:
         embedding, emb_usage = self.llm_provider.get_text_embedding(query_text)
         embedding = embedding[0]["embedding"]
 
-        # Create a filter to only search within the current repository
-        # filter_dict = {"repo_id": self.repo_id}
+        results = self.vector_store.search(embedding, limit=num_results)
 
-        query_bundle = QueryBundle(query_str=query_text, embedding=embedding)
-        retriever = index.as_retriever(similarity_top_k=num_results)
-
-        # Apply the filter during retrieval
-        nodes = retriever.retrieve(query_bundle)  # Add potential filtering
+        processed_results = []
+        for result in results:
+            processed_results.append(
+                {
+                    "function_id": result.payload["function_id"],
+                    "relevance_score": result.score,
+                }
+            )
 
-        results = []
+        # Fetch additional data from the database
         with self.engine.connect() as connection:
-            for node in nodes:
-                function_id = (
-                    node.node.id_
-                )  # Assuming we stored function_id as the node id
+            for result in processed_results:
                 query = text(
                     """
                     SELECT fa.function_name, fa.abstract_functionality, f.file_path, fa.function_signature
                     FROM function_abstractions fa
                     JOIN files f ON fa.file_id = f.file_id
                     WHERE fa.function_id = :function_id
-                    """
+                """
                 )
-                result = connection.execute(
-                    query, {"function_id": function_id}
+                db_result = connection.execute(
+                    query, {"function_id": result["function_id"]}
                 ).fetchone()
-                if result:
-                    results.append(
+                if db_result:
+                    result.update(
                         {
-                            "function_name": result[0],
-                            "abstraction": result[1],
-                            "file_path": result[2],
-                            "function_signature": result[3],
-                            "relevance_score": (
-                                node.score if hasattr(node, "score") else 1.0
-                            ),
+                            "function_name": db_result[0],
+                            "abstraction": db_result[1],
+                            "file_path": db_result[2],
+                            "function_signature": db_result[3],
                         }
                     )
 
-        sorted_results = sorted(
-            results, key=lambda x: x["relevance_score"], reverse=True
+        return sorted(
+            processed_results, key=lambda x: x["relevance_score"], reverse=True
         )
-        logger.info(f"Query completed. Found {len(sorted_results)} results.")
-        return sorted_results
diff --git a/kaizen/retriever/qdrant_vector_store.py b/kaizen/retriever/qdrant_vector_store.py
new file mode 100644
index 00000000..86a424d9
--- /dev/null
+++ b/kaizen/retriever/qdrant_vector_store.py
@@ -0,0 +1,26 @@
+from qdrant_client import QdrantClient
+from qdrant_client.models import Distance, VectorParams
+from qdrant_client.http.models import PointStruct
+
+
+class QdrantVectorStore:
+    def __init__(self, collection_name, vector_size):
+        self.client = QdrantClient("localhost", port=6333)
+        self.collection_name = collection_name
+        self.client.recreate_collection(
+            collection_name=self.collection_name,
+            vectors_config=VectorParams(size=vector_size, distance=Distance.COSINE),
+        )
+
+    def add(self, nodes):
+        points = [
+            PointStruct(id=node.id_, vector=node.embedding, payload=node.metadata)
+            for node in nodes
+        ]
+        self.client.upsert(collection_name=self.collection_name, points=points)
+
+    def search(self, query_vector, limit=10):
+        results = self.client.search(
+            collection_name=self.collection_name, query_vector=query_vector, limit=limit
+        )
+        return results
diff --git a/poetry.lock b/poetry.lock
index 1b3c4d2b..9b2c9127 100644
--- a/poetry.lock
+++ b/poetry.lock
@@ -166,63 +166,6 @@ files = [
     {file = "async_timeout-4.0.3-py3-none-any.whl", hash = "sha256:7405140ff1230c310e51dc27b3145b9092d659ce68ff733fb0cefe3ee42be028"},
 ]
 
-[[package]]
-name = "asyncpg"
-version = "0.29.0"
-description = "An asyncio PostgreSQL driver"
-optional = false
-python-versions = ">=3.8.0"
-files = [
-    {file = "asyncpg-0.29.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:72fd0ef9f00aeed37179c62282a3d14262dbbafb74ec0ba16e1b1864d8a12169"},
-    {file = "asyncpg-0.29.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:52e8f8f9ff6e21f9b39ca9f8e3e33a5fcdceaf5667a8c5c32bee158e313be385"},
-    {file = "asyncpg-0.29.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a9e6823a7012be8b68301342ba33b4740e5a166f6bbda0aee32bc01638491a22"},
-    {file = "asyncpg-0.29.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:746e80d83ad5d5464cfbf94315eb6744222ab00aa4e522b704322fb182b83610"},
-    {file = "asyncpg-0.29.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:ff8e8109cd6a46ff852a5e6bab8b0a047d7ea42fcb7ca5ae6eaae97d8eacf397"},
-    {file = "asyncpg-0.29.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:97eb024685b1d7e72b1972863de527c11ff87960837919dac6e34754768098eb"},
-    {file = "asyncpg-0.29.0-cp310-cp310-win32.whl", hash = "sha256:5bbb7f2cafd8d1fa3e65431833de2642f4b2124be61a449fa064e1a08d27e449"},
-    {file = "asyncpg-0.29.0-cp310-cp310-win_amd64.whl", hash = "sha256:76c3ac6530904838a4b650b2880f8e7af938ee049e769ec2fba7cd66469d7772"},
-    {file = "asyncpg-0.29.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:d4900ee08e85af01adb207519bb4e14b1cae8fd21e0ccf80fac6aa60b6da37b4"},
-    {file = "asyncpg-0.29.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:a65c1dcd820d5aea7c7d82a3fdcb70e096f8f70d1a8bf93eb458e49bfad036ac"},
-    {file = "asyncpg-0.29.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5b52e46f165585fd6af4863f268566668407c76b2c72d366bb8b522fa66f1870"},
-    {file = "asyncpg-0.29.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:dc600ee8ef3dd38b8d67421359779f8ccec30b463e7aec7ed481c8346decf99f"},
-    {file = "asyncpg-0.29.0-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:039a261af4f38f949095e1e780bae84a25ffe3e370175193174eb08d3cecab23"},
-    {file = "asyncpg-0.29.0-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:6feaf2d8f9138d190e5ec4390c1715c3e87b37715cd69b2c3dfca616134efd2b"},
-    {file = "asyncpg-0.29.0-cp311-cp311-win32.whl", hash = "sha256:1e186427c88225ef730555f5fdda6c1812daa884064bfe6bc462fd3a71c4b675"},
-    {file = "asyncpg-0.29.0-cp311-cp311-win_amd64.whl", hash = "sha256:cfe73ffae35f518cfd6e4e5f5abb2618ceb5ef02a2365ce64f132601000587d3"},
-    {file = "asyncpg-0.29.0-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:6011b0dc29886ab424dc042bf9eeb507670a3b40aece3439944006aafe023178"},
-    {file = "asyncpg-0.29.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:b544ffc66b039d5ec5a7454667f855f7fec08e0dfaf5a5490dfafbb7abbd2cfb"},
-    {file = "asyncpg-0.29.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d84156d5fb530b06c493f9e7635aa18f518fa1d1395ef240d211cb563c4e2364"},
-    {file = "asyncpg-0.29.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:54858bc25b49d1114178d65a88e48ad50cb2b6f3e475caa0f0c092d5f527c106"},
-    {file = "asyncpg-0.29.0-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:bde17a1861cf10d5afce80a36fca736a86769ab3579532c03e45f83ba8a09c59"},
-    {file = "asyncpg-0.29.0-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:37a2ec1b9ff88d8773d3eb6d3784dc7e3fee7756a5317b67f923172a4748a175"},
-    {file = "asyncpg-0.29.0-cp312-cp312-win32.whl", hash = "sha256:bb1292d9fad43112a85e98ecdc2e051602bce97c199920586be83254d9dafc02"},
-    {file = "asyncpg-0.29.0-cp312-cp312-win_amd64.whl", hash = "sha256:2245be8ec5047a605e0b454c894e54bf2ec787ac04b1cb7e0d3c67aa1e32f0fe"},
-    {file = "asyncpg-0.29.0-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:0009a300cae37b8c525e5b449233d59cd9868fd35431abc470a3e364d2b85cb9"},
-    {file = "asyncpg-0.29.0-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:5cad1324dbb33f3ca0cd2074d5114354ed3be2b94d48ddfd88af75ebda7c43cc"},
-    {file = "asyncpg-0.29.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:012d01df61e009015944ac7543d6ee30c2dc1eb2f6b10b62a3f598beb6531548"},
-    {file = "asyncpg-0.29.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:000c996c53c04770798053e1730d34e30cb645ad95a63265aec82da9093d88e7"},
-    {file = "asyncpg-0.29.0-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:e0bfe9c4d3429706cf70d3249089de14d6a01192d617e9093a8e941fea8ee775"},
-    {file = "asyncpg-0.29.0-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:642a36eb41b6313ffa328e8a5c5c2b5bea6ee138546c9c3cf1bffaad8ee36dd9"},
-    {file = "asyncpg-0.29.0-cp38-cp38-win32.whl", hash = "sha256:a921372bbd0aa3a5822dd0409da61b4cd50df89ae85150149f8c119f23e8c408"},
-    {file = "asyncpg-0.29.0-cp38-cp38-win_amd64.whl", hash = "sha256:103aad2b92d1506700cbf51cd8bb5441e7e72e87a7b3a2ca4e32c840f051a6a3"},
-    {file = "asyncpg-0.29.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:5340dd515d7e52f4c11ada32171d87c05570479dc01dc66d03ee3e150fb695da"},
-    {file = "asyncpg-0.29.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:e17b52c6cf83e170d3d865571ba574577ab8e533e7361a2b8ce6157d02c665d3"},
-    {file = "asyncpg-0.29.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f100d23f273555f4b19b74a96840aa27b85e99ba4b1f18d4ebff0734e78dc090"},
-    {file = "asyncpg-0.29.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:48e7c58b516057126b363cec8ca02b804644fd012ef8e6c7e23386b7d5e6ce83"},
-    {file = "asyncpg-0.29.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:f9ea3f24eb4c49a615573724d88a48bd1b7821c890c2effe04f05382ed9e8810"},
-    {file = "asyncpg-0.29.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:8d36c7f14a22ec9e928f15f92a48207546ffe68bc412f3be718eedccdf10dc5c"},
-    {file = "asyncpg-0.29.0-cp39-cp39-win32.whl", hash = "sha256:797ab8123ebaed304a1fad4d7576d5376c3a006a4100380fb9d517f0b59c1ab2"},
-    {file = "asyncpg-0.29.0-cp39-cp39-win_amd64.whl", hash = "sha256:cce08a178858b426ae1aa8409b5cc171def45d4293626e7aa6510696d46decd8"},
-    {file = "asyncpg-0.29.0.tar.gz", hash = "sha256:d1c49e1f44fffafd9a55e1a9b101590859d881d639ea2922516f5d9c512d354e"},
-]
-
-[package.dependencies]
-async-timeout = {version = ">=4.0.3", markers = "python_version < \"3.12.0\""}
-
-[package.extras]
-docs = ["Sphinx (>=5.3.0,<5.4.0)", "sphinx-rtd-theme (>=1.2.2)", "sphinxcontrib-asyncio (>=0.3.0,<0.4.0)"]
-test = ["flake8 (>=6.1,<7.0)", "uvloop (>=0.15.3)"]
-
 [[package]]
 name = "attrs"
 version = "24.1.0"
@@ -960,6 +903,124 @@ files = [
 docs = ["Sphinx", "furo"]
 test = ["objgraph", "psutil"]
 
+[[package]]
+name = "grpcio"
+version = "1.66.0"
+description = "HTTP/2-based RPC framework"
+optional = false
+python-versions = ">=3.8"
+files = [
+    {file = "grpcio-1.66.0-cp310-cp310-linux_armv7l.whl", hash = "sha256:ad7256f224437b2c29c2bef98ddd3130454c5b1ab1f0471fc11794cefd4dbd3d"},
+    {file = "grpcio-1.66.0-cp310-cp310-macosx_12_0_universal2.whl", hash = "sha256:5f4b3357e59dfba9140a51597287297bc638710d6a163f99ee14efc19967a821"},
+    {file = "grpcio-1.66.0-cp310-cp310-manylinux_2_17_aarch64.whl", hash = "sha256:e8d20308eeae15b3e182f47876f05acbdec1eebd9473a9814a44e46ec4a84c04"},
+    {file = "grpcio-1.66.0-cp310-cp310-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1eb03524d0f55b965d6c86aa44e5db9e5eaa15f9ed3b164621e652e5b927f4b8"},
+    {file = "grpcio-1.66.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:37514b68a42e9cf24536345d3cf9e580ffd29117c158b4eeea34625200256067"},
+    {file = "grpcio-1.66.0-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:516fdbc8e156db71a004bc431a6303bca24cfde186babe96dde7bd01e8f0cc70"},
+    {file = "grpcio-1.66.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:d0439a970d65327de21c299ea0e0c2ad0987cdaf18ba5066621dea5f427f922b"},
+    {file = "grpcio-1.66.0-cp310-cp310-win32.whl", hash = "sha256:5f93fc84b72bbc7b84a42f3ca9dc055fa00d2303d9803be011ebf7a10a4eb833"},
+    {file = "grpcio-1.66.0-cp310-cp310-win_amd64.whl", hash = "sha256:8fc5c710ddd51b5a0dc36ef1b6663430aa620e0ce029b87b150dafd313b978c3"},
+    {file = "grpcio-1.66.0-cp311-cp311-linux_armv7l.whl", hash = "sha256:dd614370e939f9fceeeb2915111a0795271b4c11dfb5fc0f58449bee40c726a5"},
+    {file = "grpcio-1.66.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:245b08f9b3c645a6a623f3ed4fa43dcfcd6ad701eb9c32511c1bb7380e8c3d23"},
+    {file = "grpcio-1.66.0-cp311-cp311-manylinux_2_17_aarch64.whl", hash = "sha256:aaf30c75cbaf30e561ca45f21eb1f729f0fab3f15c592c1074795ed43e3ff96f"},
+    {file = "grpcio-1.66.0-cp311-cp311-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:49234580a073ce7ac490112f6c67c874cbcb27804c4525978cdb21ba7f3f193c"},
+    {file = "grpcio-1.66.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:de9e20a0acb709dcfa15a622c91f584f12c9739a79c47999f73435d2b3cc8a3b"},
+    {file = "grpcio-1.66.0-cp311-cp311-musllinux_1_1_i686.whl", hash = "sha256:bc008c6afa1e7c8df99bd9154abc4f0470d26b7730ca2521122e99e771baa8c7"},
+    {file = "grpcio-1.66.0-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:50cea8ce2552865b87e3dffbb85eb21e6b98d928621600c0feda2f02449cd837"},
+    {file = "grpcio-1.66.0-cp311-cp311-win32.whl", hash = "sha256:508411df1f2b7cfa05d4d7dbf3d576fe4f949cd61c03f3a6f0378c84e3d7b963"},
+    {file = "grpcio-1.66.0-cp311-cp311-win_amd64.whl", hash = "sha256:6d586a95c05c82a5354be48bb4537e1accaf2472d8eb7e9086d844cbff934482"},
+    {file = "grpcio-1.66.0-cp312-cp312-linux_armv7l.whl", hash = "sha256:5ea27f4ce8c0daccfdd2c7961e6ba404b6599f47c948415c4cca5728739107a3"},
+    {file = "grpcio-1.66.0-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:296a45ea835e12a1cc35ab0c57e455346c272af7b0d178e29c67742167262b4c"},
+    {file = "grpcio-1.66.0-cp312-cp312-manylinux_2_17_aarch64.whl", hash = "sha256:e36fa838ac1d6c87198ca149cbfcc92e1af06bb8c8cd852622f8e58f33ea3324"},
+    {file = "grpcio-1.66.0-cp312-cp312-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:684a4c07883cbd4ac864f0d08d927267404f5f0c76f31c85f9bbe05f2daae2f2"},
+    {file = "grpcio-1.66.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c3084e590e857ba7585ae91078e4c9b6ef55aaf1dc343ce26400ba59a146eada"},
+    {file = "grpcio-1.66.0-cp312-cp312-musllinux_1_1_i686.whl", hash = "sha256:526d4f6ca19f31b25606d5c470ecba55c0b22707b524e4de8987919e8920437d"},
+    {file = "grpcio-1.66.0-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:423ae18637cd99ddcf2e5a6851c61828c49e9b9d022d0442d979b4f230109787"},
+    {file = "grpcio-1.66.0-cp312-cp312-win32.whl", hash = "sha256:7bc9d823e05d63a87511fb456dcc48dc0fced86c282bf60229675e7ee7aac1a1"},
+    {file = "grpcio-1.66.0-cp312-cp312-win_amd64.whl", hash = "sha256:230cdd696751e7eb1395718cd308234749daa217bb8d128f00357dc4df102558"},
+    {file = "grpcio-1.66.0-cp38-cp38-linux_armv7l.whl", hash = "sha256:0f3010bf46b2a01c9e40644cb9ed91b4b8435e5c500a275da5f9f62580e31e80"},
+    {file = "grpcio-1.66.0-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:ba18cfdc09312eb2eea6fa0ce5d2eec3cf345ea78f6528b2eaed6432105e0bd0"},
+    {file = "grpcio-1.66.0-cp38-cp38-manylinux_2_17_aarch64.whl", hash = "sha256:53d4c6706b49e358a2a33345dbe9b6b3bb047cecd7e8c07ba383bd09349bfef8"},
+    {file = "grpcio-1.66.0-cp38-cp38-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:643d8d9632a688ae69661e924b862e23c83a3575b24e52917ec5bcc59543d212"},
+    {file = "grpcio-1.66.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ba60ae3b465b3e85080ae3bfbc36fd0305ae495ab16fcf8022fc7d7a23aac846"},
+    {file = "grpcio-1.66.0-cp38-cp38-musllinux_1_1_i686.whl", hash = "sha256:9d5251578767fe44602688c851c2373b5513048ac84c21a0fe946590a8e7933d"},
+    {file = "grpcio-1.66.0-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:5e8140b39f10d7be2263afa2838112de29374c5c740eb0afd99146cb5bdbd990"},
+    {file = "grpcio-1.66.0-cp38-cp38-win32.whl", hash = "sha256:5b15ef1b296c4e78f15f64fc65bf8081f8774480ffcac45642f69d9d753d9c6b"},
+    {file = "grpcio-1.66.0-cp38-cp38-win_amd64.whl", hash = "sha256:c072f90a1f0409f827ae86266984cba65e89c5831a0726b9fc7f4b5fb940b853"},
+    {file = "grpcio-1.66.0-cp39-cp39-linux_armv7l.whl", hash = "sha256:a639d3866bfb5a678b5c0b92cd7ab543033ed8988854290fd86145e71731fd4c"},
+    {file = "grpcio-1.66.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:6ed35bf7da3fb3b1949e32bdf47a8b5ffe0aed11722d948933bd068531cd4682"},
+    {file = "grpcio-1.66.0-cp39-cp39-manylinux_2_17_aarch64.whl", hash = "sha256:1c5466222470cb7fbc9cc898af1d48eefd297cb2e2f59af6d4a851c862fa90ac"},
+    {file = "grpcio-1.66.0-cp39-cp39-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:921b8f7f25d5300d7c6837a1e0639ef145fbdbfb728e0a5db2dbccc9fc0fd891"},
+    {file = "grpcio-1.66.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c3f6feb0dc8456d025e566709f7dd02885add99bedaac50229013069242a1bfd"},
+    {file = "grpcio-1.66.0-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:748452dbd5a047475d5413bdef08b0b9ceb2c0c0e249d4ee905a5fb82c6328dc"},
+    {file = "grpcio-1.66.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:832945e64176520520317b50d64ec7d79924429528d5747669b52d0bf2c7bd78"},
+    {file = "grpcio-1.66.0-cp39-cp39-win32.whl", hash = "sha256:8096a922eb91bc97c839f675c3efa1257c6ef181ae1b25d3fb97f2cae4c57c01"},
+    {file = "grpcio-1.66.0-cp39-cp39-win_amd64.whl", hash = "sha256:375b58892301a5fc6ca7d7ff689c9dc9d00895f5d560604ace9f4f0573013c63"},
+    {file = "grpcio-1.66.0.tar.gz", hash = "sha256:c1ea4c528e7db6660718e4165fd1b5ac24b79a70c870a7bc0b7bdb9babab7c1e"},
+]
+
+[package.extras]
+protobuf = ["grpcio-tools (>=1.66.0)"]
+
+[[package]]
+name = "grpcio-tools"
+version = "1.66.0"
+description = "Protobuf code generator for gRPC"
+optional = false
+python-versions = ">=3.8"
+files = [
+    {file = "grpcio_tools-1.66.0-cp310-cp310-linux_armv7l.whl", hash = "sha256:e0841fe0aa865694468243b682792d6649a9eaaeec103984a74fcf4289851a83"},
+    {file = "grpcio_tools-1.66.0-cp310-cp310-macosx_12_0_universal2.whl", hash = "sha256:30261ab79e460e93002117627ec42a960c0d3d6292e3fd44a43eae94aedbae9a"},
+    {file = "grpcio_tools-1.66.0-cp310-cp310-manylinux_2_17_aarch64.whl", hash = "sha256:879a70a153f05d61fae8e7dd88ad67c63c1a30ee22c344509ec2b898f1e29250"},
+    {file = "grpcio_tools-1.66.0-cp310-cp310-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ff704d5b2c66e15aee1f34c74d8a44f0b613e9205d69c22172ffa056f9791db4"},
+    {file = "grpcio_tools-1.66.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:24773294210f554cdf282feaa3f95b79e22de56f78ec7a2e66c990266100480b"},
+    {file = "grpcio_tools-1.66.0-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:2da55cab0569eb2bae8fc445cb9eaafad488918e4a443f831dbdd2ce60c47684"},
+    {file = "grpcio_tools-1.66.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:72e86d15d5dab2f25385e40608f5dc6b512172c3b10d01952d3d25f2d0648b7c"},
+    {file = "grpcio_tools-1.66.0-cp310-cp310-win32.whl", hash = "sha256:923c60602e2025e1082cd3a1d7a5f74314f945ebb4763a939cc3f5a667d48d7f"},
+    {file = "grpcio_tools-1.66.0-cp310-cp310-win_amd64.whl", hash = "sha256:95edac51be6cd1391726024dea3a2a852c0a4c63e90de1ec52b5857d1ad5fef1"},
+    {file = "grpcio_tools-1.66.0-cp311-cp311-linux_armv7l.whl", hash = "sha256:81123f93a4f93f8e2bd7ba4a106c1eb1529e0336368c3b93c077f7649b48d784"},
+    {file = "grpcio_tools-1.66.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:95e3d1506bb3c6574c9d359ac78eaaad18276a3aaa328852796ee10d28a10656"},
+    {file = "grpcio_tools-1.66.0-cp311-cp311-manylinux_2_17_aarch64.whl", hash = "sha256:51cdcdf9dc9087bfc5d7aa03c4c76614350e0f7ef0689763f69938d1a7ebfac4"},
+    {file = "grpcio_tools-1.66.0-cp311-cp311-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:e5ef97b6e945e77575d07dc2158773313aa1b36ddab41c59a1c51803b4620abd"},
+    {file = "grpcio_tools-1.66.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:cc188a5fbaf25e3a5f91f815d3928b1e40ba38f5a5f5b5e86f640c575f7db1c9"},
+    {file = "grpcio_tools-1.66.0-cp311-cp311-musllinux_1_1_i686.whl", hash = "sha256:fddc8f3216199f47f2370f8a22ecc10a4e0b5c434eeab0ec47a79fb292e5a6f8"},
+    {file = "grpcio_tools-1.66.0-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:87a654381cdc43a64f890e1f68ca14f09c5bcafe9fe2481f50029a220b748d15"},
+    {file = "grpcio_tools-1.66.0-cp311-cp311-win32.whl", hash = "sha256:ecb781e41b08b094742137f56740acebedc29a18480a37c16d5dfed2aef0597a"},
+    {file = "grpcio_tools-1.66.0-cp311-cp311-win_amd64.whl", hash = "sha256:cf5906367329121b90942de6a2f77b316090ce15980254c61ecd5043526dc03d"},
+    {file = "grpcio_tools-1.66.0-cp312-cp312-linux_armv7l.whl", hash = "sha256:bcb7f09c1569c2e5f1600e5b1eb6a8321e789a3e1d2f9ec5c236c62d61d22879"},
+    {file = "grpcio_tools-1.66.0-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:ca654c732029483a0355164f551b4531eae1d1f64e269d389d97d79a0b087966"},
+    {file = "grpcio_tools-1.66.0-cp312-cp312-manylinux_2_17_aarch64.whl", hash = "sha256:b117868e2040489d8d542348a45cce6225fc87e1bc5e6092ad05bea343d4723d"},
+    {file = "grpcio_tools-1.66.0-cp312-cp312-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d72c6a8e1470832199764a4ac4aa999def0ccfb0fe0266c73aae003812acb957"},
+    {file = "grpcio_tools-1.66.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7055599f250713662022f5096956c220ff0f43a7ab500d080b0f343ba8d98e14"},
+    {file = "grpcio_tools-1.66.0-cp312-cp312-musllinux_1_1_i686.whl", hash = "sha256:4ecd2caa15c2070182e49aa1771cbf8e6181e5072833222401d965c6338a075c"},
+    {file = "grpcio_tools-1.66.0-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:b7da029e5a1270a0342c01f897436ab690677502e12f18664b7387a5e6938134"},
+    {file = "grpcio_tools-1.66.0-cp312-cp312-win32.whl", hash = "sha256:bde2aca5fd16e5ab37cf83a8a7b805ccb7faceb804c562387852a3146bfd7eaf"},
+    {file = "grpcio_tools-1.66.0-cp312-cp312-win_amd64.whl", hash = "sha256:e5507e1fee9caa19e2525d280016af8f4404affaad1a7c08beb7060797bd7972"},
+    {file = "grpcio_tools-1.66.0-cp38-cp38-linux_armv7l.whl", hash = "sha256:01449e9b20347fc7661f79090a9c0317e6de2759748170ac04cc0a4db74a681f"},
+    {file = "grpcio_tools-1.66.0-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:9c026adf37d1dacc3270c60ef479945c68756a251c362aef51c250e1f69f6a18"},
+    {file = "grpcio_tools-1.66.0-cp38-cp38-manylinux_2_17_aarch64.whl", hash = "sha256:2e31ac9a93feb5a4fbbb72de7a9a39709f28eea8183bab5e88f90a7facccf00b"},
+    {file = "grpcio_tools-1.66.0-cp38-cp38-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:63897f679ea55bc25accc825329b53acef2ad1266237d90be63c5aeaaa5bf175"},
+    {file = "grpcio_tools-1.66.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7d38a0b97d16343b3389228edc58c9dfea69bd3833fe458681f9cf66d13bb2e0"},
+    {file = "grpcio_tools-1.66.0-cp38-cp38-musllinux_1_1_i686.whl", hash = "sha256:8e197458cc1747f56a5b6bddd635247f86d3eb2a8a191e3f43ce0e6f2bf374c5"},
+    {file = "grpcio_tools-1.66.0-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:fd70b60d6b62df3d232e6c4f6c061c6bb5e071af88fe6323487d0b3b97ac87d2"},
+    {file = "grpcio_tools-1.66.0-cp38-cp38-win32.whl", hash = "sha256:65dfc1019a6dc3343161360a9436ca34f4aa4ffc40f4cdcd98e1e887dbe87cf8"},
+    {file = "grpcio_tools-1.66.0-cp38-cp38-win_amd64.whl", hash = "sha256:2a76db15aea734e583158c7190615f9e82de19fbb1f8d15f7a34fa9e4c3938a5"},
+    {file = "grpcio_tools-1.66.0-cp39-cp39-linux_armv7l.whl", hash = "sha256:95f1d076a310007fff710b4eea648a98ec75e0eb755b9df9af03b38a120ed8ac"},
+    {file = "grpcio_tools-1.66.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:eaf20f8141646b1db73f36711960d1bdf96435fbce670417e0754b15fbc52e76"},
+    {file = "grpcio_tools-1.66.0-cp39-cp39-manylinux_2_17_aarch64.whl", hash = "sha256:d84db86038507c86bfa148c9b6dde5a17b8b2e529eecbf1ca427c367043a56e8"},
+    {file = "grpcio_tools-1.66.0-cp39-cp39-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:7ca7080ac2aed6d303fab162c5945d920c0243a7a393df71c9f98882583dcda5"},
+    {file = "grpcio_tools-1.66.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:af2f8f944e779cb8dd5b5e8a689514775c745068cd564df662e00cab45430d40"},
+    {file = "grpcio_tools-1.66.0-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:e67a36da1ca3501933f26bd65589b7a5abdf5cfed79fd419054a0924f79fa760"},
+    {file = "grpcio_tools-1.66.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:2e78e94d9db3d686bc76f0ecedf5634ca3fad2d94e50c564a7d87630326719e8"},
+    {file = "grpcio_tools-1.66.0-cp39-cp39-win32.whl", hash = "sha256:00aafd7714f2e2f618ec75b0f13df6a6f174f2bc50ad70c79443d8f5aa60df96"},
+    {file = "grpcio_tools-1.66.0-cp39-cp39-win_amd64.whl", hash = "sha256:a236df9ac2dd1f6009adc94bce1da10ac46dd87a04dea86bfbeadaa261c7adea"},
+    {file = "grpcio_tools-1.66.0.tar.gz", hash = "sha256:6e111f73f400d64b8dc32f5dab67c5e806c290eb2658fecdbfc44c2bb1020efc"},
+]
+
+[package.dependencies]
+grpcio = ">=1.66.0"
+protobuf = ">=5.26.1,<6.0dev"
+setuptools = "*"
+
 [[package]]
 name = "h11"
 version = "0.14.0"
@@ -971,6 +1032,32 @@ files = [
     {file = "h11-0.14.0.tar.gz", hash = "sha256:8f19fbbe99e72420ff35c00b27a34cb9937e902a8b810e2c88300c6f0a3b699d"},
 ]
 
+[[package]]
+name = "h2"
+version = "4.1.0"
+description = "HTTP/2 State-Machine based protocol implementation"
+optional = false
+python-versions = ">=3.6.1"
+files = [
+    {file = "h2-4.1.0-py3-none-any.whl", hash = "sha256:03a46bcf682256c95b5fd9e9a99c1323584c3eec6440d379b9903d709476bc6d"},
+    {file = "h2-4.1.0.tar.gz", hash = "sha256:a83aca08fbe7aacb79fec788c9c0bac936343560ed9ec18b82a13a12c28d2abb"},
+]
+
+[package.dependencies]
+hpack = ">=4.0,<5"
+hyperframe = ">=6.0,<7"
+
+[[package]]
+name = "hpack"
+version = "4.0.0"
+description = "Pure-Python HPACK header compression"
+optional = false
+python-versions = ">=3.6.1"
+files = [
+    {file = "hpack-4.0.0-py3-none-any.whl", hash = "sha256:84a076fad3dc9a9f8063ccb8041ef100867b1878b25ef0ee63847a5d53818a6c"},
+    {file = "hpack-4.0.0.tar.gz", hash = "sha256:fc41de0c63e687ebffde81187a948221294896f6bdc0ae2312708df339430095"},
+]
+
 [[package]]
 name = "httpcore"
 version = "1.0.5"
@@ -1006,6 +1093,7 @@ files = [
 [package.dependencies]
 anyio = "*"
 certifi = "*"
+h2 = {version = ">=3,<5", optional = true, markers = "extra == \"http2\""}
 httpcore = "==1.*"
 idna = "*"
 sniffio = "*"
@@ -1050,6 +1138,17 @@ testing = ["InquirerPy (==0.3.4)", "Jinja2", "Pillow", "aiohttp", "fastapi", "gr
 torch = ["safetensors[torch]", "torch"]
 typing = ["types-PyYAML", "types-requests", "types-simplejson", "types-toml", "types-tqdm", "types-urllib3", "typing-extensions (>=4.8.0)"]
 
+[[package]]
+name = "hyperframe"
+version = "6.0.1"
+description = "HTTP/2 framing layer for Python"
+optional = false
+python-versions = ">=3.6.1"
+files = [
+    {file = "hyperframe-6.0.1-py3-none-any.whl", hash = "sha256:0ec6bafd80d8ad2195c4f03aacba3a8265e57bc4cff261e802bf39970ed02a15"},
+    {file = "hyperframe-6.0.1.tar.gz", hash = "sha256:ae510046231dc8e9ecb1a6586f63d2347bf4c8905914aa84ba585ae85f28a914"},
+]
+
 [[package]]
 name = "identify"
 version = "2.6.0"
@@ -1568,24 +1667,6 @@ files = [
 llama-index-core = ">=0.10.7,<0.11.0"
 llama-parse = ">=0.4.0"
 
-[[package]]
-name = "llama-index-vector-stores-postgres"
-version = "0.1.11"
-description = "llama-index vector_stores postgres integration"
-optional = false
-python-versions = "<4.0,>=3.8.1"
-files = [
-    {file = "llama_index_vector_stores_postgres-0.1.11-py3-none-any.whl", hash = "sha256:a3856372579c541457dfde295858cb9d8719f588d559f55d57bc6174f93a8293"},
-    {file = "llama_index_vector_stores_postgres-0.1.11.tar.gz", hash = "sha256:ce23ff9549c5269bdccba638875b921faaa4a581cefb753e99f8365c82487a0e"},
-]
-
-[package.dependencies]
-asyncpg = ">=0.29.0,<0.30.0"
-llama-index-core = ">=0.10.20,<0.11.0"
-pgvector = ">=0.2.4,<0.3.0"
-psycopg2-binary = ">=2.9.9,<3.0.0"
-sqlalchemy = {version = ">=1.4.49,<2.1", extras = ["asyncio"]}
-
 [[package]]
 name = "llama-parse"
 version = "0.4.9"
@@ -2038,19 +2119,6 @@ files = [
     {file = "pathspec-0.12.1.tar.gz", hash = "sha256:a482d51503a1ab33b1c67a6c3813a26953dbdc71c31dacaef9a838c4e29f5712"},
 ]
 
-[[package]]
-name = "pgvector"
-version = "0.2.5"
-description = "pgvector support for Python"
-optional = false
-python-versions = ">=3.8"
-files = [
-    {file = "pgvector-0.2.5-py2.py3-none-any.whl", hash = "sha256:5e5e93ec4d3c45ab1fa388729d56c602f6966296e19deee8878928c6d567e41b"},
-]
-
-[package.dependencies]
-numpy = "*"
-
 [[package]]
 name = "pillow"
 version = "10.3.0"
@@ -2199,6 +2267,25 @@ files = [
 dev = ["pre-commit", "tox"]
 testing = ["pytest", "pytest-benchmark"]
 
+[[package]]
+name = "portalocker"
+version = "2.10.1"
+description = "Wraps the portalocker recipe for easy usage"
+optional = false
+python-versions = ">=3.8"
+files = [
+    {file = "portalocker-2.10.1-py3-none-any.whl", hash = "sha256:53a5984ebc86a025552264b459b46a2086e269b21823cb572f8f28ee759e45bf"},
+    {file = "portalocker-2.10.1.tar.gz", hash = "sha256:ef1bf844e878ab08aee7e40184156e1151f228f103aa5c6bd0724cc330960f8f"},
+]
+
+[package.dependencies]
+pywin32 = {version = ">=226", markers = "platform_system == \"Windows\""}
+
+[package.extras]
+docs = ["sphinx (>=1.7.1)"]
+redis = ["redis"]
+tests = ["pytest (>=5.4.1)", "pytest-cov (>=2.8.1)", "pytest-mypy (>=0.8.0)", "pytest-timeout (>=2.1.0)", "redis", "sphinx (>=6.0.0)", "types-redis"]
+
 [[package]]
 name = "pre-commit"
 version = "3.8.0"
@@ -2217,6 +2304,26 @@ nodeenv = ">=0.11.1"
 pyyaml = ">=5.1"
 virtualenv = ">=20.10.0"
 
+[[package]]
+name = "protobuf"
+version = "5.27.3"
+description = ""
+optional = false
+python-versions = ">=3.8"
+files = [
+    {file = "protobuf-5.27.3-cp310-abi3-win32.whl", hash = "sha256:dcb307cd4ef8fec0cf52cb9105a03d06fbb5275ce6d84a6ae33bc6cf84e0a07b"},
+    {file = "protobuf-5.27.3-cp310-abi3-win_amd64.whl", hash = "sha256:16ddf3f8c6c41e1e803da7abea17b1793a97ef079a912e42351eabb19b2cffe7"},
+    {file = "protobuf-5.27.3-cp38-abi3-macosx_10_9_universal2.whl", hash = "sha256:68248c60d53f6168f565a8c76dc58ba4fa2ade31c2d1ebdae6d80f969cdc2d4f"},
+    {file = "protobuf-5.27.3-cp38-abi3-manylinux2014_aarch64.whl", hash = "sha256:b8a994fb3d1c11156e7d1e427186662b64694a62b55936b2b9348f0a7c6625ce"},
+    {file = "protobuf-5.27.3-cp38-abi3-manylinux2014_x86_64.whl", hash = "sha256:a55c48f2a2092d8e213bd143474df33a6ae751b781dd1d1f4d953c128a415b25"},
+    {file = "protobuf-5.27.3-cp38-cp38-win32.whl", hash = "sha256:043853dcb55cc262bf2e116215ad43fa0859caab79bb0b2d31b708f128ece035"},
+    {file = "protobuf-5.27.3-cp38-cp38-win_amd64.whl", hash = "sha256:c2a105c24f08b1e53d6c7ffe69cb09d0031512f0b72f812dd4005b8112dbe91e"},
+    {file = "protobuf-5.27.3-cp39-cp39-win32.whl", hash = "sha256:c84eee2c71ed83704f1afbf1a85c3171eab0fd1ade3b399b3fad0884cbcca8bf"},
+    {file = "protobuf-5.27.3-cp39-cp39-win_amd64.whl", hash = "sha256:af7c0b7cfbbb649ad26132e53faa348580f844d9ca46fd3ec7ca48a1ea5db8a1"},
+    {file = "protobuf-5.27.3-py3-none-any.whl", hash = "sha256:8572c6533e544ebf6899c360e91d6bcbbee2549251643d32c52cf8a5de295ba5"},
+    {file = "protobuf-5.27.3.tar.gz", hash = "sha256:82460903e640f2b7e34ee81a947fdaad89de796d324bcbc38ff5430bcdead82c"},
+]
+
 [[package]]
 name = "psycopg2-binary"
 version = "2.9.9"
@@ -2622,6 +2729,29 @@ files = [
     {file = "pytz-2024.1.tar.gz", hash = "sha256:2a29735ea9c18baf14b448846bde5a48030ed267578472d8955cd0e7443a9812"},
 ]
 
+[[package]]
+name = "pywin32"
+version = "306"
+description = "Python for Window Extensions"
+optional = false
+python-versions = "*"
+files = [
+    {file = "pywin32-306-cp310-cp310-win32.whl", hash = "sha256:06d3420a5155ba65f0b72f2699b5bacf3109f36acbe8923765c22938a69dfc8d"},
+    {file = "pywin32-306-cp310-cp310-win_amd64.whl", hash = "sha256:84f4471dbca1887ea3803d8848a1616429ac94a4a8d05f4bc9c5dcfd42ca99c8"},
+    {file = "pywin32-306-cp311-cp311-win32.whl", hash = "sha256:e65028133d15b64d2ed8f06dd9fbc268352478d4f9289e69c190ecd6818b6407"},
+    {file = "pywin32-306-cp311-cp311-win_amd64.whl", hash = "sha256:a7639f51c184c0272e93f244eb24dafca9b1855707d94c192d4a0b4c01e1100e"},
+    {file = "pywin32-306-cp311-cp311-win_arm64.whl", hash = "sha256:70dba0c913d19f942a2db25217d9a1b726c278f483a919f1abfed79c9cf64d3a"},
+    {file = "pywin32-306-cp312-cp312-win32.whl", hash = "sha256:383229d515657f4e3ed1343da8be101000562bf514591ff383ae940cad65458b"},
+    {file = "pywin32-306-cp312-cp312-win_amd64.whl", hash = "sha256:37257794c1ad39ee9be652da0462dc2e394c8159dfd913a8a4e8eb6fd346da0e"},
+    {file = "pywin32-306-cp312-cp312-win_arm64.whl", hash = "sha256:5821ec52f6d321aa59e2db7e0a35b997de60c201943557d108af9d4ae1ec7040"},
+    {file = "pywin32-306-cp37-cp37m-win32.whl", hash = "sha256:1c73ea9a0d2283d889001998059f5eaaba3b6238f767c9cf2833b13e6a685f65"},
+    {file = "pywin32-306-cp37-cp37m-win_amd64.whl", hash = "sha256:72c5f621542d7bdd4fdb716227be0dd3f8565c11b280be6315b06ace35487d36"},
+    {file = "pywin32-306-cp38-cp38-win32.whl", hash = "sha256:e4c092e2589b5cf0d365849e73e02c391c1349958c5ac3e9d5ccb9a28e017b3a"},
+    {file = "pywin32-306-cp38-cp38-win_amd64.whl", hash = "sha256:e8ac1ae3601bee6ca9f7cb4b5363bf1c0badb935ef243c4733ff9a393b1690c0"},
+    {file = "pywin32-306-cp39-cp39-win32.whl", hash = "sha256:e25fd5b485b55ac9c057f67d94bc203f3f6595078d1fb3b458c9c28b7153a802"},
+    {file = "pywin32-306-cp39-cp39-win_amd64.whl", hash = "sha256:39b61c15272833b5c329a2989999dcae836b1eed650252ab1b7bfbe1d59f30f4"},
+]
+
 [[package]]
 name = "pyyaml"
 version = "6.0.1"
@@ -2682,6 +2812,33 @@ files = [
     {file = "PyYAML-6.0.1.tar.gz", hash = "sha256:bfdf460b1736c775f2ba9f6a92bca30bc2095067b8a9d77876d1fad6cc3b4a43"},
 ]
 
+[[package]]
+name = "qdrant-client"
+version = "1.11.0"
+description = "Client library for the Qdrant vector search engine"
+optional = false
+python-versions = ">=3.8"
+files = [
+    {file = "qdrant_client-1.11.0-py3-none-any.whl", hash = "sha256:1f574ccebb91c0bc8a620c9a41a5a010084fbc4d8c6f1cd0ab7b2eeb97336fc0"},
+    {file = "qdrant_client-1.11.0.tar.gz", hash = "sha256:7c1d4d7a96cfd1ee0cde2a21c607e9df86bcca795ad8d1fd274d295ab64b8458"},
+]
+
+[package.dependencies]
+grpcio = ">=1.41.0"
+grpcio-tools = ">=1.41.0"
+httpx = {version = ">=0.20.0", extras = ["http2"]}
+numpy = [
+    {version = ">=1.21", markers = "python_version >= \"3.8\" and python_version < \"3.12\""},
+    {version = ">=1.26", markers = "python_version >= \"3.12\""},
+]
+portalocker = ">=2.7.0,<3.0.0"
+pydantic = ">=1.10.8"
+urllib3 = ">=1.26.14,<3"
+
+[package.extras]
+fastembed = ["fastembed (==0.3.4)"]
+fastembed-gpu = ["fastembed-gpu (==0.3.4)"]
+
 [[package]]
 name = "redis"
 version = "5.0.8"
@@ -2936,6 +3093,22 @@ files = [
     {file = "rpds_py-0.19.1.tar.gz", hash = "sha256:31dd5794837f00b46f4096aa8ccaa5972f73a938982e32ed817bb520c465e520"},
 ]
 
+[[package]]
+name = "setuptools"
+version = "73.0.1"
+description = "Easily download, build, install, upgrade, and uninstall Python packages"
+optional = false
+python-versions = ">=3.8"
+files = [
+    {file = "setuptools-73.0.1-py3-none-any.whl", hash = "sha256:b208925fcb9f7af924ed2dc04708ea89791e24bde0d3020b27df0e116088b34e"},
+    {file = "setuptools-73.0.1.tar.gz", hash = "sha256:d59a3e788ab7e012ab2c4baed1b376da6366883ee20d7a5fc426816e3d7b1193"},
+]
+
+[package.extras]
+core = ["importlib-metadata (>=6)", "importlib-resources (>=5.10.2)", "jaraco.text (>=3.7)", "more-itertools (>=8.8)", "packaging (>=24)", "platformdirs (>=2.6.2)", "tomli (>=2.0.1)", "wheel (>=0.43.0)"]
+doc = ["furo", "jaraco.packaging (>=9.3)", "jaraco.tidelift (>=1.4)", "pygments-github-lexers (==0.0.5)", "pyproject-hooks (!=1.1)", "rst.linker (>=1.9)", "sphinx (>=3.5)", "sphinx-favicon", "sphinx-inline-tabs", "sphinx-lint", "sphinx-notfound-page (>=1,<2)", "sphinx-reredirects", "sphinxcontrib-towncrier", "towncrier (<24.7)"]
+test = ["build[virtualenv] (>=1.0.3)", "filelock (>=3.4.0)", "importlib-metadata", "ini2toml[lite] (>=0.14)", "jaraco.develop (>=7.21)", "jaraco.envs (>=2.2)", "jaraco.path (>=3.2.0)", "jaraco.test", "mypy (==1.11.*)", "packaging (>=23.2)", "pip (>=19.1)", "pyproject-hooks (!=1.1)", "pytest (>=6,!=8.1.*)", "pytest-checkdocs (>=2.4)", "pytest-cov", "pytest-enabler (>=2.2)", "pytest-home (>=0.5)", "pytest-mypy", "pytest-perf", "pytest-ruff (<0.4)", "pytest-ruff (>=0.2.1)", "pytest-ruff (>=0.3.2)", "pytest-subprocess", "pytest-timeout", "pytest-xdist (>=3)", "tomli", "tomli-w (>=1.0.0)", "virtualenv (>=13.0.0)", "wheel (>=0.44.0)"]
+
 [[package]]
 name = "six"
 version = "1.16.0"
@@ -3731,4 +3904,4 @@ typescript = ["tree-sitter-typescript"]
 [metadata]
 lock-version = "2.0"
 python-versions = "^3.9.0"
-content-hash = "19b210560f02e7e1b4210a4cca0e50df1f6669e3b9586e24ca20a39265170490"
+content-hash = "b8b1f2e9a8da22e59f874abab8847f1096da0c2782924889778f48722c7f4e44"
diff --git a/pyproject.toml b/pyproject.toml
index 273795bd..848c9d62 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -26,7 +26,6 @@ pip = "^24.0"
 fuzzywuzzy = "^0.18.0"
 llama-index-core = "0.10.65"
 llama-index-readers-file = "^0.1.25"
-llama-index-vector-stores-postgres = "^0.1.11"
 sqlalchemy = "^2.0.31"
 esprima = "^4.0.1"
 escodegen = "^1.0.11"
@@ -41,6 +40,8 @@ tree-sitter-rust = "^0.21.2"
 llama-index-llms-litellm = "^0.1.4"
 llama-index-embeddings-litellm = "^0.1.1"
 pre-commit = "^3.8.0"
+qdrant-client = "^1.11.0"
+psycopg2-binary = "^2.9.9"
 
 [tool.poetry.extras]
 python = ["tree-sitter-python"]

From 8f3cd2005d04d51ab9eb61812f22e320657bd2b4 Mon Sep 17 00:00:00 2001
From: Saurav Panda <sgp65@cornell.edu>
Date: Tue, 27 Aug 2024 02:34:00 -0700
Subject: [PATCH 15/19] feat: added experimental eval

---
 .../no_eval/pr_335_comments.json              |  17 -
 .../no_eval/pr_335_review.md                  |  98 --
 .../with_eval/pr_335_comments.json            |  17 -
 .../with_eval/pr_335_review.md                | 114 ---
 .../code_review/dataset/pr_222/issues.json    | 152 +++
 .../code_review/dataset/pr_232/issues.json    | 107 ++
 .../code_review/dataset/pr_252/issues.json    |  92 ++
 .../code_review/dataset/pr_335/issues.json    |  92 ++
 .../code_review/dataset/pr_400/issues.json    |  52 +
 .../code_review/dataset/pr_476/issues.json    |  68 ++
 .../code_review/dataset/pr_5/issues.json      |  93 ++
 .experiments/code_review/evaluate.py          | 179 ++++
 .../gpt-4o-mini/no_eval/pr_222/comments.json  | 102 ++
 .../gpt-4o-mini/no_eval/pr_222/issues.json    | 342 +++++++
 .../gpt-4o-mini/no_eval/pr_222/review.md      | 402 ++++++++
 .../gpt-4o-mini/no_eval/pr_335/comments.json  |  17 +
 .../gpt-4o-mini/no_eval/pr_335/issues.json    |  77 ++
 .../gpt-4o-mini/no_eval/pr_335/review.md      | 139 +++
 .../gpt-4o-mini/no_eval/pr_440/comments.json  |   1 +
 .../gpt-4o-mini/no_eval/pr_440/issues.json    |  47 +
 .../gpt-4o-mini/no_eval/pr_440/review.md      | 100 ++
 .../gpt-4o-mini/no_eval/pr_476/comments.json  |  16 +
 .../gpt-4o-mini/no_eval/pr_476/issues.json    |  91 ++
 .../gpt-4o-mini/no_eval/pr_476/review.md      | 145 +++
 .../gpt-4o-mini/no_eval/pr_5/comments.json    |  47 +
 .../gpt-4o-mini/no_eval/pr_5/issues.json      | 107 ++
 .../gpt-4o-mini/no_eval/pr_5/review.md        | 155 +++
 .../gpt-4o/no_eval/pr_222/comments.json       | 132 +++
 .../gpt-4o/no_eval/pr_222/issues.json         | 387 +++++++
 .../gpt-4o/no_eval/pr_222/review.md           | 560 ++++++++++
 .../gpt-4o/no_eval/pr_335/comments.json       |  32 +
 .../gpt-4o/no_eval/pr_335/issues.json         |  92 ++
 .../gpt-4o/no_eval/pr_335/review.md           | 151 +++
 .../gpt-4o/no_eval/pr_440/comments.json       |   1 +
 .../gpt-4o/no_eval/pr_440/issues.json         |  47 +
 .../gpt-4o/no_eval/pr_440/review.md           |  92 ++
 .../gpt-4o/no_eval/pr_476/comments.json       |  31 +
 .../gpt-4o/no_eval/pr_476/issues.json         |  91 ++
 .../gpt-4o/no_eval/pr_476/review.md           | 144 +++
 .../gpt-4o/no_eval/pr_5/comments.json         |  47 +
 .../gpt-4o/no_eval/pr_5/issues.json           | 107 ++
 .../code_review/gpt-4o/no_eval/pr_5/review.md | 182 ++++
 .../haiku/no_eval/pr_222/comments.json        |  72 ++
 .../haiku/no_eval/pr_222/issues.json          | 312 ++++++
 .../haiku/no_eval/pr_222/review.md            | 247 +++++
 .../haiku/no_eval/pr_252/comments.json        |   1 +
 .../haiku/no_eval/pr_252/issues.json          |  32 +
 .../haiku/no_eval/pr_252/review.md            |  70 ++
 .../haiku/no_eval/pr_335/comments.json        |   1 +
 .../haiku/no_eval/pr_335/issues.json          |  62 ++
 .../haiku/no_eval/pr_335/review.md            |  78 ++
 .../haiku/no_eval/pr_400/comments.json        |  16 +
 .../haiku/no_eval/pr_400/issues.json          | 406 ++++++++
 .../haiku/no_eval/pr_400/review.md            | 177 ++++
 .../haiku/no_eval/pr_440/comments.json        |   1 +
 .../haiku/no_eval/pr_440/issues.json          |  62 ++
 .../haiku/no_eval/pr_440/review.md            | 118 +++
 .../haiku/no_eval/pr_476/comments.json        |  16 +
 .../haiku/no_eval/pr_476/issues.json          | 106 ++
 .../haiku/no_eval/pr_476/review.md            | 103 ++
 .../haiku/no_eval/pr_5/comments.json          |  32 +
 .../haiku/no_eval/pr_5/issues.json            | 107 ++
 .../code_review/haiku/no_eval/pr_5/review.md  | 211 ++++
 .../llama-405b/no_eval/pr_222/comments.json   | 117 +++
 .../llama-405b/no_eval/pr_222/issues.json     | 282 ++++++
 .../llama-405b/no_eval/pr_222/review.md       | 195 ++++
 .../llama-405b/no_eval/pr_232/comments.json   |   1 +
 .../llama-405b/no_eval/pr_232/issues.json     | 137 +++
 .../llama-405b/no_eval/pr_232/review.md       | 100 ++
 .../llama-405b/no_eval/pr_252/comments.json   |   1 +
 .../llama-405b/no_eval/pr_252/issues.json     |  62 ++
 .../llama-405b/no_eval/pr_252/review.md       |  78 ++
 .../llama-405b/no_eval/pr_335/comments.json   |   1 +
 .../llama-405b/no_eval/pr_335/issues.json     |  47 +
 .../llama-405b/no_eval/pr_335/review.md       |  62 ++
 .../llama-405b/no_eval/pr_400/comments.json   |  16 +
 .../llama-405b/no_eval/pr_400/issues.json     | 316 ++++++
 .../llama-405b/no_eval/pr_400/review.md       | 197 ++++
 .../llama-405b/no_eval/pr_440/comments.json   |   1 +
 .../llama-405b/no_eval/pr_440/issues.json     |  47 +
 .../llama-405b/no_eval/pr_440/review.md       |  92 ++
 .../llama-405b/no_eval/pr_476/comments.json   |  16 +
 .../llama-405b/no_eval/pr_476/issues.json     |  61 ++
 .../llama-405b/no_eval/pr_476/review.md       | 115 +++
 .../llama-405b/no_eval/pr_5/comments.json     |  47 +
 .../llama-405b/no_eval/pr_5/issues.json       | 107 ++
 .../llama-405b/no_eval/pr_5/review.md         | 193 ++++
 .experiments/code_review/main.py              |  36 +-
 .experiments/code_review/print_info.py        |  40 +
 .../sonnet-3.5/no_eval/pr_222/comments.json   |  72 ++
 .../sonnet-3.5/no_eval/pr_222/issues.json     | 312 ++++++
 .../sonnet-3.5/no_eval/pr_222/review.md       | 481 +++++++++
 .../sonnet-3.5/no_eval/pr_232/comments.json   |   1 +
 .../sonnet-3.5/no_eval/pr_232/issues.json     | 227 +++++
 .../sonnet-3.5/no_eval/pr_232/review.md       | 253 +++++
 .../sonnet-3.5/no_eval/pr_252/comments.json   |   1 +
 .../sonnet-3.5/no_eval/pr_252/issues.json     |  62 ++
 .../sonnet-3.5/no_eval/pr_252/review.md       | 126 +++
 .../sonnet-3.5/no_eval/pr_335/comments.json   |   1 +
 .../sonnet-3.5/no_eval/pr_335/issues.json     |  62 ++
 .../sonnet-3.5/no_eval/pr_335/review.md       |  89 ++
 .../sonnet-3.5/no_eval/pr_400/comments.json   |  31 +
 .../sonnet-3.5/no_eval/pr_400/issues.json     | 406 ++++++++
 .../sonnet-3.5/no_eval/pr_400/review.md       | 435 ++++++++
 .../sonnet-3.5/no_eval/pr_476/comments.json   |  16 +
 .../sonnet-3.5/no_eval/pr_476/issues.json     | 106 ++
 .../sonnet-3.5/no_eval/pr_476/review.md       | 154 +++
 .../sonnet-3.5/no_eval/pr_5/comments.json     |  47 +
 .../sonnet-3.5/no_eval/pr_5/issues.json       | 107 ++
 .../sonnet-3.5/no_eval/pr_5/review.md         | 234 +++++
 github_app/github_helper/pull_requests.py     |   9 +-
 kaizen/formatters/code_review_formatter.py    |  23 +-
 kaizen/llms/prompts/code_review_prompts.py    |  45 +-
 kaizen/llms/provider.py                       |  11 +-
 kaizen/reviewer/code_review.py                |   2 +-
 poetry.lock                                   | 323 +++++-
 pyproject.toml                                |   3 +
 test.txt                                      | 956 ++++++++++++++++++
 118 files changed, 13866 insertions(+), 291 deletions(-)
 delete mode 100644 .experiments/code_review/code_reviews_20240815_142254/no_eval/pr_335_comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240815_142254/no_eval/pr_335_review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240815_142254/with_eval/pr_335_comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240815_142254/with_eval/pr_335_review.md
 create mode 100644 .experiments/code_review/dataset/pr_222/issues.json
 create mode 100644 .experiments/code_review/dataset/pr_232/issues.json
 create mode 100644 .experiments/code_review/dataset/pr_252/issues.json
 create mode 100644 .experiments/code_review/dataset/pr_335/issues.json
 create mode 100644 .experiments/code_review/dataset/pr_400/issues.json
 create mode 100644 .experiments/code_review/dataset/pr_476/issues.json
 create mode 100644 .experiments/code_review/dataset/pr_5/issues.json
 create mode 100644 .experiments/code_review/evaluate.py
 create mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_222/comments.json
 create mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_222/issues.json
 create mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_222/review.md
 create mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_335/comments.json
 create mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_335/issues.json
 create mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_335/review.md
 create mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_440/comments.json
 create mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_440/issues.json
 create mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_440/review.md
 create mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_476/comments.json
 create mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_476/issues.json
 create mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_476/review.md
 create mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_5/comments.json
 create mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_5/issues.json
 create mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_5/review.md
 create mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_222/comments.json
 create mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_222/issues.json
 create mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_222/review.md
 create mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_335/comments.json
 create mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_335/issues.json
 create mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_335/review.md
 create mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_440/comments.json
 create mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_440/issues.json
 create mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_440/review.md
 create mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_476/comments.json
 create mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_476/issues.json
 create mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_476/review.md
 create mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_5/comments.json
 create mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_5/issues.json
 create mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_5/review.md
 create mode 100644 .experiments/code_review/haiku/no_eval/pr_222/comments.json
 create mode 100644 .experiments/code_review/haiku/no_eval/pr_222/issues.json
 create mode 100644 .experiments/code_review/haiku/no_eval/pr_222/review.md
 create mode 100644 .experiments/code_review/haiku/no_eval/pr_252/comments.json
 create mode 100644 .experiments/code_review/haiku/no_eval/pr_252/issues.json
 create mode 100644 .experiments/code_review/haiku/no_eval/pr_252/review.md
 create mode 100644 .experiments/code_review/haiku/no_eval/pr_335/comments.json
 create mode 100644 .experiments/code_review/haiku/no_eval/pr_335/issues.json
 create mode 100644 .experiments/code_review/haiku/no_eval/pr_335/review.md
 create mode 100644 .experiments/code_review/haiku/no_eval/pr_400/comments.json
 create mode 100644 .experiments/code_review/haiku/no_eval/pr_400/issues.json
 create mode 100644 .experiments/code_review/haiku/no_eval/pr_400/review.md
 create mode 100644 .experiments/code_review/haiku/no_eval/pr_440/comments.json
 create mode 100644 .experiments/code_review/haiku/no_eval/pr_440/issues.json
 create mode 100644 .experiments/code_review/haiku/no_eval/pr_440/review.md
 create mode 100644 .experiments/code_review/haiku/no_eval/pr_476/comments.json
 create mode 100644 .experiments/code_review/haiku/no_eval/pr_476/issues.json
 create mode 100644 .experiments/code_review/haiku/no_eval/pr_476/review.md
 create mode 100644 .experiments/code_review/haiku/no_eval/pr_5/comments.json
 create mode 100644 .experiments/code_review/haiku/no_eval/pr_5/issues.json
 create mode 100644 .experiments/code_review/haiku/no_eval/pr_5/review.md
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_222/comments.json
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_222/issues.json
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_222/review.md
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_232/comments.json
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_232/issues.json
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_232/review.md
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_252/comments.json
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_252/issues.json
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_252/review.md
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_335/comments.json
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_335/issues.json
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_335/review.md
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_400/comments.json
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_400/issues.json
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_400/review.md
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_440/comments.json
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_440/issues.json
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_440/review.md
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_476/comments.json
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_476/issues.json
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_476/review.md
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_5/comments.json
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_5/issues.json
 create mode 100644 .experiments/code_review/llama-405b/no_eval/pr_5/review.md
 create mode 100644 .experiments/code_review/print_info.py
 create mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_222/comments.json
 create mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_222/issues.json
 create mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_222/review.md
 create mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_232/comments.json
 create mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_232/issues.json
 create mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_232/review.md
 create mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_252/comments.json
 create mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_252/issues.json
 create mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_252/review.md
 create mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_335/comments.json
 create mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_335/issues.json
 create mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_335/review.md
 create mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_400/comments.json
 create mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_400/issues.json
 create mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_400/review.md
 create mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_476/comments.json
 create mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_476/issues.json
 create mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_476/review.md
 create mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_5/comments.json
 create mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_5/issues.json
 create mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_5/review.md
 create mode 100644 test.txt

diff --git a/.experiments/code_review/code_reviews_20240815_142254/no_eval/pr_335_comments.json b/.experiments/code_review/code_reviews_20240815_142254/no_eval/pr_335_comments.json
deleted file mode 100644
index 1f6094e7..00000000
--- a/.experiments/code_review/code_reviews_20240815_142254/no_eval/pr_335_comments.json
+++ /dev/null
@@ -1,17 +0,0 @@
-[
-  {
-    "topic": "Error Handling",
-    "comment": "Removed parameter 'reeval_response' without handling its previous functionality.",
-    "confidence": "critical",
-    "reason": "The removal of 'reeval_response' may lead to unexpected behavior if the function relies on it.",
-    "solution": "Evaluate the necessity of the 'reeval_response' parameter and ensure that its removal does not affect the logic of the code.",
-    "actual_code": "desc = self._process_full_diff(prompt, user, reeval_response)",
-    "fixed_code": "desc = self._process_full_diff(prompt, user)",
-    "file_name": "kaizen/generator/pr_description.py",
-    "start_line": 54,
-    "end_line": 54,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 8
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240815_142254/no_eval/pr_335_review.md b/.experiments/code_review/code_reviews_20240815_142254/no_eval/pr_335_review.md
deleted file mode 100644
index 4f72039b..00000000
--- a/.experiments/code_review/code_reviews_20240815_142254/no_eval/pr_335_review.md
+++ /dev/null
@@ -1,98 +0,0 @@
-# 🔍 Code Review Summary
-
-## 📊 Stats
-- Total Issues: 4
-- Critical: 1
-- Important: 2
-- Minor: 1
-- Files Affected: 1
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Error Handling (1 issues)</strong></summary>
-
-### 1. Removed parameter 'reeval_response' without handling its previous functionality.
-📁 **File:** `kaizen/generator/pr_description.py:54`
-⚖️ **Severity:** 8/10
-🔍 **Description:** The removal of 'reeval_response' may lead to unexpected behavior if the function relies on it.
-💡 **Solution:** Evaluate the necessity of the 'reeval_response' parameter and ensure that its removal does not affect the logic of the code.
-
-**Current Code:**
-```python
-desc = self._process_full_diff(prompt, user, reeval_response)
-```
-
-**Suggested Code:**
-```python
-desc = self._process_full_diff(prompt, user)
-```
-
-</details>
-
-## 🟠 Important Issues
-
-<details>
-<summary><strong>Imports (2 issues)</strong></summary>
-
-### 1. Inconsistent naming of imported prompts.
-📁 **File:** `kaizen/generator/pr_description.py:8`
-⚖️ **Severity:** 5/10
-🔍 **Description:** The change from `code_review_prompts` to `pr_desc_prompts` may lead to confusion if not documented properly.
-💡 **Solution:** Ensure that the new prompt names are well-documented and consistent across the codebase.
-
-**Current Code:**
-```python
-from kaizen.llms.prompts.code_review_prompts import (
-    PR_DESCRIPTION_PROMPT,
-    MERGE_PR_DESCRIPTION_PROMPT,
-    PR_FILE_DESCRIPTION_PROMPT,
-    PR_DESC_EVALUATION_PROMPT,
-    CODE_REVIEW_SYSTEM_PROMPT,
-)
-```
-
-**Suggested Code:**
-```python
-from kaizen.llms.prompts.pr_desc_prompts import (
-    PR_DESCRIPTION_PROMPT,
-    MERGE_PR_DESCRIPTION_PROMPT,
-    PR_FILE_DESCRIPTION_PROMPT,
-    PR_DESCRIPTION_SYSTEM_PROMPT,
-)
-```
-
-### 2. Inconsistent handling of response extraction.
-📁 **File:** `kaizen/generator/pr_description.py:110`
-⚖️ **Severity:** 7/10
-🔍 **Description:** The change from 'chat_completion_with_json' to 'chat_completion' may alter the expected response format.
-💡 **Solution:** Ensure that the new method returns the same structure as the previous one or update the handling logic accordingly.
-
-**Current Code:**
-```python
-resp, usage = self.provider.chat_completion_with_json(prompt, user=user)
-```
-
-**Suggested Code:**
-```python
-resp, usage = self.provider.chat_completion(prompt, user=user)
-```
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
diff --git a/.experiments/code_review/code_reviews_20240815_142254/with_eval/pr_335_comments.json b/.experiments/code_review/code_reviews_20240815_142254/with_eval/pr_335_comments.json
deleted file mode 100644
index 2eaf7ef9..00000000
--- a/.experiments/code_review/code_reviews_20240815_142254/with_eval/pr_335_comments.json
+++ /dev/null
@@ -1,17 +0,0 @@
-[
-  {
-    "topic": "Functionality",
-    "comment": "Changing the method from 'chat_completion_with_json' to 'chat_completion' may alter expected behavior.",
-    "confidence": "critical",
-    "reason": "If 'chat_completion_with_json' was designed to handle specific JSON formatting, switching to 'chat_completion' may lead to data handling issues.",
-    "solution": "Review the implementation of 'chat_completion' to ensure it meets the requirements previously handled by 'chat_completion_with_json'.",
-    "actual_code": "resp, usage = self.provider.chat_completion_with_json(prompt, user=user)",
-    "fixed_code": "resp, usage = self.provider.chat_completion(prompt, user=user)",
-    "file_name": "kaizen/generator/pr_description.py",
-    "start_line": 83,
-    "end_line": 83,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 8
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240815_142254/with_eval/pr_335_review.md b/.experiments/code_review/code_reviews_20240815_142254/with_eval/pr_335_review.md
deleted file mode 100644
index 8eefd8ac..00000000
--- a/.experiments/code_review/code_reviews_20240815_142254/with_eval/pr_335_review.md
+++ /dev/null
@@ -1,114 +0,0 @@
-# 🔍 Code Review Summary
-
-## 📊 Stats
-- Total Issues: 5
-- Critical: 1
-- Important: 3
-- Minor: 1
-- Files Affected: 1
-## 🏆 Code Quality
-[████████████████░░░░] 80% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Functionality (1 issues)</strong></summary>
-
-### 1. Changing the method from 'chat_completion_with_json' to 'chat_completion' may alter expected behavior.
-📁 **File:** `kaizen/generator/pr_description.py:83`
-⚖️ **Severity:** 8/10
-🔍 **Description:** If 'chat_completion_with_json' was designed to handle specific JSON formatting, switching to 'chat_completion' may lead to data handling issues.
-💡 **Solution:** Review the implementation of 'chat_completion' to ensure it meets the requirements previously handled by 'chat_completion_with_json'.
-
-**Current Code:**
-```python
-resp, usage = self.provider.chat_completion_with_json(prompt, user=user)
-```
-
-**Suggested Code:**
-```python
-resp, usage = self.provider.chat_completion(prompt, user=user)
-```
-
-</details>
-
-## 🟠 Important Issues
-
-<details>
-<summary><strong>Imports (3 issues)</strong></summary>
-
-### 1. Updated import statements may lead to confusion regarding the source of prompts.
-📁 **File:** `kaizen/generator/pr_description.py:8`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Changing the import path for prompts can lead to issues if the new module does not contain the expected constants.
-💡 **Solution:** Ensure that the new import path is correct and that all necessary constants are defined in the new module.
-
-**Current Code:**
-```python
-from kaizen.llms.prompts.code_review_prompts import (
-    PR_DESCRIPTION_PROMPT,
-    MERGE_PR_DESCRIPTION_PROMPT,
-    PR_FILE_DESCRIPTION_PROMPT,
-    PR_DESC_EVALUATION_PROMPT,
-    CODE_REVIEW_SYSTEM_PROMPT,
-)
-```
-
-**Suggested Code:**
-```python
-from kaizen.llms.prompts.pr_desc_prompts import (
-    PR_DESCRIPTION_PROMPT,
-    MERGE_PR_DESCRIPTION_PROMPT,
-    PR_FILE_DESCRIPTION_PROMPT,
-    PR_DESCRIPTION_SYSTEM_PROMPT,
-)
-```
-
-### 2. Raising a generic Exception can obscure the cause of errors.
-📁 **File:** `kaizen/generator/pr_description.py:51`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Using a generic Exception does not provide specific information about the error, making debugging difficult.
-💡 **Solution:** Use a more specific exception type or create a custom exception class to provide better context.
-
-**Current Code:**
-```python
-raise Exception("Both diff_text and pull_request_files are empty!")
-```
-
-**Suggested Code:**
-```python
-raise ValueError("Both diff_text and pull_request_files are empty!")
-```
-
-### 3. Removing 'reeval_response' from multiple function signatures may lead to loss of intended functionality.
-📁 **File:** `kaizen/generator/pr_description.py:40`
-⚖️ **Severity:** 6/10
-🔍 **Description:** If 'reeval_response' was previously used to control logic, its removal could lead to unintended behavior.
-💡 **Solution:** Carefully assess the logic that relies on 'reeval_response' to determine if it should be retained.
-
-**Current Code:**
-```python
-def _process_full_diff(self, prompt: str, user: Optional[str], reeval_response: bool) -> str:
-```
-
-**Suggested Code:**
-```python
-def _process_full_diff(self, prompt: str, user: Optional[str]) -> str:
-```
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
diff --git a/.experiments/code_review/dataset/pr_222/issues.json b/.experiments/code_review/dataset/pr_222/issues.json
new file mode 100644
index 00000000..c50260a4
--- /dev/null
+++ b/.experiments/code_review/dataset/pr_222/issues.json
@@ -0,0 +1,152 @@
+[
+  {
+    "topic": "Security",
+    "comment": "Hardcoding API keys in `config.json` can lead to security vulnerabilities.",
+    "confidence": "critical",
+    "reason": "Exposing API keys in the codebase can lead to unauthorized access. This was highlighted by multiple models and is a critical security issue.",
+    "solution": "Use environment variables to store API keys instead of hardcoding them.",
+    "actual_code": "\"api_key\": \"os.environ/AZURE_API_KEY\"",
+    "fixed_code": "\"api_key\": \"${AZURE_API_KEY}\"",
+    "file_name": "config.json",
+    "start_line": 13,
+    "end_line": 13,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "SQL Injection",
+    "comment": "Potential SQL injection vulnerability in the query construction.",
+    "confidence": "critical",
+    "reason": "Using string interpolation for SQL queries can lead to SQL injection attacks. This was identified by multiple models as a critical issue.",
+    "solution": "Use parameterized queries to avoid SQL injection vulnerabilities.",
+    "actual_code": "query = f\"\"\"\nSELECT \n    e.node_id,\n    e.text,\n    e.metadata,\n    1 - (e.embedding <=> %s::vector) as similarity\nFROM \n{self.table_name}e\nJOIN \n    function_abstractions fa ON e.node_id = fa.function_id::text\nJOIN \n    files f ON fa.file_id = f.file_id\nWHERE \n    f.repo_id = %s\nORDER BY \n    similarity DESC\nLIMIT \n    %s\n\"\"\"",
+    "fixed_code": "query = \"\"\"\nSELECT \n    e.node_id,\n    e.text,\n    e.metadata,\n    1 - (e.embedding <=> %s::vector) as similarity\nFROM \n    %s e\nJOIN \n    function_abstractions fa ON e.node_id = fa.function_id::text\nJOIN \n    files f ON fa.file_id = f.file_id\nWHERE \n    f.repo_id = %s\nORDER BY \n    similarity DESC\nLIMIT \n    %s\n\"\"\"",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 19,
+    "end_line": 37,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Lack of error handling in database operations.",
+    "confidence": "important",
+    "reason": "Multiple models identified the need for better error handling in database operations to prevent crashes and improve debugging.",
+    "solution": "Add try-except blocks to handle potential database errors.",
+    "actual_code": "",
+    "fixed_code": "try:\n    with self.get_client() as client:\n        with client.cursor() as cur:\n            cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))\n            results = cur.fetchall()\nexcept Exception as e:\n    # Handle exception (e.g., log the error, re-raise, etc.)\n    raise e",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 39,
+    "end_line": 42,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Code Readability",
+    "comment": "The `chunk_code` function in `code_chunker.py` has nested functions and complex logic that can be refactored for better readability.",
+    "confidence": "important",
+    "reason": "Complex functions with nested logic can be hard to maintain and understand. This was noted by multiple models.",
+    "solution": "Refactor the `chunk_code` function to extract nested functions into separate helper functions.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/code_chunker.py",
+    "start_line": 7,
+    "end_line": 62,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "Type Annotations",
+    "comment": "Missing or incomplete type annotations for method parameters and return types.",
+    "confidence": "important",
+    "reason": "Type annotations improve code readability and help with static analysis. This was mentioned by several models.",
+    "solution": "Add or improve type annotations to method parameters and return types.",
+    "actual_code": "def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:",
+    "fixed_code": "def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[Dict[str, Any]]:",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 13,
+    "end_line": 13,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Documentation",
+    "comment": "Lack of comments and docstrings in complex functions and methods.",
+    "confidence": "important",
+    "reason": "Several models noted the need for better documentation to improve code understanding and maintainability.",
+    "solution": "Add comments explaining complex logic and docstrings to functions and methods.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 1,
+    "end_line": 1,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "Code Duplication",
+    "comment": "Duplicate code found in test cases and database connection string creation.",
+    "confidence": "important",
+    "reason": "Code duplication was identified by multiple models as an issue that can lead to maintenance problems.",
+    "solution": "Refactor duplicate code into reusable functions or constants.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "tests/retriever/test_chunker.py",
+    "start_line": 98,
+    "end_line": 101,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "Dependency Management",
+    "comment": "New dependencies added without justification.",
+    "confidence": "important",
+    "reason": "Adding dependencies increases the attack surface and maintenance burden. This was noted as an important consideration by multiple models.",
+    "solution": "Review and justify new dependencies, ensuring they are necessary and compatible.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "pyproject.toml",
+    "start_line": 27,
+    "end_line": 49,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "Performance",
+    "comment": "Potential performance issues in database operations and code parsing.",
+    "confidence": "moderate",
+    "reason": "Several models identified areas where performance could be improved, particularly in database operations and file parsing.",
+    "solution": "Optimize database queries, consider batching operations, and review file parsing logic for potential improvements.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 1,
+    "end_line": 1,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Improve error handling in the parse_file method and LanguageLoader class.",
+    "confidence": "important",
+    "reason": "Better error handling was suggested by multiple models to improve debugging and prevent unexpected behavior.",
+    "solution": "Implement more specific exception handling and provide detailed error messages.",
+    "actual_code": "except Exception as e:\n    logger.error(f\"Error processing file{file_path}:{str(e)}\")\n    logger.error(traceback.format_exc())",
+    "fixed_code": "except Exception as e:\n    logger.error(f\"Error processing file{file_path}:{str(e)}\")\n    logger.error(traceback.format_exc())\n    raise",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 108,
+    "end_line": 110,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/dataset/pr_232/issues.json b/.experiments/code_review/dataset/pr_232/issues.json
new file mode 100644
index 00000000..8ddf3398
--- /dev/null
+++ b/.experiments/code_review/dataset/pr_232/issues.json
@@ -0,0 +1,107 @@
+[
+  {
+    "topic": "Unused Imports",
+    "comment": "There are several unused imports across multiple files that should be removed.",
+    "confidence": "important",
+    "reason": "Removing unused imports improves code cleanliness, readability, and potentially reduces bundle size. This issue was identified by both models across multiple files.",
+    "solution": "Remove all unused imports from the affected files.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "page.tsx, queryinput.tsx, apps/web/app/(dash)/home/page.tsx, apps/web/app/(dash)/home/queryinput.tsx, packages/ui/shadcn/combobox.tsx",
+    "start_line": 0,
+    "end_line": 0,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Type Annotations and Definitions",
+    "comment": "Some variables, functions, and components are missing proper type annotations or definitions.",
+    "confidence": "important",
+    "reason": "Proper type annotations improve code readability, maintainability, and help catch type-related errors at compile-time. This issue was noted by both models.",
+    "solution": "Add or improve type annotations for variables, functions, and components where they are missing or inadequate.",
+    "actual_code": "const ComboboxWithCreate = ({",
+    "fixed_code": "const ComboboxWithCreate: React.FC<ComboboxWithCreateProps> = ({",
+    "file_name": "queryinput.tsx, packages/ui/shadcn/combobox.tsx, apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 32,
+    "end_line": 32,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Code Organization and Structure",
+    "comment": "Some files contain multiple unrelated components or have poor code organization.",
+    "confidence": "important",
+    "reason": "Proper code organization improves readability, maintainability, and reusability. This issue was identified by both models.",
+    "solution": "Separate unrelated components into their own files and improve overall code structure.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "page.tsx, apps/web/app/(dash)/menu.tsx",
+    "start_line": 0,
+    "end_line": 0,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Improve error handling in various parts of the code, particularly in the handleSubmit function.",
+    "confidence": "important",
+    "reason": "Proper error handling is crucial for preventing crashes and providing useful feedback. This issue was highlighted by both models.",
+    "solution": "Implement robust error handling, especially in critical functions like handleSubmit.",
+    "actual_code": "throw new Error(`Memory creation failed: ${cont.error}`);\nreturn cont;",
+    "fixed_code": "throw new Error(`Memory creation failed: ${cont.error}`);",
+    "file_name": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 230,
+    "end_line": 231,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "State Management",
+    "comment": "Consider improving state management to avoid prop drilling and improve component encapsulation.",
+    "confidence": "moderate",
+    "reason": "Better state management can improve code maintainability and reduce complexity. This was suggested by the Sonnet model.",
+    "solution": "Consider using React Context or a state management library for managing global state.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 163,
+    "end_line": 167,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Performance Optimization",
+    "comment": "Some computations, like filtering options, could be optimized to improve performance.",
+    "confidence": "moderate",
+    "reason": "Optimizing expensive computations can lead to better performance, especially for larger datasets.",
+    "solution": "Use memoization techniques like useMemo for expensive computations that don't need to be recalculated on every render.",
+    "actual_code": "const filteredOptions = options.filter(\n\t\t(option) => !selectedSpaces.includes(parseInt(option.value)),\n\t);",
+    "fixed_code": "const filteredOptions = useMemo(() => options.filter(\n\t\t(option) => !selectedSpaces.includes(parseInt(option.value)),\n\t),[options, selectedSpaces]);",
+    "file_name": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 55,
+    "end_line": 57,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Accessibility",
+    "comment": "Some UI elements lack proper accessibility attributes.",
+    "confidence": "moderate",
+    "reason": "Improving accessibility ensures the application is usable by all users, including those with disabilities.",
+    "solution": "Add appropriate aria-labels and other accessibility attributes to interactive elements.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 65,
+    "end_line": 72,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/dataset/pr_252/issues.json b/.experiments/code_review/dataset/pr_252/issues.json
new file mode 100644
index 00000000..568ce90d
--- /dev/null
+++ b/.experiments/code_review/dataset/pr_252/issues.json
@@ -0,0 +1,92 @@
+[
+  {
+    "topic": "Code Structure and Consistency",
+    "comment": "There are inconsistencies in code formatting and structure across different function calls.",
+    "confidence": "moderate",
+    "reason": "Consistent code structure and formatting improves readability and maintainability. This issue was noted by multiple models.",
+    "solution": "Standardize the formatting of function calls, particularly for `generate_twitter_post` and `generate_linkedin_post`. Consider using multi-line formatting for both for consistency.",
+    "actual_code": "twitter_post = work_summary_generator.generate_twitter_post(summary, user=\"oss_example\")\n\nlinkedin_post = work_summary_generator.generate_linkedin_post(\n    summary, user=\"oss_example\"\n)",
+    "fixed_code": "twitter_post = work_summary_generator.generate_twitter_post(\n    summary, user=\"oss_example\"\n)\n\nlinkedin_post = work_summary_generator.generate_linkedin_post(\n    summary, user=\"oss_example\"\n)",
+    "file_name": "examples/work_summarizer/main.py",
+    "start_line": 59,
+    "end_line": 62,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Code Organization",
+    "comment": "The `WorkSummaryGenerator` class has multiple responsibilities and could be refactored for better organization.",
+    "confidence": "important",
+    "reason": "Separation of Concerns (SoC) principle improves code maintainability and readability.",
+    "solution": "Refactor the `WorkSummaryGenerator` class into separate classes or functions for each responsibility (e.g., summary generation, Twitter post generation, LinkedIn post generation).",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/reviewer/work_summarizer.py",
+    "start_line": 0,
+    "end_line": 0,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "The `generate_twitter_post` and `generate_linkedin_post` methods lack error handling.",
+    "confidence": "important",
+    "reason": "Proper error handling improves code robustness and helps with debugging.",
+    "solution": "Add try-except blocks to handle and log any exceptions during the post generation process.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/reviewer/work_summarizer.py",
+    "start_line": 58,
+    "end_line": 74,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 7
+  },
+  {
+    "topic": "Code Duplication",
+    "comment": "There is code duplication in the `generate_twitter_post` and `generate_linkedin_post` methods, and a duplicated print statement for LinkedIn post.",
+    "confidence": "important",
+    "reason": "Code duplication violates the DRY principle and can lead to maintenance issues.",
+    "solution": "Extract common code from `generate_twitter_post` and `generate_linkedin_post` into a shared method. Remove the duplicated print statement for the LinkedIn post.",
+    "actual_code": "print(f\" LinkedIn Post: \\n{linkedin_post}\\n\")",
+    "fixed_code": "",
+    "file_name": "kaizen/reviewer/work_summarizer.py, examples/work_summarizer/main.py",
+    "start_line": 58,
+    "end_line": 74,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Code Documentation",
+    "comment": "The `severity_level` field in the code review prompt lacks detailed explanation.",
+    "confidence": "moderate",
+    "reason": "Clear documentation helps users understand how to use features correctly.",
+    "solution": "Add a more detailed explanation of what each severity level represents in the code review prompt.",
+    "actual_code": "For \"severity_level\" score in range of 1 to 10, 1 being not severe and 10 being critical.",
+    "fixed_code": "For \"severity_level\" score in range of 1 to 10:\n1-3: Minor issues (style, small optimizations)\n4-6: Moderate issues (potential bugs, performance concerns)\n7-8: Major issues (definite bugs, security vulnerabilities)\n9-10: Critical issues (severe security risks, system-breaking bugs)",
+    "file_name": "kaizen/llms/prompts/code_review_prompts.py",
+    "start_line": 100,
+    "end_line": 100,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Prompt Formatting",
+    "comment": "The TWITTER_POST_PROMPT and LINKEDIN_POST_PROMPT could be improved for better readability.",
+    "confidence": "moderate",
+    "reason": "Well-formatted prompts are easier to read and maintain.",
+    "solution": "Break the prompts into multiple lines, use string formatting, and add comments to explain different sections.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/llms/prompts/work_summary_prompts.py",
+    "start_line": 44,
+    "end_line": 65,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/dataset/pr_335/issues.json b/.experiments/code_review/dataset/pr_335/issues.json
new file mode 100644
index 00000000..cb9843ab
--- /dev/null
+++ b/.experiments/code_review/dataset/pr_335/issues.json
@@ -0,0 +1,92 @@
+[
+  {
+    "topic": "Import Changes",
+    "comment": "Import statements have been changed and some may be unused.",
+    "confidence": "important",
+    "reason": "Changing import paths can lead to runtime errors and unused imports clutter the code. This issue was identified by multiple models.",
+    "solution": "Verify that all new import paths are correct, remove any unused imports, and ensure consistency across the codebase.",
+    "actual_code": "from kaizen.llms.prompts.pr_desc_prompts import (",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 8,
+    "end_line": 8,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "Removal of Reevaluation Logic",
+    "comment": "The 'reeval_response' parameter and associated logic have been removed from multiple functions.",
+    "confidence": "critical",
+    "reason": "Removing this parameter and logic could significantly change the behavior of the PR description generation. This was noted as a critical issue by multiple models.",
+    "solution": "Carefully review the impact of removing the reevaluation logic. Ensure that the quality of PR descriptions is maintained without this feature. Consider adding unit tests to verify the new behavior.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 43,
+    "end_line": 96,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "API Change",
+    "comment": "Changed from 'chat_completion_with_json' to 'chat_completion'",
+    "confidence": "important",
+    "reason": "This API change could affect the format of the response and how it's processed. Multiple models highlighted this as an important change.",
+    "solution": "Ensure that the new chat_completion method returns the expected format. Update any dependent code that might be affected by this change. Verify that the response parsing is adjusted accordingly.",
+    "actual_code": "resp, usage = self.provider.chat_completion(prompt, user=user)",
+    "fixed_code": "resp, usage = self.provider.chat_completion(prompt, user=user)\ndesc = parser.extract_code_from_markdown(resp)",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 79,
+    "end_line": 80,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 7
+  },
+  {
+    "topic": "Prompt Changes",
+    "comment": "Significant changes to PR description prompts and system prompts.",
+    "confidence": "important",
+    "reason": "The prompts have been restructured and moved to a new file. This could impact the quality and structure of generated PR descriptions.",
+    "solution": "Review the new prompt structure to ensure it meets all requirements. Test thoroughly to verify that the generated PR descriptions maintain or improve quality. Update any related documentation.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/llms/prompts/pr_desc_prompts.py",
+    "start_line": 1,
+    "end_line": 92,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 7
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Potential lack of error handling for exceptions in PR description generation.",
+    "confidence": "important",
+    "reason": "Proper error handling is crucial for preventing unexpected crashes and providing useful feedback.",
+    "solution": "Implement try-except blocks where appropriate to handle potential exceptions gracefully. Consider using more specific exception types.",
+    "actual_code": "raise Exception(\"Both diff_text and pull_request_files are empty!\")",
+    "fixed_code": "raise ValueError(\"Both diff_text and pull_request_files are empty!\")",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 51,
+    "end_line": 51,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Code Style and Documentation",
+    "comment": "Various minor issues with code style, variable naming, and documentation.",
+    "confidence": "moderate",
+    "reason": "Consistent code style and proper documentation improve readability and maintainability.",
+    "solution": "Review and update variable names to follow PEP 8 conventions. Add docstrings or comments explaining the purpose of new prompts and significant changes.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/llms/prompts/pr_desc_prompts.py, kaizen/generator/pr_description.py",
+    "start_line": 1,
+    "end_line": 1,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/dataset/pr_400/issues.json b/.experiments/code_review/dataset/pr_400/issues.json
new file mode 100644
index 00000000..1bc6c452
--- /dev/null
+++ b/.experiments/code_review/dataset/pr_400/issues.json
@@ -0,0 +1,52 @@
+[
+  {
+    "topic": "Test Coverage",
+    "comment": "Improved test coverage with more comprehensive test cases, including edge cases and error handling scenarios.",
+    "confidence": "important",
+    "reason": "Comprehensive test coverage is crucial for maintaining code quality, catching potential bugs, and ensuring the reliability of the code.",
+    "solution": "Continue to add test cases for edge cases and ensure all new functionality is covered by tests.",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py, .kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py, .kaizen/unit_test/kaizen/helpers/test_create_test_files.py, .kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "sentiment": "positive",
+    "severity_level": 7
+  },
+  {
+    "topic": "Code Structure and Organization",
+    "comment": "Improved code structure with better modularity, organization, and use of pytest fixtures.",
+    "confidence": "important",
+    "reason": "Good organization improves readability, maintainability, and allows for better separation of concerns.",
+    "solution": "Continue to monitor for potential further improvements in code structure and organization.",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py, kaizen/generator/unit_test.py",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Insufficient error handling in some methods, particularly for file operations.",
+    "confidence": "important",
+    "reason": "Proper error handling ensures the program remains stable and provides useful error messages.",
+    "solution": "Implement try-except blocks to handle potential errors, especially in file operations and network requests.",
+    "file_name": "kaizen/generator/unit_test.py",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Logging Configuration",
+    "comment": "The logging configuration might override existing setups and is mixed with import statements.",
+    "confidence": "important",
+    "reason": "Inconsistent logging configuration can lead to loss of important log information and poor code organization.",
+    "solution": "Adjust the logging configuration to respect the LOGLEVEL environment variable and move it to a separate section after all imports.",
+    "file_name": "kaizen/llms/provider.py",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Configuration Update",
+    "comment": "Changes made to sensitive configuration file",
+    "confidence": "critical",
+    "reason": "Changes to config.json may affect system functionality and security",
+    "solution": "Review the changes in config.json carefully, especially the new 'base_model' fields, and ensure that the code using this configuration is updated accordingly.",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/dataset/pr_476/issues.json b/.experiments/code_review/dataset/pr_476/issues.json
new file mode 100644
index 00000000..c426f116
--- /dev/null
+++ b/.experiments/code_review/dataset/pr_476/issues.json
@@ -0,0 +1,68 @@
+[
+  {
+    "topic": "Error Handling",
+    "comment": "Broad exception handling with generic error message",
+    "confidence": "critical",
+    "reason": "Using a generic 'except Exception' block with a non-specific error message can mask important errors and make debugging difficult.",
+    "solution": "Catch specific exceptions where possible and provide more informative error messages. Consider using proper logging instead of print statements.",
+    "actual_code": "except Exception:\n    print(\"Error\")",
+    "fixed_code": "except KeyError as e:\n    logger.error(f\"Invalid confidence level: {e}\")\nexcept Exception as e:\n    logger.error(f\"Unexpected error: {e}\")",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 140,
+    "end_line": 141,
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "Code Efficiency",
+    "comment": "Inefficient sorting implementation",
+    "confidence": "important",
+    "reason": "The custom sorting logic in 'sort_files' function is unnecessarily complex and inefficient for large lists.",
+    "solution": "Use Python's built-in sorted() function with a key function for better performance and readability.",
+    "actual_code": "def sort_files(files):\n    sorted_files = []\n    for file in files:\n        min_index = len(sorted_files)\n        file_name = file[\"filename\"]\n        for i, sorted_file in enumerate(sorted_files):\n            if file_name < sorted_file[\"filename\"]:\n                min_index = i\n                break\n        sorted_files.insert(min_index, file)\n    return sorted_files",
+    "fixed_code": "def sort_files(files):\n    return sorted(files, key=lambda x: x[\"filename\"])",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 184,
+    "end_line": 194,
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Code Simplification",
+    "comment": "Overly verbose implementation of generate_tests function",
+    "confidence": "moderate",
+    "reason": "The current implementation of generate_tests function can be simplified using a list comprehension.",
+    "solution": "Use a list comprehension to create the list of filenames.",
+    "actual_code": "def generate_tests(pr_files):\n    return [f[\"filename\"] for f in pr_files]",
+    "fixed_code": "def generate_tests(pr_files):\n    return [f[\"filename\"] for f in pr_files]",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 199,
+    "end_line": 200,
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Logging and Debugging",
+    "comment": "Inconsistent use of print statements for debugging",
+    "confidence": "important",
+    "reason": "Using print statements for debugging can clutter the code and make it difficult to control log levels in different environments.",
+    "solution": "Replace print statements with proper logging calls using Python's logging module.",
+    "actual_code": "print(\"diff: \", diff_text)\nprint(\"pr_files\", pr_files)",
+    "fixed_code": "import logging\n\nlogger = logging.getLogger(__name__)\nlogger.debug(f\"diff: {diff_text}\")\nlogger.debug(f\"pr_files: {pr_files}\")",
+    "file_name": "examples/code_review/main.py",
+    "start_line": 21,
+    "end_line": 22,
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Configuration Changes",
+    "comment": "Changes made to sensitive configuration file",
+    "confidence": "critical",
+    "reason": "Changes to config.json may affect system functionality and security. The removal of 'enable_observability_logging' option needs to be properly documented.",
+    "solution": "Review all changes in config.json carefully. If removing features, provide a migration guide or deprecation notice for existing users.",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/dataset/pr_5/issues.json b/.experiments/code_review/dataset/pr_5/issues.json
new file mode 100644
index 00000000..693b3000
--- /dev/null
+++ b/.experiments/code_review/dataset/pr_5/issues.json
@@ -0,0 +1,93 @@
+[
+  {
+    "topic": "Unused Import",
+    "comment": "The 'random' module is imported but never used in the code.",
+    "confidence": "trivial",
+    "reason": "The 'random' module is imported but not utilized in the code.",
+    "solution": "Remove the unused import statement for 'random'.",
+    "actual_code": "import random  # Unused import",
+    "fixed_code": "",
+    "file_name": "main.py",
+    "start_line": 8,
+    "end_line": 8,
+    "severity_level": 1
+  },
+  {
+    "topic": "API Call Error Handling",
+    "comment": "The API call to 'completion' lacks a retry mechanism.",
+    "confidence": "critical",
+    "reason": "API calls can fail due to network issues or server errors, and without a retry mechanism, the function may fail unexpectedly.",
+    "solution": "Implement a retry mechanism with exponential backoff for the API call.",
+    "actual_code": "response = completion(\n    model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n)",
+    "fixed_code": "import time\n\nfor attempt in range(3):\n    try:\n        response = completion(\n            model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n        )\n        break\n    except Exception as e:\n        if attempt < 2:\n            time.sleep(2 ** attempt)\n        else:\n            raise e",
+    "file_name": "main.py",
+    "start_line": 66,
+    "end_line": 68,
+    "severity_level": 9
+  },
+  {
+    "topic": "Silent Failure in JSON Parsing",
+    "comment": "The exception handling for JSON decoding fails silently without logging.",
+    "confidence": "critical",
+    "reason": "Silent failures make it difficult to diagnose issues when they occur.",
+    "solution": "Add logging to capture the exception details.",
+    "actual_code": "except json.JSONDecodeError:\n    result = {",
+    "fixed_code": "except json.JSONDecodeError as e:\n    print(f\"Failed to parse content for applicant: {e}\")\n    result = {",
+    "file_name": "main.py",
+    "start_line": 82,
+    "end_line": 84,
+    "severity_level": 8
+  },
+  {
+    "topic": "Inefficient Progress Printing",
+    "comment": "The progress printing method is inefficient.",
+    "confidence": "important",
+    "reason": "Printing progress in this manner can be slow and resource-intensive.",
+    "solution": "Use a more efficient method for printing progress, such as updating the progress less frequently or using a dedicated progress reporting library like tqdm.",
+    "actual_code": "print(f\"\\rProgress:[{'=' * int(50 * progress):<50}]{progress:.0%}\", end=\"\", flush=True)",
+    "fixed_code": "if index % max(1, len(df) // 100) == 0:  # Update every 1%\n    print(f\"\\rProgress:[{'=' * int(50 * progress):<50}]{progress:.0%}\", end=\"\", flush=True)",
+    "file_name": "main.py",
+    "start_line": 121,
+    "end_line": 122,
+    "severity_level": 5
+  },
+  {
+    "topic": "Redundant Code",
+    "comment": "The check for an empty DataFrame is redundant.",
+    "confidence": "moderate",
+    "reason": "The code already handles an empty DataFrame gracefully, so this check is unnecessary.",
+    "solution": "Remove the redundant check for an empty DataFrame.",
+    "actual_code": "if len(df) == 0:\n    return",
+    "fixed_code": "",
+    "file_name": "main.py",
+    "start_line": 142,
+    "end_line": 143,
+    "severity_level": 3
+  },
+  {
+    "topic": "Division by Zero",
+    "comment": "Potential division by zero when calculating total tokens.",
+    "confidence": "critical",
+    "reason": "If 'total_tokens' is zero, it will cause a division by zero error.",
+    "solution": "Add a check to ensure 'total_tokens' is not zero before performing the division.",
+    "actual_code": "print(f\"Total tokens used: {total_tokens:,}\")\nprint(f\"  - Input tokens: {total_input_tokens:,}\")\nprint(f\"  - Output tokens: {total_output_tokens:,}\")",
+    "fixed_code": "print(f\"Total tokens used: {total_tokens:,}\")\nif total_tokens > 0:\n    print(f\"  - Input tokens: {total_input_tokens:,} ({total_input_tokens/total_tokens:.2%})\")\n    print(f\"  - Output tokens: {total_output_tokens:,} ({total_output_tokens/total_tokens:.2%})\")\nelse:\n    print(\"  - No tokens used.\")",
+    "file_name": "main.py",
+    "start_line": 158,
+    "end_line": 163,
+    "severity_level": 7
+  },
+  {
+    "topic": "File Not Found Handling",
+    "comment": "No error handling for file not found.",
+    "confidence": "important",
+    "reason": "If the specified file does not exist, the program will crash.",
+    "solution": "Add error handling to check if the file exists before processing.",
+    "actual_code": "main(input_file)",
+    "fixed_code": "try:\n    main(input_file)\nexcept FileNotFoundError:\n    print(f\"Error: The file '{input_file}' does not exist. Please check the file path and try again.\")\nexcept Exception as e:\n    print(f\"An error occurred: {e}\")",
+    "file_name": "main.py",
+    "start_line": 174,
+    "end_line": 175,
+    "severity_level": 6
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/evaluate.py b/.experiments/code_review/evaluate.py
new file mode 100644
index 00000000..d36faecb
--- /dev/null
+++ b/.experiments/code_review/evaluate.py
@@ -0,0 +1,179 @@
+import json
+import os
+from fuzzywuzzy import fuzz
+
+
+def load_json(file_path):
+    with open(file_path, "r") as f:
+        return json.load(f)
+
+
+def fuzzy_match(str1, str2, threshold=50):
+    return fuzz.partial_ratio(str1.lower(), str2.lower()) >= threshold
+
+
+def compare_issues(ground_truth, model_issues):
+    matched = []
+    unmatched_ground_truth = []
+    unmatched_model = []
+
+    for gt_issue in ground_truth:
+        found_match = False
+        for model_issue in model_issues:
+            if (
+                fuzzy_match(gt_issue["topic"], model_issue["topic"])
+                and fuzzy_match(gt_issue["comment"], model_issue["comment"])
+                and gt_issue["file_name"] == model_issue["file_name"]
+                and abs(
+                    int(gt_issue.get("start_line", 0))
+                    - int(model_issue.get("start_line", -10))
+                )
+                <= 1
+                and abs(
+                    int(gt_issue.get("end_line", 0))
+                    - int(model_issue.get("end_line", -10))
+                )
+                <= 1
+                and abs(
+                    int(gt_issue.get("severity_level", 0))
+                    - int(model_issue.get("severity_level", -10))
+                )
+                <= 1
+                and gt_issue.get("sentiment", "bad")
+                == model_issue.get("sentiment", "hmm")
+            ):
+                matched.append((gt_issue, model_issue))
+                found_match = True
+                break
+
+        if not found_match:
+            unmatched_ground_truth.append(gt_issue)
+
+    for model_issue in model_issues:
+        if not any(model_issue in pair for pair in matched):
+            unmatched_model.append(model_issue)
+
+    return matched, unmatched_ground_truth, unmatched_model
+
+
+def evaluate_model(ground_truth, model_issues):
+    matched, unmatched_gt, unmatched_model = compare_issues(ground_truth, model_issues)
+
+    total_issues = len(ground_truth)
+    issues_found = len(model_issues)
+    correct_issues = len(matched)
+    false_positives = len(unmatched_model)
+    false_negatives = len(unmatched_gt)
+
+    precision = (
+        correct_issues / (correct_issues + false_positives)
+        if (correct_issues + false_positives) > 0
+        else 0
+    )
+    recall = correct_issues / total_issues
+    f1_score = (
+        2 * (precision * recall) / (precision + recall)
+        if (precision + recall) > 0
+        else 0
+    )
+
+    return {
+        "total_issues": total_issues,
+        "issues_found": issues_found,
+        "correct_issues": correct_issues,
+        "false_positives": false_positives,
+        "false_negatives": false_negatives,
+        "precision": precision,
+        "recall": recall,
+        "f1_score": f1_score,
+    }
+
+
+def main(folder_name):
+    dataset_path = ".experiments/code_review/dataset"
+    model_base_path = os.path.join(
+        ".experiments", "code_review", folder_name, "no_eval"
+    )
+
+    overall_results = {
+        "total_issues": 0,
+        "correct_issues": 0,
+        "false_positives": 0,
+        "false_negatives": 0,
+    }
+
+    pr_count = 0
+
+    for pr_folder in os.listdir(dataset_path):
+        if pr_folder.startswith("pr_"):
+            pr_number = pr_folder.split("_")[1]
+            ground_truth_path = os.path.join(dataset_path, pr_folder, "issues.json")
+            model_path = os.path.join(model_base_path, f"pr_{pr_number}", "issues.json")
+
+            if not os.path.exists(ground_truth_path):
+                print(f"Ground truth file not found for PR {pr_number}")
+                continue
+            if not os.path.exists(model_path):
+                print(
+                    f"Model output file not found for {folder_name} on PR {pr_number}"
+                )
+                continue
+
+            ground_truth = load_json(ground_truth_path)
+            model_issues = load_json(model_path)
+
+            results = evaluate_model(ground_truth, model_issues)
+
+            print(f"\nEvaluation Results for {folder_name} on PR {pr_number}:")
+            print(f"  Issues Found: {results['issues_found']}")
+            print(
+                f"  Correct issues: {results['correct_issues']}/{results['total_issues']}"
+            )
+            print(f"  False positives: {results['false_positives']}")
+            print(f"  False negatives: {results['false_negatives']}")
+            print(f"  Precision: {results['precision']:.2f}")
+            print(f"  Recall: {results['recall']:.2f}")
+            print(f"  F1 Score: {results['f1_score']:.2f}")
+
+            for key in [
+                "total_issues",
+                "correct_issues",
+                "false_positives",
+                "false_negatives",
+            ]:
+                overall_results[key] += results[key]
+
+            pr_count += 1
+
+    if pr_count > 0:
+        overall_precision = overall_results["correct_issues"] / (
+            overall_results["correct_issues"] + overall_results["false_positives"]
+        )
+        overall_recall = (
+            overall_results["correct_issues"] / overall_results["total_issues"]
+        )
+        overall_f1 = (
+            2
+            * (overall_precision * overall_recall)
+            / (overall_precision + overall_recall)
+            if (overall_precision + overall_recall) > 0
+            else 0
+        )
+
+        print(f"\nOverall Results for {folder_name}:")
+        print(f"  Total PRs evaluated: {pr_count}")
+        print(
+            f"  Correct issues: {overall_results['correct_issues']}/{overall_results['total_issues']}"
+        )
+        print(f"  False positives: {overall_results['false_positives']}")
+        print(f"  False negatives: {overall_results['false_negatives']}")
+        print(f"  Precision: {overall_precision:.2f}")
+        print(f"  Recall: {overall_recall:.2f}")
+        print(f"  F1 Score: {overall_f1:.2f}")
+    else:
+        print(f"No valid PRs found for evaluation of {folder_name}")
+
+
+if __name__ == "__main__":
+    folder_name = input("Enter the model name (e.g., gpt-4o): ")
+    main(folder_name)
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_222/comments.json b/.experiments/code_review/gpt-4o-mini/no_eval/pr_222/comments.json
new file mode 100644
index 00000000..e09425aa
--- /dev/null
+++ b/.experiments/code_review/gpt-4o-mini/no_eval/pr_222/comments.json
@@ -0,0 +1,102 @@
+[
+  {
+    "topic": "Environment Variables",
+    "comment": "Using 'os.environ' directly in JSON is not valid.",
+    "confidence": "critical",
+    "reason": "Environment variables should be accessed in the code, not hardcoded in configuration files.",
+    "solution": "Use a placeholder or variable in the code to fetch environment variables.",
+    "actual_code": "                    \"api_key\": \"os.environ/AZURE_API_KEY\",",
+    "fixed_code": "                    \"api_key\": \"AZURE_API_KEY\",  // Access this in the code",
+    "file_name": "config.json",
+    "start_line": 13,
+    "end_line": 13,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Lack of error handling in database operations.",
+    "confidence": "critical",
+    "reason": "Database operations can fail due to various reasons (e.g., connection issues, SQL errors). Without error handling, the application may crash or behave unexpectedly.",
+    "solution": "Wrap database operations in try-except blocks to handle potential exceptions gracefully.",
+    "actual_code": "41   +1:[+]                 cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))",
+    "fixed_code": "41   +1:[+]                 try:\n41.1   +1:[+]                     cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))\n41.2   +1:[+]                 except Exception as e:\n41.3   +1:[+]                     # Handle the exception (e.g., log it, raise a custom error, etc.)\n41.4   +1:[+]                     raise RuntimeError('Database query failed') from e",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 39,
+    "end_line": 41,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Docker",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to Dockerfile, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_name": "Dockerfile",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Docker",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to docker-compose.yml, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "15",
+    "end_line": "15",
+    "side": "RIGHT",
+    "file_name": "docker-compose.yml",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Version Control",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to .gitignore, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "164",
+    "end_line": "164",
+    "side": "RIGHT",
+    "file_name": ".gitignore",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Database",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to db_setup/init.sql, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "db_setup/init.sql",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_222/issues.json b/.experiments/code_review/gpt-4o-mini/no_eval/pr_222/issues.json
new file mode 100644
index 00000000..061bf872
--- /dev/null
+++ b/.experiments/code_review/gpt-4o-mini/no_eval/pr_222/issues.json
@@ -0,0 +1,342 @@
+[
+  {
+    "topic": "Environment Variables",
+    "comment": "Using 'os.environ' directly in JSON is not valid.",
+    "confidence": "critical",
+    "reason": "Environment variables should be accessed in the code, not hardcoded in configuration files.",
+    "solution": "Use a placeholder or variable in the code to fetch environment variables.",
+    "actual_code": "                    \"api_key\": \"os.environ/AZURE_API_KEY\",",
+    "fixed_code": "                    \"api_key\": \"AZURE_API_KEY\",  // Access this in the code",
+    "file_name": "config.json",
+    "start_line": 13,
+    "end_line": 13,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Dockerfile",
+    "comment": "Consider using multi-stage builds to reduce image size.",
+    "confidence": "important",
+    "reason": "Multi-stage builds can help keep the final image smaller by excluding build dependencies.",
+    "solution": "Use a multi-stage build pattern to install dependencies and copy only necessary files.",
+    "actual_code": "RUN apt-get update && apt-get install -y \\",
+    "fixed_code": "FROM python:3.9 AS builder\nRUN apt-get update && apt-get install -y build-essential git\n\nFROM python:3.9\nCOPY --from=builder /app /app",
+    "file_name": "Dockerfile",
+    "start_line": 8,
+    "end_line": 12,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "Docker Compose",
+    "comment": "Ensure proper indentation for YAML files.",
+    "confidence": "important",
+    "reason": "Improper indentation can lead to YAML parsing errors.",
+    "solution": "Review the indentation levels for all entries in the YAML file.",
+    "actual_code": "     networks:\n       - app-network",
+    "fixed_code": "     networks:\n       - app-network",
+    "file_name": "docker-compose-dev.yml",
+    "start_line": 15,
+    "end_line": 16,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "SQL Initialization",
+    "comment": "Add comments to SQL statements for clarity.",
+    "confidence": "moderate",
+    "reason": "Comments can help future developers understand the purpose of each SQL command.",
+    "solution": "Add comments above each CREATE TABLE statement.",
+    "actual_code": "CREATE TABLE repositories (",
+    "fixed_code": "-- Table to store repository information\nCREATE TABLE repositories (",
+    "file_name": "db_setup/init.sql",
+    "start_line": 4,
+    "end_line": 4,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Shell Script",
+    "comment": "Add error handling to the shell script.",
+    "confidence": "important",
+    "reason": "Error handling can prevent the script from failing silently.",
+    "solution": "Use 'set -e' at the beginning of the script to exit on errors.",
+    "actual_code": "#!/bin/bash",
+    "fixed_code": "#!/bin/bash\nset -e",
+    "file_name": "install_tree_sitter_languages.sh",
+    "start_line": 1,
+    "end_line": 1,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 7
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Lack of error handling in database operations.",
+    "confidence": "critical",
+    "reason": "Database operations can fail due to various reasons (e.g., connection issues, SQL errors). Without error handling, the application may crash or behave unexpectedly.",
+    "solution": "Wrap database operations in try-except blocks to handle potential exceptions gracefully.",
+    "actual_code": "41   +1:[+]                 cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))",
+    "fixed_code": "41   +1:[+]                 try:\n41.1   +1:[+]                     cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))\n41.2   +1:[+]                 except Exception as e:\n41.3   +1:[+]                     # Handle the exception (e.g., log it, raise a custom error, etc.)\n41.4   +1:[+]                     raise RuntimeError('Database query failed') from e",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 39,
+    "end_line": 41,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "Code Readability",
+    "comment": "Consider adding docstrings to methods for better understanding.",
+    "confidence": "important",
+    "reason": "Docstrings help other developers understand the purpose and usage of methods, especially in larger codebases.",
+    "solution": "Add docstrings to the `__init__` and `custom_query` methods to describe their parameters and return values.",
+    "actual_code": "8    +1:[+]     def __init__(self, *args, **kwargs):",
+    "fixed_code": "8    +1:[+]     def __init__(self, *args, **kwargs):\n8.1   +1:[+]         \"\"\"Initialize the CustomPGVectorStore with table name and other parameters.\"\"\"",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 8,
+    "end_line": 8,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Code Readability",
+    "comment": "Consider adding type hints for method parameters and return types.",
+    "confidence": "important",
+    "reason": "Type hints improve code clarity and help with static type checking, making it easier for developers to understand expected types.",
+    "solution": "Add type hints to the `custom_query` method parameters and return type.",
+    "actual_code": "13   +1:[+]     def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:",
+    "fixed_code": "13   +1:[+]     def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[Dict[str, Any]]:",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 13,
+    "end_line": 13,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Code Readability",
+    "comment": "Consider using f-strings for consistency in SQL query construction.",
+    "confidence": "moderate",
+    "reason": "Using f-strings consistently improves readability and reduces the risk of SQL injection if not handled properly.",
+    "solution": "Use f-strings for constructing the SQL query instead of concatenation.",
+    "actual_code": "19   +1:[+]         query = f\"\"\"\n26   +1:[+]{self.table_name}e\n32   +1:[+]             f.repo_id = %s\n36   +1:[+]             %s\n\"\"\"",
+    "fixed_code": "19   +1:[+]         query = f\"\"\"\n19.1   +1:[+]         SELECT \n19.2   +1:[+]             e.node_id,\n19.3   +1:[+]             e.text,\n19.4   +1:[+]             e.metadata,\n19.5   +1:[+]             1 - (e.embedding <=> %s::vector) as similarity\n19.6   +1:[+]         FROM \n19.7   +1:[+]{self.table_name}e\n19.8   +1:[+]         JOIN \n19.9   +1:[+]             function_abstractions fa ON e.node_id = fa.function_id::text\n19.10   +1:[+]         JOIN \n19.11   +1:[+]             files f ON fa.file_id = f.file_id\n19.12   +1:[+]         WHERE \n19.13   +1:[+]             f.repo_id = %s\n19.14   +1:[+]         ORDER BY \n19.15   +1:[+]             similarity DESC\n19.16   +1:[+]         LIMIT \n19.17   +1:[+]             %s\n\"\"\"",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 19,
+    "end_line": 37,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Potential unhandled exceptions in `generate_abstraction` method.",
+    "confidence": "important",
+    "reason": "Raising the exception without handling it can lead to application crashes.",
+    "solution": "Wrap the call to `self.llm_provider.chat_completion` in a try-except block to handle specific exceptions gracefully.",
+    "actual_code": "            raise e",
+    "fixed_code": "            logger.error(f'Error in generating abstraction:{str(e)}')\n            return None, None",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 218,
+    "end_line": 219,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Logging",
+    "comment": "Inconsistent logging levels for errors and debugging.",
+    "confidence": "important",
+    "reason": "Using `logger.debug` for important errors can lead to missed critical information in production logs.",
+    "solution": "Use `logger.error` for logging errors and `logger.debug` for detailed debugging information.",
+    "actual_code": "             logger.debug(f\"Successfully parsed file:{file_path}\")",
+    "fixed_code": "             logger.info(f\"Successfully parsed file:{file_path}\")",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 107,
+    "end_line": 107,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Code Readability",
+    "comment": "Lack of comments explaining complex logic in methods.",
+    "confidence": "moderate",
+    "reason": "While the code is mostly clear, some complex sections could benefit from additional comments for future maintainability.",
+    "solution": "Add comments to explain the purpose and logic of complex code blocks, especially in `store_code_in_db` and `query` methods.",
+    "actual_code": "",
+    "fixed_code": "    # This method stores the code and its abstraction in the database.\n    # Ensure to handle potential conflicts and return the function ID.",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 246,
+    "end_line": 246,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Environment Variables",
+    "comment": "Direct use of environment variables without validation.",
+    "confidence": "important",
+    "reason": "Using environment variables directly can lead to runtime errors if they are not set or misspelled.",
+    "solution": "Implement checks to ensure that required environment variables are set before using them.",
+    "actual_code": "         self.engine = create_engine(\n             f\"postgresql://{os.environ['POSTGRES_USER']}:{os.environ['POSTGRES_PASSWORD']}@{os.environ['POSTGRES_HOST']}:{os.environ['POSTGRES_PORT']}/{os.environ['POSTGRES_DB']}\",",
+    "fixed_code": "         required_env_vars =['POSTGRES_USER', 'POSTGRES_PASSWORD', 'POSTGRES_HOST', 'POSTGRES_PORT', 'POSTGRES_DB']\n         for var in required_env_vars:\n             if var not in os.environ:\n                 raise EnvironmentError(f'Missing required environment variable:{var}')",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 35,
+    "end_line": 39,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Broad exception handling in load_language and get_parser methods.",
+    "confidence": "important",
+    "reason": "Using a generic Exception can obscure the root cause of issues and make debugging difficult.",
+    "solution": "Catch specific exceptions where possible, and log the relevant error messages.",
+    "actual_code": "except Exception as e:",
+    "fixed_code": "except (ImportError, ValueError) as e:",
+    "file_name": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 28,
+    "end_line": 28,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Logging",
+    "comment": "Consider adding more context to log messages.",
+    "confidence": "moderate",
+    "reason": "Current log messages may not provide enough context for troubleshooting.",
+    "solution": "Include the function name or additional context in the log messages.",
+    "actual_code": "logger.error(f\"Failed to load language{language}:{str(e)}\")",
+    "fixed_code": "logger.error(f\"{__name__}.load_language failed for{language}:{str(e)}\")",
+    "file_name": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 29,
+    "end_line": 29,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Code Duplication",
+    "comment": "Duplicated code in traverse_tree for handling different node types.",
+    "confidence": "important",
+    "reason": "Code duplication can lead to maintenance challenges and potential inconsistencies.",
+    "solution": "Consider refactoring to a helper function that handles common logic.",
+    "actual_code": "return{\"type\": \"function\", \"name\": (node.child_by_field_name(\"name\").text.decode(\"utf8\") if node.child_by_field_name(\"name\") else \"anonymous\"), \"code\": code_bytes[node.start_byte : node.end_byte].decode(\"utf8\")}",
+    "fixed_code": "def extract_node_info(node, code_bytes): ...",
+    "file_name": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 55,
+    "end_line": 68,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Function Naming",
+    "comment": "Function names could be more descriptive.",
+    "confidence": "low",
+    "reason": "Descriptive names improve code readability and maintainability.",
+    "solution": "Consider renaming functions to reflect their purpose more clearly.",
+    "actual_code": "def parse_code(code: str, language: str) -> Dict[str, Any]:",
+    "fixed_code": "def parse_source_code(source_code: str, language: str) -> Dict[str, Any]:",
+    "file_name": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 91,
+    "end_line": 91,
+    "side": "LEFT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Dependency Management",
+    "comment": "Ensure that dependencies in pyproject.toml are up-to-date.",
+    "confidence": "important",
+    "reason": "Using outdated dependencies can lead to security vulnerabilities and compatibility issues.",
+    "solution": "Regularly review and update dependencies to the latest stable versions.",
+    "actual_code": "python = \"^3.8.1\"",
+    "fixed_code": "python = \"^3.9.0\"",
+    "file_name": "pyproject.toml",
+    "start_line": 13,
+    "end_line": 13,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Docker",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to Dockerfile, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_name": "Dockerfile",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Docker",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to docker-compose.yml, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "15",
+    "end_line": "15",
+    "side": "RIGHT",
+    "file_name": "docker-compose.yml",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Version Control",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to .gitignore, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "164",
+    "end_line": "164",
+    "side": "RIGHT",
+    "file_name": ".gitignore",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Database",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to db_setup/init.sql, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "db_setup/init.sql",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_222/review.md b/.experiments/code_review/gpt-4o-mini/no_eval/pr_222/review.md
new file mode 100644
index 00000000..a5f856a0
--- /dev/null
+++ b/.experiments/code_review/gpt-4o-mini/no_eval/pr_222/review.md
@@ -0,0 +1,402 @@
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 23
+- Critical: 7
+- Important: 11
+- Minor: 4
+- Files Affected: 11
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Environment Variables (7 issues)</strong></summary>
+
+### 1. Using 'os.environ' directly in JSON is not valid.
+📁 **File:** `config.json:13`
+⚖️ **Severity:** 8/10
+🔍 **Description:** Environment variables should be accessed in the code, not hardcoded in configuration files.
+💡 **Solution:** Use a placeholder or variable in the code to fetch environment variables.
+
+**Current Code:**
+```python
+                    "api_key": "os.environ/AZURE_API_KEY",
+```
+
+**Suggested Code:**
+```python
+                    "api_key": "AZURE_API_KEY",  // Access this in the code
+```
+
+### 2. Lack of error handling in database operations.
+📁 **File:** `kaizen/retriever/custom_vector_store.py:39`
+⚖️ **Severity:** 9/10
+🔍 **Description:** Database operations can fail due to various reasons (e.g., connection issues, SQL errors). Without error handling, the application may crash or behave unexpectedly.
+💡 **Solution:** Wrap database operations in try-except blocks to handle potential exceptions gracefully.
+
+**Current Code:**
+```python
+41   +1:[+]                 cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))
+```
+
+**Suggested Code:**
+```python
+41   +1:[+]                 try:
+41.1   +1:[+]                     cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))
+41.2   +1:[+]                 except Exception as e:
+41.3   +1:[+]                     # Handle the exception (e.g., log it, raise a custom error, etc.)
+41.4   +1:[+]                     raise RuntimeError('Database query failed') from e
+```
+
+### 3. Changes made to sensitive file
+📁 **File:** `config.json:4`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to config.json, which needs review
+💡 **Solution:** NA
+
+### 4. Changes made to sensitive file
+📁 **File:** `Dockerfile:4`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to Dockerfile, which needs review
+💡 **Solution:** NA
+
+### 5. Changes made to sensitive file
+📁 **File:** `docker-compose.yml:15`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to docker-compose.yml, which needs review
+💡 **Solution:** NA
+
+### 6. Changes made to sensitive file
+📁 **File:** `.gitignore:164`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to .gitignore, which needs review
+💡 **Solution:** NA
+
+### 7. Changes made to sensitive file
+📁 **File:** `db_setup/init.sql:1`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to db_setup/init.sql, which needs review
+💡 **Solution:** NA
+
+</details>
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Dockerfile (11 issues)</strong></summary>
+
+### 1. Consider using multi-stage builds to reduce image size.
+📁 **File:** `Dockerfile:8`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Multi-stage builds can help keep the final image smaller by excluding build dependencies.
+💡 **Solution:** Use a multi-stage build pattern to install dependencies and copy only necessary files.
+
+**Current Code:**
+```python
+RUN apt-get update && apt-get install -y \
+```
+
+**Suggested Code:**
+```python
+FROM python:3.9 AS builder
+RUN apt-get update && apt-get install -y build-essential git
+
+FROM python:3.9
+COPY --from=builder /app /app
+```
+
+### 2. Ensure proper indentation for YAML files.
+📁 **File:** `docker-compose-dev.yml:15`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Improper indentation can lead to YAML parsing errors.
+💡 **Solution:** Review the indentation levels for all entries in the YAML file.
+
+**Current Code:**
+```python
+     networks:
+       - app-network
+```
+
+**Suggested Code:**
+```python
+     networks:
+       - app-network
+```
+
+### 3. Add error handling to the shell script.
+📁 **File:** `install_tree_sitter_languages.sh:1`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Error handling can prevent the script from failing silently.
+💡 **Solution:** Use 'set -e' at the beginning of the script to exit on errors.
+
+**Current Code:**
+```python
+#!/bin/bash
+```
+
+**Suggested Code:**
+```python
+#!/bin/bash
+set -e
+```
+
+### 4. Consider adding docstrings to methods for better understanding.
+📁 **File:** `kaizen/retriever/custom_vector_store.py:8`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Docstrings help other developers understand the purpose and usage of methods, especially in larger codebases.
+💡 **Solution:** Add docstrings to the `__init__` and `custom_query` methods to describe their parameters and return values.
+
+**Current Code:**
+```python
+8    +1:[+]     def __init__(self, *args, **kwargs):
+```
+
+**Suggested Code:**
+```python
+8    +1:[+]     def __init__(self, *args, **kwargs):
+8.1   +1:[+]         """Initialize the CustomPGVectorStore with table name and other parameters."""
+```
+
+### 5. Consider adding type hints for method parameters and return types.
+📁 **File:** `kaizen/retriever/custom_vector_store.py:13`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Type hints improve code clarity and help with static type checking, making it easier for developers to understand expected types.
+💡 **Solution:** Add type hints to the `custom_query` method parameters and return type.
+
+**Current Code:**
+```python
+13   +1:[+]     def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:
+```
+
+**Suggested Code:**
+```python
+13   +1:[+]     def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[Dict[str, Any]]:
+```
+
+### 6. Potential unhandled exceptions in `generate_abstraction` method.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:218`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Raising the exception without handling it can lead to application crashes.
+💡 **Solution:** Wrap the call to `self.llm_provider.chat_completion` in a try-except block to handle specific exceptions gracefully.
+
+**Current Code:**
+```python
+            raise e
+```
+
+**Suggested Code:**
+```python
+            logger.error(f'Error in generating abstraction:{str(e)}')
+            return None, None
+```
+
+### 7. Inconsistent logging levels for errors and debugging.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:107`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Using `logger.debug` for important errors can lead to missed critical information in production logs.
+💡 **Solution:** Use `logger.error` for logging errors and `logger.debug` for detailed debugging information.
+
+**Current Code:**
+```python
+             logger.debug(f"Successfully parsed file:{file_path}")
+```
+
+**Suggested Code:**
+```python
+             logger.info(f"Successfully parsed file:{file_path}")
+```
+
+### 8. Direct use of environment variables without validation.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:35`
+⚖️ **Severity:** 8/10
+🔍 **Description:** Using environment variables directly can lead to runtime errors if they are not set or misspelled.
+💡 **Solution:** Implement checks to ensure that required environment variables are set before using them.
+
+**Current Code:**
+```python
+         self.engine = create_engine(
+             f"postgresql://{os.environ['POSTGRES_USER']}:{os.environ['POSTGRES_PASSWORD']}@{os.environ['POSTGRES_HOST']}:{os.environ['POSTGRES_PORT']}/{os.environ['POSTGRES_DB']}",
+```
+
+**Suggested Code:**
+```python
+         required_env_vars =['POSTGRES_USER', 'POSTGRES_PASSWORD', 'POSTGRES_HOST', 'POSTGRES_PORT', 'POSTGRES_DB']
+         for var in required_env_vars:
+             if var not in os.environ:
+                 raise EnvironmentError(f'Missing required environment variable:{var}')
+```
+
+### 9. Broad exception handling in load_language and get_parser methods.
+📁 **File:** `kaizen/retriever/tree_sitter_utils.py:28`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Using a generic Exception can obscure the root cause of issues and make debugging difficult.
+💡 **Solution:** Catch specific exceptions where possible, and log the relevant error messages.
+
+**Current Code:**
+```python
+except Exception as e:
+```
+
+**Suggested Code:**
+```python
+except (ImportError, ValueError) as e:
+```
+
+### 10. Duplicated code in traverse_tree for handling different node types.
+📁 **File:** `kaizen/retriever/tree_sitter_utils.py:55`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Code duplication can lead to maintenance challenges and potential inconsistencies.
+💡 **Solution:** Consider refactoring to a helper function that handles common logic.
+
+**Current Code:**
+```python
+return{"type": "function", "name": (node.child_by_field_name("name").text.decode("utf8") if node.child_by_field_name("name") else "anonymous"), "code": code_bytes[node.start_byte : node.end_byte].decode("utf8")}
+```
+
+**Suggested Code:**
+```python
+def extract_node_info(node, code_bytes): ...
+```
+
+### 11. Ensure that dependencies in pyproject.toml are up-to-date.
+📁 **File:** `pyproject.toml:13`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Using outdated dependencies can lead to security vulnerabilities and compatibility issues.
+💡 **Solution:** Regularly review and update dependencies to the latest stable versions.
+
+**Current Code:**
+```python
+python = "^3.8.1"
+```
+
+**Suggested Code:**
+```python
+python = "^3.9.0"
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (5 issues)</strong></summary>
+
+<details>
+<summary><strong>SQL Initialization (4 issues)</strong></summary>
+
+### 1. Add comments to SQL statements for clarity.
+📁 **File:** `db_setup/init.sql:4`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Comments can help future developers understand the purpose of each SQL command.
+💡 **Solution:** Add comments above each CREATE TABLE statement.
+
+**Current Code:**
+```python
+CREATE TABLE repositories (
+```
+
+**Suggested Code:**
+```python
+-- Table to store repository information
+CREATE TABLE repositories (
+```
+
+### 2. Consider using f-strings for consistency in SQL query construction.
+📁 **File:** `kaizen/retriever/custom_vector_store.py:19`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Using f-strings consistently improves readability and reduces the risk of SQL injection if not handled properly.
+💡 **Solution:** Use f-strings for constructing the SQL query instead of concatenation.
+
+**Current Code:**
+```python
+19   +1:[+]         query = f"""
+26   +1:[+]{self.table_name}e
+32   +1:[+]             f.repo_id = %s
+36   +1:[+]             %s
+"""
+```
+
+**Suggested Code:**
+```python
+19   +1:[+]         query = f"""
+19.1   +1:[+]         SELECT 
+19.2   +1:[+]             e.node_id,
+19.3   +1:[+]             e.text,
+19.4   +1:[+]             e.metadata,
+19.5   +1:[+]             1 - (e.embedding <=> %s::vector) as similarity
+19.6   +1:[+]         FROM 
+19.7   +1:[+]{self.table_name}e
+19.8   +1:[+]         JOIN 
+19.9   +1:[+]             function_abstractions fa ON e.node_id = fa.function_id::text
+19.10   +1:[+]         JOIN 
+19.11   +1:[+]             files f ON fa.file_id = f.file_id
+19.12   +1:[+]         WHERE 
+19.13   +1:[+]             f.repo_id = %s
+19.14   +1:[+]         ORDER BY 
+19.15   +1:[+]             similarity DESC
+19.16   +1:[+]         LIMIT 
+19.17   +1:[+]             %s
+"""
+```
+
+### 3. Lack of comments explaining complex logic in methods.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:246`
+⚖️ **Severity:** 4/10
+🔍 **Description:** While the code is mostly clear, some complex sections could benefit from additional comments for future maintainability.
+💡 **Solution:** Add comments to explain the purpose and logic of complex code blocks, especially in `store_code_in_db` and `query` methods.
+
+**Current Code:**
+```python
+
+```
+
+**Suggested Code:**
+```python
+    # This method stores the code and its abstraction in the database.
+    # Ensure to handle potential conflicts and return the function ID.
+```
+
+### 4. Consider adding more context to log messages.
+📁 **File:** `kaizen/retriever/tree_sitter_utils.py:29`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Current log messages may not provide enough context for troubleshooting.
+💡 **Solution:** Include the function name or additional context in the log messages.
+
+**Current Code:**
+```python
+logger.error(f"Failed to load language{language}:{str(e)}")
+```
+
+**Suggested Code:**
+```python
+logger.error(f"{__name__}.load_language failed for{language}:{str(e)}")
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage 
+{"prompt_tokens": 21545, "completion_tokens": 3763, "total_tokens": 25308}
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_335/comments.json b/.experiments/code_review/gpt-4o-mini/no_eval/pr_335/comments.json
new file mode 100644
index 00000000..dfddeae0
--- /dev/null
+++ b/.experiments/code_review/gpt-4o-mini/no_eval/pr_335/comments.json
@@ -0,0 +1,17 @@
+[
+  {
+    "topic": "Parameter Removal",
+    "comment": "Removal of 'reeval_response' parameter may affect functionality.",
+    "confidence": "critical",
+    "reason": "This parameter is used in multiple methods, and its removal could lead to unexpected behavior.",
+    "solution": "Evaluate the necessity of this parameter and ensure that its removal does not break existing functionality.",
+    "actual_code": "reeval_response: bool = False,",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 43,
+    "end_line": 43,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 9
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_335/issues.json b/.experiments/code_review/gpt-4o-mini/no_eval/pr_335/issues.json
new file mode 100644
index 00000000..b93239cf
--- /dev/null
+++ b/.experiments/code_review/gpt-4o-mini/no_eval/pr_335/issues.json
@@ -0,0 +1,77 @@
+[
+  {
+    "topic": "Imports",
+    "comment": "Inconsistent import statements.",
+    "confidence": "important",
+    "reason": "Changing import paths can lead to confusion and potential import errors.",
+    "solution": "Ensure that all import paths are updated consistently throughout the codebase.",
+    "actual_code": "from kaizen.llms.prompts.code_review_prompts import (",
+    "fixed_code": "from kaizen.llms.prompts.pr_desc_prompts import (",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 8,
+    "end_line": 8,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "Parameter Removal",
+    "comment": "Removal of 'reeval_response' parameter may affect functionality.",
+    "confidence": "critical",
+    "reason": "This parameter is used in multiple methods, and its removal could lead to unexpected behavior.",
+    "solution": "Evaluate the necessity of this parameter and ensure that its removal does not break existing functionality.",
+    "actual_code": "reeval_response: bool = False,",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 43,
+    "end_line": 43,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Lack of error handling for potential exceptions.",
+    "confidence": "important",
+    "reason": "The absence of error handling can cause the application to crash unexpectedly.",
+    "solution": "Implement try-except blocks where appropriate to handle potential exceptions gracefully.",
+    "actual_code": "raise Exception(\"Both diff_text and pull_request_files are empty!\")",
+    "fixed_code": "raise ValueError(\"Both diff_text and pull_request_files are empty!\")",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 51,
+    "end_line": 51,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Functionality Changes",
+    "comment": "Changes to function calls may alter expected behavior.",
+    "confidence": "important",
+    "reason": "Switching from 'chat_completion_with_json' to 'chat_completion' may change the output format.",
+    "solution": "Review the expected output of the new function and ensure compatibility with existing code.",
+    "actual_code": "resp, usage = self.provider.chat_completion_with_json(prompt, user=user)",
+    "fixed_code": "resp, usage = self.provider.chat_completion(prompt, user=user)",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 83,
+    "end_line": 83,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 8
+  },
+  {
+    "topic": "Documentation",
+    "comment": "Missing documentation for new prompts.",
+    "confidence": "moderate",
+    "reason": "Lack of documentation can make it harder for other developers to understand the purpose of new prompts.",
+    "solution": "Add docstrings or comments explaining the purpose of each new prompt.",
+    "actual_code": "",
+    "fixed_code": "# This prompt is used to generate a PR description based on the provided details.",
+    "file_name": "kaizen/llms/prompts/pr_desc_prompts.py",
+    "start_line": 1,
+    "end_line": 1,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_335/review.md b/.experiments/code_review/gpt-4o-mini/no_eval/pr_335/review.md
new file mode 100644
index 00000000..7fdf7ed4
--- /dev/null
+++ b/.experiments/code_review/gpt-4o-mini/no_eval/pr_335/review.md
@@ -0,0 +1,139 @@
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 5
+- Critical: 1
+- Important: 3
+- Minor: 1
+- Files Affected: 2
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Parameter Removal (1 issues)</strong></summary>
+
+### 1. Removal of 'reeval_response' parameter may affect functionality.
+📁 **File:** `kaizen/generator/pr_description.py:43`
+⚖️ **Severity:** 9/10
+🔍 **Description:** This parameter is used in multiple methods, and its removal could lead to unexpected behavior.
+💡 **Solution:** Evaluate the necessity of this parameter and ensure that its removal does not break existing functionality.
+
+**Current Code:**
+```python
+reeval_response: bool = False,
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Imports (3 issues)</strong></summary>
+
+### 1. Inconsistent import statements.
+📁 **File:** `kaizen/generator/pr_description.py:8`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Changing import paths can lead to confusion and potential import errors.
+💡 **Solution:** Ensure that all import paths are updated consistently throughout the codebase.
+
+**Current Code:**
+```python
+from kaizen.llms.prompts.code_review_prompts import (
+```
+
+**Suggested Code:**
+```python
+from kaizen.llms.prompts.pr_desc_prompts import (
+```
+
+### 2. Lack of error handling for potential exceptions.
+📁 **File:** `kaizen/generator/pr_description.py:51`
+⚖️ **Severity:** 7/10
+🔍 **Description:** The absence of error handling can cause the application to crash unexpectedly.
+💡 **Solution:** Implement try-except blocks where appropriate to handle potential exceptions gracefully.
+
+**Current Code:**
+```python
+raise Exception("Both diff_text and pull_request_files are empty!")
+```
+
+**Suggested Code:**
+```python
+raise ValueError("Both diff_text and pull_request_files are empty!")
+```
+
+### 3. Changes to function calls may alter expected behavior.
+📁 **File:** `kaizen/generator/pr_description.py:83`
+⚖️ **Severity:** 8/10
+🔍 **Description:** Switching from 'chat_completion_with_json' to 'chat_completion' may change the output format.
+💡 **Solution:** Review the expected output of the new function and ensure compatibility with existing code.
+
+**Current Code:**
+```python
+resp, usage = self.provider.chat_completion_with_json(prompt, user=user)
+```
+
+**Suggested Code:**
+```python
+resp, usage = self.provider.chat_completion(prompt, user=user)
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (1 issues)</strong></summary>
+
+<details>
+<summary><strong>Documentation (1 issues)</strong></summary>
+
+### 1. Missing documentation for new prompts.
+📁 **File:** `kaizen/llms/prompts/pr_desc_prompts.py:1`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Lack of documentation can make it harder for other developers to understand the purpose of new prompts.
+💡 **Solution:** Add docstrings or comments explaining the purpose of each new prompt.
+
+**Current Code:**
+```python
+
+```
+
+**Suggested Code:**
+```python
+# This prompt is used to generate a PR description based on the provided details.
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage 
+{"prompt_tokens": 6751, "completion_tokens": 826, "total_tokens": 7577}
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_440/comments.json b/.experiments/code_review/gpt-4o-mini/no_eval/pr_440/comments.json
new file mode 100644
index 00000000..0637a088
--- /dev/null
+++ b/.experiments/code_review/gpt-4o-mini/no_eval/pr_440/comments.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_440/issues.json b/.experiments/code_review/gpt-4o-mini/no_eval/pr_440/issues.json
new file mode 100644
index 00000000..7e262d38
--- /dev/null
+++ b/.experiments/code_review/gpt-4o-mini/no_eval/pr_440/issues.json
@@ -0,0 +1,47 @@
+[
+  {
+    "topic": "Repository Setup",
+    "comment": "The setup_repository method is commented out in the previous version, which may lead to confusion about its necessity.",
+    "confidence": "important",
+    "reason": "Commenting out essential setup code can lead to runtime errors if the repository is not properly initialized.",
+    "solution": "Ensure that the setup_repository method is called appropriately to avoid potential issues.",
+    "actual_code": "analyzer.setup_repository(\"./github_app/\")",
+    "fixed_code": "analyzer.setup_repository(\"./github_app/\")",
+    "file_name": "examples/ragify_codebase/main.py",
+    "start_line": 7,
+    "end_line": 7,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Code Comment",
+    "comment": "The TODO comment lacks specificity regarding how to handle duplicates.",
+    "confidence": "moderate",
+    "reason": "Vague comments can lead to misunderstandings and may not provide enough guidance for future developers.",
+    "solution": "Specify the conditions under which duplicates should be checked and how to handle them.",
+    "actual_code": "# TODO: DONT PUSH DUPLICATE",
+    "fixed_code": "# TODO: Implement a check to prevent pushing duplicate embeddings based on a unique identifier.",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 157,
+    "end_line": 157,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Dependency Update",
+    "comment": "The dependency version for llama-index-core has been updated without a clear reason.",
+    "confidence": "important",
+    "reason": "Updating dependencies can introduce breaking changes; it's important to ensure compatibility.",
+    "solution": "Review the changelog for llama-index-core to confirm that the new version does not introduce breaking changes.",
+    "actual_code": "llama-index-core = \"^0.10.47\"",
+    "fixed_code": "llama-index-core = \"0.10.65\"",
+    "file_name": "pyproject.toml",
+    "start_line": 27,
+    "end_line": 27,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_440/review.md b/.experiments/code_review/gpt-4o-mini/no_eval/pr_440/review.md
new file mode 100644
index 00000000..08c9e2bc
--- /dev/null
+++ b/.experiments/code_review/gpt-4o-mini/no_eval/pr_440/review.md
@@ -0,0 +1,100 @@
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 3
+- Critical: 0
+- Important: 2
+- Minor: 1
+- Files Affected: 3
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Repository Setup (2 issues)</strong></summary>
+
+### 1. The setup_repository method is commented out in the previous version, which may lead to confusion about its necessity.
+📁 **File:** `examples/ragify_codebase/main.py:7`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Commenting out essential setup code can lead to runtime errors if the repository is not properly initialized.
+💡 **Solution:** Ensure that the setup_repository method is called appropriately to avoid potential issues.
+
+**Current Code:**
+```python
+analyzer.setup_repository("./github_app/")
+```
+
+**Suggested Code:**
+```python
+analyzer.setup_repository("./github_app/")
+```
+
+### 2. The dependency version for llama-index-core has been updated without a clear reason.
+📁 **File:** `pyproject.toml:27`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Updating dependencies can introduce breaking changes; it's important to ensure compatibility.
+💡 **Solution:** Review the changelog for llama-index-core to confirm that the new version does not introduce breaking changes.
+
+**Current Code:**
+```python
+llama-index-core = "^0.10.47"
+```
+
+**Suggested Code:**
+```python
+llama-index-core = "0.10.65"
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (1 issues)</strong></summary>
+
+<details>
+<summary><strong>Code Comment (1 issues)</strong></summary>
+
+### 1. The TODO comment lacks specificity regarding how to handle duplicates.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:157`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Vague comments can lead to misunderstandings and may not provide enough guidance for future developers.
+💡 **Solution:** Specify the conditions under which duplicates should be checked and how to handle them.
+
+**Current Code:**
+```python
+# TODO: DONT PUSH DUPLICATE
+```
+
+**Suggested Code:**
+```python
+# TODO: Implement a check to prevent pushing duplicate embeddings based on a unique identifier.
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage 
+{"prompt_tokens": 1364, "completion_tokens": 544, "total_tokens": 1908}
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_476/comments.json b/.experiments/code_review/gpt-4o-mini/no_eval/pr_476/comments.json
new file mode 100644
index 00000000..a9d40eac
--- /dev/null
+++ b/.experiments/code_review/gpt-4o-mini/no_eval/pr_476/comments.json
@@ -0,0 +1,16 @@
+[
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_476/issues.json b/.experiments/code_review/gpt-4o-mini/no_eval/pr_476/issues.json
new file mode 100644
index 00000000..498888fb
--- /dev/null
+++ b/.experiments/code_review/gpt-4o-mini/no_eval/pr_476/issues.json
@@ -0,0 +1,91 @@
+[
+  {
+    "topic": "Error Handling",
+    "comment": "Broad exception handling can obscure specific errors.",
+    "confidence": "important",
+    "reason": "Using a generic Exception can make debugging difficult and hide underlying issues.",
+    "solution": "Catch specific exceptions where possible.",
+    "actual_code": "except Exception:",
+    "fixed_code": "except KeyError:",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 140,
+    "end_line": 141,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Function Signature",
+    "comment": "The function 'post_pull_request' has an additional parameter that should be documented.",
+    "confidence": "important",
+    "reason": "New parameters should be documented to ensure clarity for future maintainers.",
+    "solution": "Update the function docstring to include the 'tests' parameter.",
+    "actual_code": "def post_pull_request(url, data, installation_id, tests=None):",
+    "fixed_code": "def post_pull_request(url, data, installation_id, tests=None):  # tests: List of test files",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 106,
+    "end_line": 106,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Code Readability",
+    "comment": "The new function 'sort_files' lacks a docstring.",
+    "confidence": "important",
+    "reason": "Docstrings are essential for understanding the purpose and usage of functions.",
+    "solution": "Add a docstring to describe the function's purpose and parameters.",
+    "actual_code": "def sort_files(files):",
+    "fixed_code": "def sort_files(files):  # Sorts a list of file dictionaries by filename.",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 184,
+    "end_line": 184,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Variable Naming",
+    "comment": "The variable 'tests' could be more descriptive.",
+    "confidence": "moderate",
+    "reason": "Descriptive variable names improve code readability and maintainability.",
+    "solution": "Consider renaming 'tests' to 'generated_tests' for clarity.",
+    "actual_code": "tests = generate_tests(pr_files)",
+    "fixed_code": "generated_tests = generate_tests(pr_files)",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 58,
+    "end_line": 58,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 3
+  },
+  {
+    "topic": "Logging",
+    "comment": "Consider using logging instead of print statements for error reporting.",
+    "confidence": "important",
+    "reason": "Using logging allows for better control over the output and can be configured for different environments.",
+    "solution": "Replace print statements with appropriate logging calls.",
+    "actual_code": "print(\"Error\")",
+    "fixed_code": "logger.error(\"Error occurred\")",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 141,
+    "end_line": 141,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_476/review.md b/.experiments/code_review/gpt-4o-mini/no_eval/pr_476/review.md
new file mode 100644
index 00000000..5d27ff4f
--- /dev/null
+++ b/.experiments/code_review/gpt-4o-mini/no_eval/pr_476/review.md
@@ -0,0 +1,145 @@
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 6
+- Critical: 1
+- Important: 4
+- Minor: 1
+- Files Affected: 2
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Configuration (1 issues)</strong></summary>
+
+### 1. Changes made to sensitive file
+📁 **File:** `config.json:1`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to config.json, which needs review
+💡 **Solution:** NA
+
+</details>
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Error Handling (4 issues)</strong></summary>
+
+### 1. Broad exception handling can obscure specific errors.
+📁 **File:** `github_app/github_helper/pull_requests.py:140`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Using a generic Exception can make debugging difficult and hide underlying issues.
+💡 **Solution:** Catch specific exceptions where possible.
+
+**Current Code:**
+```python
+except Exception:
+```
+
+**Suggested Code:**
+```python
+except KeyError:
+```
+
+### 2. The function 'post_pull_request' has an additional parameter that should be documented.
+📁 **File:** `github_app/github_helper/pull_requests.py:106`
+⚖️ **Severity:** 5/10
+🔍 **Description:** New parameters should be documented to ensure clarity for future maintainers.
+💡 **Solution:** Update the function docstring to include the 'tests' parameter.
+
+**Current Code:**
+```python
+def post_pull_request(url, data, installation_id, tests=None):
+```
+
+**Suggested Code:**
+```python
+def post_pull_request(url, data, installation_id, tests=None):  # tests: List of test files
+```
+
+### 3. The new function 'sort_files' lacks a docstring.
+📁 **File:** `github_app/github_helper/pull_requests.py:184`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Docstrings are essential for understanding the purpose and usage of functions.
+💡 **Solution:** Add a docstring to describe the function's purpose and parameters.
+
+**Current Code:**
+```python
+def sort_files(files):
+```
+
+**Suggested Code:**
+```python
+def sort_files(files):  # Sorts a list of file dictionaries by filename.
+```
+
+### 4. Consider using logging instead of print statements for error reporting.
+📁 **File:** `github_app/github_helper/pull_requests.py:141`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Using logging allows for better control over the output and can be configured for different environments.
+💡 **Solution:** Replace print statements with appropriate logging calls.
+
+**Current Code:**
+```python
+print("Error")
+```
+
+**Suggested Code:**
+```python
+logger.error("Error occurred")
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (1 issues)</strong></summary>
+
+<details>
+<summary><strong>Variable Naming (1 issues)</strong></summary>
+
+### 1. The variable 'tests' could be more descriptive.
+📁 **File:** `github_app/github_helper/pull_requests.py:58`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Descriptive variable names improve code readability and maintainability.
+💡 **Solution:** Consider renaming 'tests' to 'generated_tests' for clarity.
+
+**Current Code:**
+```python
+tests = generate_tests(pr_files)
+```
+
+**Suggested Code:**
+```python
+generated_tests = generate_tests(pr_files)
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage 
+{"prompt_tokens": 4007, "completion_tokens": 796, "total_tokens": 4803}
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_5/comments.json b/.experiments/code_review/gpt-4o-mini/no_eval/pr_5/comments.json
new file mode 100644
index 00000000..c38f9fea
--- /dev/null
+++ b/.experiments/code_review/gpt-4o-mini/no_eval/pr_5/comments.json
@@ -0,0 +1,47 @@
+[
+  {
+    "topic": "Error Handling",
+    "comment": "API call may fail without a retry mechanism.",
+    "confidence": "critical",
+    "reason": "The completion function call does not handle potential API failures.",
+    "solution": "Implement a retry mechanism or error handling for API calls.",
+    "actual_code": "response = completion(",
+    "fixed_code": "# Implement retry logic here",
+    "file_name": "main.py",
+    "start_line": 66,
+    "end_line": 68,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Silent Failure",
+    "comment": "Silent failure without logging in case of JSON parsing error.",
+    "confidence": "critical",
+    "reason": "The code does not log the error, making debugging difficult.",
+    "solution": "Log the error message to provide feedback in case of failure.",
+    "actual_code": "print(f\"Failed to parse content for applicant\")",
+    "fixed_code": "print(f\"Failed to parse content for applicant:{e}\")",
+    "file_name": "main.py",
+    "start_line": 86,
+    "end_line": 86,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Division by Zero",
+    "comment": "Potential division by zero error.",
+    "confidence": "critical",
+    "reason": "If total_tokens is zero, this will raise an error.",
+    "solution": "Add a check to prevent division by zero.",
+    "actual_code": "print(f\"Total tokens used:{total_tokens:,}\")",
+    "fixed_code": "if total_tokens > 0: print(f\"Total tokens used:{total_tokens:,}\")",
+    "file_name": "main.py",
+    "start_line": 159,
+    "end_line": 159,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 9
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_5/issues.json b/.experiments/code_review/gpt-4o-mini/no_eval/pr_5/issues.json
new file mode 100644
index 00000000..5dd7c900
--- /dev/null
+++ b/.experiments/code_review/gpt-4o-mini/no_eval/pr_5/issues.json
@@ -0,0 +1,107 @@
+[
+  {
+    "topic": "Imports",
+    "comment": "Unused import detected.",
+    "confidence": "trivial",
+    "reason": "The import of 'random' is not utilized in the code.",
+    "solution": "Remove the unused import to clean up the code.",
+    "actual_code": "import random  # Unused import",
+    "fixed_code": "",
+    "file_name": "main.py",
+    "start_line": 8,
+    "end_line": 8,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 2
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "API call may fail without a retry mechanism.",
+    "confidence": "critical",
+    "reason": "The completion function call does not handle potential API failures.",
+    "solution": "Implement a retry mechanism or error handling for API calls.",
+    "actual_code": "response = completion(",
+    "fixed_code": "# Implement retry logic here",
+    "file_name": "main.py",
+    "start_line": 66,
+    "end_line": 68,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Silent Failure",
+    "comment": "Silent failure without logging in case of JSON parsing error.",
+    "confidence": "critical",
+    "reason": "The code does not log the error, making debugging difficult.",
+    "solution": "Log the error message to provide feedback in case of failure.",
+    "actual_code": "print(f\"Failed to parse content for applicant\")",
+    "fixed_code": "print(f\"Failed to parse content for applicant:{e}\")",
+    "file_name": "main.py",
+    "start_line": 86,
+    "end_line": 86,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Inefficient Progress Reporting",
+    "comment": "Inefficient way to print progress.",
+    "confidence": "important",
+    "reason": "The current method of printing progress can be improved for better performance.",
+    "solution": "Consider using a logging library or a more efficient progress reporting method.",
+    "actual_code": "print(f\"\\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}\", end=\"\", flush=True)",
+    "fixed_code": "# Use logging or a more efficient progress reporting",
+    "file_name": "main.py",
+    "start_line": 121,
+    "end_line": 121,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Redundant Code",
+    "comment": "Redundant check for empty DataFrame.",
+    "confidence": "moderate",
+    "reason": "The check for an empty DataFrame is unnecessary as the process will handle it.",
+    "solution": "Remove the redundant check to simplify the code.",
+    "actual_code": "if len(df) == 0:",
+    "fixed_code": "",
+    "file_name": "main.py",
+    "start_line": 142,
+    "end_line": 143,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Division by Zero",
+    "comment": "Potential division by zero error.",
+    "confidence": "critical",
+    "reason": "If total_tokens is zero, this will raise an error.",
+    "solution": "Add a check to prevent division by zero.",
+    "actual_code": "print(f\"Total tokens used:{total_tokens:,}\")",
+    "fixed_code": "if total_tokens > 0: print(f\"Total tokens used:{total_tokens:,}\")",
+    "file_name": "main.py",
+    "start_line": 159,
+    "end_line": 159,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "File Handling",
+    "comment": "No error handling for file not found.",
+    "confidence": "important",
+    "reason": "If the file does not exist, it will raise an unhandled exception.",
+    "solution": "Add error handling to manage file not found scenarios.",
+    "actual_code": "main(input_file)",
+    "fixed_code": "# Add error handling for file not found",
+    "file_name": "main.py",
+    "start_line": 175,
+    "end_line": 175,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 6
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_5/review.md b/.experiments/code_review/gpt-4o-mini/no_eval/pr_5/review.md
new file mode 100644
index 00000000..fcc46911
--- /dev/null
+++ b/.experiments/code_review/gpt-4o-mini/no_eval/pr_5/review.md
@@ -0,0 +1,155 @@
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 7
+- Critical: 3
+- Important: 2
+- Minor: 1
+- Files Affected: 1
+## 🏆 Code Quality
+[███████████████░░░░░] 75% (Fair)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Error Handling (3 issues)</strong></summary>
+
+### 1. API call may fail without a retry mechanism.
+📁 **File:** `main.py:66`
+⚖️ **Severity:** 8/10
+🔍 **Description:** The completion function call does not handle potential API failures.
+💡 **Solution:** Implement a retry mechanism or error handling for API calls.
+
+**Current Code:**
+```python
+response = completion(
+```
+
+**Suggested Code:**
+```python
+# Implement retry logic here
+```
+
+### 2. Silent failure without logging in case of JSON parsing error.
+📁 **File:** `main.py:86`
+⚖️ **Severity:** 7/10
+🔍 **Description:** The code does not log the error, making debugging difficult.
+💡 **Solution:** Log the error message to provide feedback in case of failure.
+
+**Current Code:**
+```python
+print(f"Failed to parse content for applicant")
+```
+
+**Suggested Code:**
+```python
+print(f"Failed to parse content for applicant:{e}")
+```
+
+### 3. Potential division by zero error.
+📁 **File:** `main.py:159`
+⚖️ **Severity:** 9/10
+🔍 **Description:** If total_tokens is zero, this will raise an error.
+💡 **Solution:** Add a check to prevent division by zero.
+
+**Current Code:**
+```python
+print(f"Total tokens used:{total_tokens:,}")
+```
+
+**Suggested Code:**
+```python
+if total_tokens > 0: print(f"Total tokens used:{total_tokens:,}")
+```
+
+</details>
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Inefficient Progress Reporting (2 issues)</strong></summary>
+
+### 1. Inefficient way to print progress.
+📁 **File:** `main.py:121`
+⚖️ **Severity:** 5/10
+🔍 **Description:** The current method of printing progress can be improved for better performance.
+💡 **Solution:** Consider using a logging library or a more efficient progress reporting method.
+
+**Current Code:**
+```python
+print(f"\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}", end="", flush=True)
+```
+
+**Suggested Code:**
+```python
+# Use logging or a more efficient progress reporting
+```
+
+### 2. No error handling for file not found.
+📁 **File:** `main.py:175`
+⚖️ **Severity:** 6/10
+🔍 **Description:** If the file does not exist, it will raise an unhandled exception.
+💡 **Solution:** Add error handling to manage file not found scenarios.
+
+**Current Code:**
+```python
+main(input_file)
+```
+
+**Suggested Code:**
+```python
+# Add error handling for file not found
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (2 issues)</strong></summary>
+
+<details>
+<summary><strong>Redundant Code (1 issues)</strong></summary>
+
+### 1. Redundant check for empty DataFrame.
+📁 **File:** `main.py:142`
+⚖️ **Severity:** 4/10
+🔍 **Description:** The check for an empty DataFrame is unnecessary as the process will handle it.
+💡 **Solution:** Remove the redundant check to simplify the code.
+
+**Current Code:**
+```python
+if len(df) == 0:
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage 
+{"prompt_tokens": 6154, "completion_tokens": 1046, "total_tokens": 7200}
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_222/comments.json b/.experiments/code_review/gpt-4o/no_eval/pr_222/comments.json
new file mode 100644
index 00000000..5ebff860
--- /dev/null
+++ b/.experiments/code_review/gpt-4o/no_eval/pr_222/comments.json
@@ -0,0 +1,132 @@
+[
+  {
+    "topic": "Security",
+    "comment": "Hardcoding API keys in `config.json` can lead to security vulnerabilities.",
+    "confidence": "critical",
+    "reason": "Exposing API keys in the codebase can lead to unauthorized access.",
+    "solution": "Use environment variables to store API keys instead of hardcoding them.",
+    "actual_code": "\"api_key\": \"os.environ/AZURE_API_KEY\"",
+    "fixed_code": "\"api_key\": \"${AZURE_API_KEY}\"",
+    "file_name": "config.json",
+    "start_line": 13,
+    "end_line": 13,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "Security",
+    "comment": "Database connection strings should not be constructed using string interpolation.",
+    "confidence": "critical",
+    "reason": "Using string interpolation for connection strings can expose the application to SQL injection attacks.",
+    "solution": "Use parameterized queries or a configuration management tool to handle sensitive information.",
+    "actual_code": "self.engine = create_engine(\n    f\"postgresql://{os.environ['POSTGRES_USER']}:{os.environ['POSTGRES_PASSWORD']}@{os.environ['POSTGRES_HOST']}:{os.environ['POSTGRES_PORT']}/{os.environ['POSTGRES_DB']}\",\n    pool_size=10,\n    max_overflow=20,\n)",
+    "fixed_code": "self.engine = create_engine(\n    'postgresql://{user}:{password}@{host}:{port}/{db}'.format(\n        user=os.environ['POSTGRES_USER'],\n        password=os.environ['POSTGRES_PASSWORD'],\n        host=os.environ['POSTGRES_HOST'],\n        port=os.environ['POSTGRES_PORT'],\n        db=os.environ['POSTGRES_DB']\n    ),\n    pool_size=10,\n    max_overflow=20,\n)",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 35,
+    "end_line": 39,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "SQL Injection",
+    "comment": "Potential SQL injection vulnerability in the query construction.",
+    "confidence": "critical",
+    "reason": "Using f-strings to construct SQL queries can lead to SQL injection attacks if user input is not properly sanitized.",
+    "solution": "Use parameterized queries to avoid SQL injection vulnerabilities.",
+    "actual_code": "query = f\"\"\"\nSELECT \n    e.node_id,\n    e.text,\n    e.metadata,\n    1 - (e.embedding <=> %s::vector) as similarity\nFROM \n{self.table_name}e\nJOIN \n    function_abstractions fa ON e.node_id = fa.function_id::text\nJOIN \n    files f ON fa.file_id = f.file_id\nWHERE \n    f.repo_id = %s\nORDER BY \n    similarity DESC\nLIMIT \n    %s\n\"\"\"",
+    "fixed_code": "query = \"\"\"\nSELECT \n    e.node_id,\n    e.text,\n    e.metadata,\n    1 - (e.embedding <=> %s::vector) as similarity\nFROM \n    %s e\nJOIN \n    function_abstractions fa ON e.node_id = fa.function_id::text\nJOIN \n    files f ON fa.file_id = f.file_id\nWHERE \n    f.repo_id = %s\nORDER BY \n    similarity DESC\nLIMIT \n    %s\n\"\"\"",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 19,
+    "end_line": 37,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "Exception Handling",
+    "comment": "Broad exception handling should be avoided.",
+    "confidence": "critical",
+    "reason": "Catching all exceptions can hide bugs and make debugging difficult.",
+    "solution": "Catch specific exceptions instead of using a broad except clause.",
+    "actual_code": "except Exception as e:",
+    "fixed_code": "except ImportError as e:\n    logger.error(f\"Failed to import module:{str(e)}\")\n    raise\nexcept ValueError as e:\n    logger.error(f\"Invalid value:{str(e)}\")\n    raise",
+    "file_name": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 28,
+    "end_line": 30,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Docker",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to Dockerfile, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_name": "Dockerfile",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Docker",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to docker-compose.yml, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "15",
+    "end_line": "15",
+    "side": "RIGHT",
+    "file_name": "docker-compose.yml",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Version Control",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to .gitignore, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "164",
+    "end_line": "164",
+    "side": "RIGHT",
+    "file_name": ".gitignore",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Database",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to db_setup/init.sql, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "db_setup/init.sql",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_222/issues.json b/.experiments/code_review/gpt-4o/no_eval/pr_222/issues.json
new file mode 100644
index 00000000..a46082c3
--- /dev/null
+++ b/.experiments/code_review/gpt-4o/no_eval/pr_222/issues.json
@@ -0,0 +1,387 @@
+[
+  {
+    "topic": "Documentation",
+    "comment": "The `install_tree_sitter_languages.sh` script lacks comments explaining the purpose of each step.",
+    "confidence": "moderate",
+    "reason": "Comments improve readability and maintainability, especially for complex scripts.",
+    "solution": "Add comments explaining the purpose of each step in the script.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "install_tree_sitter_languages.sh",
+    "start_line": 1,
+    "end_line": 47,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "The `install_tree_sitter_languages.sh` script does not handle errors during the installation process.",
+    "confidence": "important",
+    "reason": "Error handling ensures that the script fails gracefully and provides useful error messages.",
+    "solution": "Add error handling for each critical step in the script.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "install_tree_sitter_languages.sh",
+    "start_line": 1,
+    "end_line": 47,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Security",
+    "comment": "Hardcoding API keys in `config.json` can lead to security vulnerabilities.",
+    "confidence": "critical",
+    "reason": "Exposing API keys in the codebase can lead to unauthorized access.",
+    "solution": "Use environment variables to store API keys instead of hardcoding them.",
+    "actual_code": "\"api_key\": \"os.environ/AZURE_API_KEY\"",
+    "fixed_code": "\"api_key\": \"${AZURE_API_KEY}\"",
+    "file_name": "config.json",
+    "start_line": 13,
+    "end_line": 13,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "Performance",
+    "comment": "The Dockerfile installs build dependencies and then removes them, which is good practice but can be optimized.",
+    "confidence": "moderate",
+    "reason": "Optimizing Dockerfile layers can reduce image size and build time.",
+    "solution": "Combine RUN commands to reduce the number of layers.",
+    "actual_code": "RUN apt-get update && apt-get install -y \\\n    build-essential \\\n    git \\\n    postgresql-server-dev-16\nRUN apt-get remove -y build-essential git postgresql-server-dev-16 \\\n    && apt-get autoremove -y \\\n    && rm -rf /var/lib/apt/lists/* /pgvector",
+    "fixed_code": "RUN apt-get update && apt-get install -y \\\n    build-essential \\\n    git \\\n    postgresql-server-dev-16 \\\n    && git clone https://github.com/pgvector/pgvector.git \\\n    && cd pgvector \\\n    && make \\\n    && make install \\\n    && apt-get remove -y build-essential git postgresql-server-dev-16 \\\n    && apt-get autoremove -y \\\n    && rm -rf /var/lib/apt/lists/* /pgvector",
+    "file_name": "Dockerfile-postgres",
+    "start_line": 4,
+    "end_line": 18,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 5
+  },
+  {
+    "topic": "Code Readability",
+    "comment": "The `chunk_code` function in `code_chunker.py` has nested functions and complex logic that can be refactored for better readability.",
+    "confidence": "important",
+    "reason": "Refactoring complex functions into smaller, well-named functions improves readability and maintainability.",
+    "solution": "Refactor the `chunk_code` function to extract nested functions into separate helper functions.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/code_chunker.py",
+    "start_line": 7,
+    "end_line": 62,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "SQL Injection",
+    "comment": "Potential SQL injection vulnerability in the query construction.",
+    "confidence": "critical",
+    "reason": "Using f-strings to construct SQL queries can lead to SQL injection attacks if user input is not properly sanitized.",
+    "solution": "Use parameterized queries to avoid SQL injection vulnerabilities.",
+    "actual_code": "query = f\"\"\"\nSELECT \n    e.node_id,\n    e.text,\n    e.metadata,\n    1 - (e.embedding <=> %s::vector) as similarity\nFROM \n{self.table_name}e\nJOIN \n    function_abstractions fa ON e.node_id = fa.function_id::text\nJOIN \n    files f ON fa.file_id = f.file_id\nWHERE \n    f.repo_id = %s\nORDER BY \n    similarity DESC\nLIMIT \n    %s\n\"\"\"",
+    "fixed_code": "query = \"\"\"\nSELECT \n    e.node_id,\n    e.text,\n    e.metadata,\n    1 - (e.embedding <=> %s::vector) as similarity\nFROM \n    %s e\nJOIN \n    function_abstractions fa ON e.node_id = fa.function_id::text\nJOIN \n    files f ON fa.file_id = f.file_id\nWHERE \n    f.repo_id = %s\nORDER BY \n    similarity DESC\nLIMIT \n    %s\n\"\"\"",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 19,
+    "end_line": 37,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "Code Readability",
+    "comment": "The normalization of the query embedding can be simplified for better readability.",
+    "confidence": "moderate",
+    "reason": "Simplifying code improves readability and maintainability.",
+    "solution": "Combine the normalization steps into a single line.",
+    "actual_code": "query_embedding_np = np.array(query_embedding)\nquery_embedding_normalized = query_embedding_np / np.linalg.norm(query_embedding_np)",
+    "fixed_code": "query_embedding_normalized = np.array(query_embedding) / np.linalg.norm(query_embedding)",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 15,
+    "end_line": 16,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 3
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Lack of error handling in database operations.",
+    "confidence": "important",
+    "reason": "Database operations can fail; it's important to handle exceptions to avoid crashes.",
+    "solution": "Add try-except blocks to handle potential database errors.",
+    "actual_code": "",
+    "fixed_code": "try:\n    with self.get_client() as client:\n        with client.cursor() as cur:\n            cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))\n            results = cur.fetchall()\nexcept Exception as e:\n    # Handle exception (e.g., log the error, re-raise, etc.)\n    raise e",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 39,
+    "end_line": 42,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Type Annotations",
+    "comment": "Missing type annotations for method parameters and return types.",
+    "confidence": "moderate",
+    "reason": "Type annotations improve code readability and help with static analysis.",
+    "solution": "Add type annotations to the method parameters and return types.",
+    "actual_code": "def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:",
+    "fixed_code": "def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[Dict[str, Any]]:",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 13,
+    "end_line": 13,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Dictionary Default Values",
+    "comment": "Using `None` as a default return value for `get_feedback` method.",
+    "confidence": "low",
+    "reason": "Returning `None` can lead to potential `NoneType` errors if not handled properly.",
+    "solution": "Return an empty dictionary instead of `None`.",
+    "actual_code": "return self.feedback_store.get(code_id, None)",
+    "fixed_code": "return self.feedback_store.get(code_id,{})",
+    "file_name": "kaizen/retriever/feedback_system.py",
+    "start_line": 18,
+    "end_line": 18,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 2
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Exception handling in `generate_abstraction` method is too generic.",
+    "confidence": "important",
+    "reason": "Catching all exceptions without specific handling can obscure the root cause of errors and make debugging difficult.",
+    "solution": "Catch specific exceptions and handle them appropriately.",
+    "actual_code": "except Exception as e:\n    raise e",
+    "fixed_code": "except SomeSpecificException as e:\n    logger.error(f\"Specific error occurred:{str(e)}\")\n    raise e",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 218,
+    "end_line": 219,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 7
+  },
+  {
+    "topic": "Security",
+    "comment": "Database connection strings should not be constructed using string interpolation.",
+    "confidence": "critical",
+    "reason": "Using string interpolation for connection strings can expose the application to SQL injection attacks.",
+    "solution": "Use parameterized queries or a configuration management tool to handle sensitive information.",
+    "actual_code": "self.engine = create_engine(\n    f\"postgresql://{os.environ['POSTGRES_USER']}:{os.environ['POSTGRES_PASSWORD']}@{os.environ['POSTGRES_HOST']}:{os.environ['POSTGRES_PORT']}/{os.environ['POSTGRES_DB']}\",\n    pool_size=10,\n    max_overflow=20,\n)",
+    "fixed_code": "self.engine = create_engine(\n    'postgresql://{user}:{password}@{host}:{port}/{db}'.format(\n        user=os.environ['POSTGRES_USER'],\n        password=os.environ['POSTGRES_PASSWORD'],\n        host=os.environ['POSTGRES_HOST'],\n        port=os.environ['POSTGRES_PORT'],\n        db=os.environ['POSTGRES_DB']\n    ),\n    pool_size=10,\n    max_overflow=20,\n)",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 35,
+    "end_line": 39,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "Performance",
+    "comment": "Using `os.walk` and `ThreadPoolExecutor` for file parsing can be optimized.",
+    "confidence": "important",
+    "reason": "The current implementation may not efficiently utilize available CPU cores and can be improved for better performance.",
+    "solution": "Consider using asynchronous I/O operations or more efficient file traversal methods.",
+    "actual_code": "with ThreadPoolExecutor(max_workers=os.cpu_count()) as executor:\n    futures =[]\n    for root, _, files in os.walk(repo_path):\n        for file in files:\n            if file.endswith((\".py\", \".js\", \".ts\", \".rs\")):\n                file_path = os.path.join(root, file)\n                futures.append(executor.submit(self.parse_file, file_path))",
+    "fixed_code": "import asyncio\nfrom aiofiles import open as aio_open\n\nasync def parse_repository_async(self, repo_path: str):\n    self.total_usage = self.llm_provider.DEFAULT_USAGE\n    logger.info(f\"Starting repository setup for:{repo_path}\")\n    await self.parse_repository(repo_path)\n    self.store_function_relationships()\n    logger.info(\"Repository setup completed successfully\")\n\nasync def parse_file_async(self, file_path: str):\n    logger.debug(f\"Parsing file:{file_path}\")\n    try:\n        async with aio_open(file_path, \"r\", encoding=\"utf-8\") as file:\n            content = await file.read()\n        language = self.get_language_from_extension(file_path)\n        chunked_code = chunk_code(content, language)\n        for section, items in chunked_code.items():\n            if isinstance(items, dict):\n                for name, code_info in items.items():\n                    await self.process_code_block_async(code_info, file_path, section, name)\n            elif isinstance(items, list):\n                for i, code_info in enumerate(items):\n                    await self.process_code_block_async(code_info, file_path, section, f\"{section}_{i}\")\n        logger.debug(f\"Successfully parsed file:{file_path}\")\n    except Exception as e:\n        logger.error(f\"Error processing file{file_path}:{str(e)}\")\n        logger.error(traceback.format_exc())",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 71,
+    "end_line": 79,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "Logging",
+    "comment": "Logging level for parsing files should be more granular.",
+    "confidence": "moderate",
+    "reason": "Using `logger.debug` for file parsing can help in better debugging without cluttering the log files.",
+    "solution": "Change logging level to `debug` for detailed logs during file parsing.",
+    "actual_code": "logger.info(f\"Parsing repository:{repo_path}\")",
+    "fixed_code": "logger.debug(f\"Parsing repository:{repo_path}\")",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 70,
+    "end_line": 70,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Code Readability",
+    "comment": "The `generate_abstraction` method is too long and complex.",
+    "confidence": "important",
+    "reason": "Long methods can be difficult to read and maintain. They should be broken down into smaller, more manageable functions.",
+    "solution": "Refactor the `generate_abstraction` method into smaller helper methods.",
+    "actual_code": "def generate_abstraction(\n    self, code_block: str, language: str, max_tokens: int = 300\n) -> str:\n    prompt = f\"\"\"Generate a concise yet comprehensive abstract description of the following{language}code block. \n    Include information about:\n    1. The purpose or functionality of the code\n    2. Input parameters and return values (if applicable)\n    3. Any important algorithms or data structures used\n    4. Key dependencies or external libraries used\n    5. Any notable design patterns or architectural choices\n    6. Potential edge cases or error handling\n\n    Code:\n    ```{language}\n{code_block}\n    ```\n    \"\"\"\n\n    estimated_prompt_tokens = len(tokenizer.encode(prompt))\n    adjusted_max_tokens = min(max(150, estimated_prompt_tokens), 1000)\n\n    try:\n        abstraction, usage = self.llm_provider.chat_completion(\n            prompt=\"\",\n            messages=[\n{\n                    \"role\": \"system\",\n                    \"content\": \"You are an expert programmer tasked with generating comprehensive and accurate abstractions of code snippets.\",\n},\n{\"role\": \"user\", \"content\": prompt},\n            ],\n            custom_model={\"max_tokens\": adjusted_max_tokens, \"model\": \"small\"},\n        )\n        return abstraction, usage\n\n    except Exception as e:\n        raise e",
+    "fixed_code": "def generate_abstraction(self, code_block: str, language: str, max_tokens: int = 300) -> str:\n    prompt = self._create_prompt(code_block, language)\n    estimated_prompt_tokens = len(tokenizer.encode(prompt))\n    adjusted_max_tokens = min(max(150, estimated_prompt_tokens), 1000)\n\n    try:\n        abstraction, usage = self._get_abstraction_from_llm(prompt, adjusted_max_tokens)\n        return abstraction, usage\n    except Exception as e:\n        raise e\n\n def _create_prompt(self, code_block: str, language: str) -> str:\n    return f\"\"\"Generate a concise yet comprehensive abstract description of the following{language}code block. \n    Include information about:\n    1. The purpose or functionality of the code\n    2. Input parameters and return values (if applicable)\n    3. Any important algorithms or data structures used\n    4. Key dependencies or external libraries used\n    5. Any notable design patterns or architectural choices\n    6. Potential edge cases or error handling\n\n    Code:\n    ```{language}\n{code_block}\n    ```\n    \"\"\"\n\n def _get_abstraction_from_llm(self, prompt: str, max_tokens: int) -> str:\n    return self.llm_provider.chat_completion(\n        prompt=\"\",\n        messages=[\n{\n                \"role\": \"system\",\n                \"content\": \"You are an expert programmer tasked with generating comprehensive and accurate abstractions of code snippets.\",\n},\n{\"role\": \"user\", \"content\": prompt},\n        ],\n        custom_model={\"max_tokens\": max_tokens, \"model\": \"small\"},\n    )",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 184,
+    "end_line": 219,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Logging Configuration",
+    "comment": "Logging configuration should be done in the main entry point of the application, not in the module.",
+    "confidence": "important",
+    "reason": "Configuring logging in a module can lead to unexpected behavior if the module is imported multiple times.",
+    "solution": "Move the logging configuration to the main entry point of the application.",
+    "actual_code": "logging.basicConfig(level=logging.INFO)",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 8,
+    "end_line": 8,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "Exception Handling",
+    "comment": "Broad exception handling should be avoided.",
+    "confidence": "critical",
+    "reason": "Catching all exceptions can hide bugs and make debugging difficult.",
+    "solution": "Catch specific exceptions instead of using a broad except clause.",
+    "actual_code": "except Exception as e:",
+    "fixed_code": "except ImportError as e:\n    logger.error(f\"Failed to import module:{str(e)}\")\n    raise\nexcept ValueError as e:\n    logger.error(f\"Invalid value:{str(e)}\")\n    raise",
+    "file_name": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 28,
+    "end_line": 30,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Function Documentation",
+    "comment": "Public functions should have docstrings.",
+    "confidence": "moderate",
+    "reason": "Docstrings provide a convenient way of associating documentation with functions.",
+    "solution": "Add docstrings to public functions.",
+    "actual_code": "",
+    "fixed_code": "def load_language(language: str) -> Language:\n    \"\"\"\n    Load the specified language.\n    :param language: The name of the language to load.\n    :return: The loaded Language object.\n    \"\"\"",
+    "file_name": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 15,
+    "end_line": 15,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Code Duplication",
+    "comment": "Duplicate code found in test cases.",
+    "confidence": "important",
+    "reason": "Duplicate code can lead to maintenance issues and bugs.",
+    "solution": "Refactor the duplicate code into a helper function.",
+    "actual_code": "print_chunks(\"JavaScript\", chunk_code(javascript_code, \"javascript\"))",
+    "fixed_code": "def test_chunk(language, code):\n    print_chunks(language, chunk_code(code, language))\n\ntest_chunk(\"Python\", python_code)\ntest_chunk(\"JavaScript\", javascript_code)\ntest_chunk(\"React\", react_nextjs_code)",
+    "file_name": "tests/retriever/test_chunker.py",
+    "start_line": 98,
+    "end_line": 101,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Versioning",
+    "comment": "Version bump should be accompanied by a changelog update.",
+    "confidence": "moderate",
+    "reason": "A changelog helps track changes and improvements in the project.",
+    "solution": "Update the changelog to reflect the changes made in this version.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "pyproject.toml",
+    "start_line": 3,
+    "end_line": 3,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 3
+  },
+  {
+    "topic": "Dependency Management",
+    "comment": "New dependencies added without justification.",
+    "confidence": "important",
+    "reason": "Adding dependencies increases the attack surface and maintenance burden.",
+    "solution": "Provide justification for new dependencies in the pull request description.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "pyproject.toml",
+    "start_line": 27,
+    "end_line": 49,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Docker",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to Dockerfile, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_name": "Dockerfile",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Docker",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to docker-compose.yml, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "15",
+    "end_line": "15",
+    "side": "RIGHT",
+    "file_name": "docker-compose.yml",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Version Control",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to .gitignore, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "164",
+    "end_line": "164",
+    "side": "RIGHT",
+    "file_name": ".gitignore",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Database",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to db_setup/init.sql, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "db_setup/init.sql",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_222/review.md b/.experiments/code_review/gpt-4o/no_eval/pr_222/review.md
new file mode 100644
index 00000000..5a796ee4
--- /dev/null
+++ b/.experiments/code_review/gpt-4o/no_eval/pr_222/review.md
@@ -0,0 +1,560 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/222
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 26
+- Critical: 9
+- Important: 9
+- Minor: 7
+- Files Affected: 14
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Security (9 issues)</strong></summary>
+
+### 1. Hardcoding API keys in `config.json` can lead to security vulnerabilities.
+📁 **File:** `config.json:13`
+⚖️ **Severity:** 9/10
+🔍 **Description:** Exposing API keys in the codebase can lead to unauthorized access.
+💡 **Solution:** Use environment variables to store API keys instead of hardcoding them.
+
+**Current Code:**
+```python
+"api_key": "os.environ/AZURE_API_KEY"
+```
+
+**Suggested Code:**
+```python
+"api_key": "${AZURE_API_KEY}"
+```
+
+### 2. Potential SQL injection vulnerability in the query construction.
+📁 **File:** `kaizen/retriever/custom_vector_store.py:19`
+⚖️ **Severity:** 9/10
+🔍 **Description:** Using f-strings to construct SQL queries can lead to SQL injection attacks if user input is not properly sanitized.
+💡 **Solution:** Use parameterized queries to avoid SQL injection vulnerabilities.
+
+**Current Code:**
+```python
+query = f"""
+SELECT 
+    e.node_id,
+    e.text,
+    e.metadata,
+    1 - (e.embedding <=> %s::vector) as similarity
+FROM 
+{self.table_name}e
+JOIN 
+    function_abstractions fa ON e.node_id = fa.function_id::text
+JOIN 
+    files f ON fa.file_id = f.file_id
+WHERE 
+    f.repo_id = %s
+ORDER BY 
+    similarity DESC
+LIMIT 
+    %s
+"""
+```
+
+**Suggested Code:**
+```python
+query = """
+SELECT 
+    e.node_id,
+    e.text,
+    e.metadata,
+    1 - (e.embedding <=> %s::vector) as similarity
+FROM 
+    %s e
+JOIN 
+    function_abstractions fa ON e.node_id = fa.function_id::text
+JOIN 
+    files f ON fa.file_id = f.file_id
+WHERE 
+    f.repo_id = %s
+ORDER BY 
+    similarity DESC
+LIMIT 
+    %s
+"""
+```
+
+### 3. Database connection strings should not be constructed using string interpolation.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:35`
+⚖️ **Severity:** 9/10
+🔍 **Description:** Using string interpolation for connection strings can expose the application to SQL injection attacks.
+💡 **Solution:** Use parameterized queries or a configuration management tool to handle sensitive information.
+
+**Current Code:**
+```python
+self.engine = create_engine(
+    f"postgresql://{os.environ['POSTGRES_USER']}:{os.environ['POSTGRES_PASSWORD']}@{os.environ['POSTGRES_HOST']}:{os.environ['POSTGRES_PORT']}/{os.environ['POSTGRES_DB']}",
+    pool_size=10,
+    max_overflow=20,
+)
+```
+
+**Suggested Code:**
+```python
+self.engine = create_engine(
+    'postgresql://{user}:{password}@{host}:{port}/{db}'.format(
+        user=os.environ['POSTGRES_USER'],
+        password=os.environ['POSTGRES_PASSWORD'],
+        host=os.environ['POSTGRES_HOST'],
+        port=os.environ['POSTGRES_PORT'],
+        db=os.environ['POSTGRES_DB']
+    ),
+    pool_size=10,
+    max_overflow=20,
+)
+```
+
+### 4. Broad exception handling should be avoided.
+📁 **File:** `kaizen/retriever/tree_sitter_utils.py:28`
+⚖️ **Severity:** 8/10
+🔍 **Description:** Catching all exceptions can hide bugs and make debugging difficult.
+💡 **Solution:** Catch specific exceptions instead of using a broad except clause.
+
+**Current Code:**
+```python
+except Exception as e:
+```
+
+**Suggested Code:**
+```python
+except ImportError as e:
+    logger.error(f"Failed to import module:{str(e)}")
+    raise
+except ValueError as e:
+    logger.error(f"Invalid value:{str(e)}")
+    raise
+```
+
+### 5. Changes made to sensitive file
+📁 **File:** `config.json:4`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to config.json, which needs review
+💡 **Solution:** NA
+
+### 6. Changes made to sensitive file
+📁 **File:** `Dockerfile:4`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to Dockerfile, which needs review
+💡 **Solution:** NA
+
+### 7. Changes made to sensitive file
+📁 **File:** `docker-compose.yml:15`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to docker-compose.yml, which needs review
+💡 **Solution:** NA
+
+### 8. Changes made to sensitive file
+📁 **File:** `.gitignore:164`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to .gitignore, which needs review
+💡 **Solution:** NA
+
+### 9. Changes made to sensitive file
+📁 **File:** `db_setup/init.sql:1`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to db_setup/init.sql, which needs review
+💡 **Solution:** NA
+
+</details>
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Error Handling (9 issues)</strong></summary>
+
+### 1. The `install_tree_sitter_languages.sh` script does not handle errors during the installation process.
+📁 **File:** `install_tree_sitter_languages.sh:1`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Error handling ensures that the script fails gracefully and provides useful error messages.
+💡 **Solution:** Add error handling for each critical step in the script.
+
+### 2. The `chunk_code` function in `code_chunker.py` has nested functions and complex logic that can be refactored for better readability.
+📁 **File:** `kaizen/retriever/code_chunker.py:7`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Refactoring complex functions into smaller, well-named functions improves readability and maintainability.
+💡 **Solution:** Refactor the `chunk_code` function to extract nested functions into separate helper functions.
+
+### 3. Lack of error handling in database operations.
+📁 **File:** `kaizen/retriever/custom_vector_store.py:39`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Database operations can fail; it's important to handle exceptions to avoid crashes.
+💡 **Solution:** Add try-except blocks to handle potential database errors.
+
+**Current Code:**
+```python
+
+```
+
+**Suggested Code:**
+```python
+try:
+    with self.get_client() as client:
+        with client.cursor() as cur:
+            cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))
+            results = cur.fetchall()
+except Exception as e:
+    # Handle exception (e.g., log the error, re-raise, etc.)
+    raise e
+```
+
+### 4. Exception handling in `generate_abstraction` method is too generic.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:218`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Catching all exceptions without specific handling can obscure the root cause of errors and make debugging difficult.
+💡 **Solution:** Catch specific exceptions and handle them appropriately.
+
+**Current Code:**
+```python
+except Exception as e:
+    raise e
+```
+
+**Suggested Code:**
+```python
+except SomeSpecificException as e:
+    logger.error(f"Specific error occurred:{str(e)}")
+    raise e
+```
+
+### 5. Using `os.walk` and `ThreadPoolExecutor` for file parsing can be optimized.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:71`
+⚖️ **Severity:** 6/10
+🔍 **Description:** The current implementation may not efficiently utilize available CPU cores and can be improved for better performance.
+💡 **Solution:** Consider using asynchronous I/O operations or more efficient file traversal methods.
+
+**Current Code:**
+```python
+with ThreadPoolExecutor(max_workers=os.cpu_count()) as executor:
+    futures =[]
+    for root, _, files in os.walk(repo_path):
+        for file in files:
+            if file.endswith((".py", ".js", ".ts", ".rs")):
+                file_path = os.path.join(root, file)
+                futures.append(executor.submit(self.parse_file, file_path))
+```
+
+**Suggested Code:**
+```python
+import asyncio
+from aiofiles import open as aio_open
+
+async def parse_repository_async(self, repo_path: str):
+    self.total_usage = self.llm_provider.DEFAULT_USAGE
+    logger.info(f"Starting repository setup for:{repo_path}")
+    await self.parse_repository(repo_path)
+    self.store_function_relationships()
+    logger.info("Repository setup completed successfully")
+
+async def parse_file_async(self, file_path: str):
+    logger.debug(f"Parsing file:{file_path}")
+    try:
+        async with aio_open(file_path, "r", encoding="utf-8") as file:
+            content = await file.read()
+        language = self.get_language_from_extension(file_path)
+        chunked_code = chunk_code(content, language)
+        for section, items in chunked_code.items():
+            if isinstance(items, dict):
+                for name, code_info in items.items():
+                    await self.process_code_block_async(code_info, file_path, section, name)
+            elif isinstance(items, list):
+                for i, code_info in enumerate(items):
+                    await self.process_code_block_async(code_info, file_path, section, f"{section}_{i}")
+        logger.debug(f"Successfully parsed file:{file_path}")
+    except Exception as e:
+        logger.error(f"Error processing file{file_path}:{str(e)}")
+        logger.error(traceback.format_exc())
+```
+
+### 6. The `generate_abstraction` method is too long and complex.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:184`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Long methods can be difficult to read and maintain. They should be broken down into smaller, more manageable functions.
+💡 **Solution:** Refactor the `generate_abstraction` method into smaller helper methods.
+
+**Current Code:**
+```python
+def generate_abstraction(
+    self, code_block: str, language: str, max_tokens: int = 300
+) -> str:
+    prompt = f"""Generate a concise yet comprehensive abstract description of the following{language}code block. 
+    Include information about:
+    1. The purpose or functionality of the code
+    2. Input parameters and return values (if applicable)
+    3. Any important algorithms or data structures used
+    4. Key dependencies or external libraries used
+    5. Any notable design patterns or architectural choices
+    6. Potential edge cases or error handling
+
+    Code:
+    ```{language}
+{code_block}
+    ```
+    """
+
+    estimated_prompt_tokens = len(tokenizer.encode(prompt))
+    adjusted_max_tokens = min(max(150, estimated_prompt_tokens), 1000)
+
+    try:
+        abstraction, usage = self.llm_provider.chat_completion(
+            prompt="",
+            messages=[
+{
+                    "role": "system",
+                    "content": "You are an expert programmer tasked with generating comprehensive and accurate abstractions of code snippets.",
+},
+{"role": "user", "content": prompt},
+            ],
+            custom_model={"max_tokens": adjusted_max_tokens, "model": "small"},
+        )
+        return abstraction, usage
+
+    except Exception as e:
+        raise e
+```
+
+**Suggested Code:**
+```python
+def generate_abstraction(self, code_block: str, language: str, max_tokens: int = 300) -> str:
+    prompt = self._create_prompt(code_block, language)
+    estimated_prompt_tokens = len(tokenizer.encode(prompt))
+    adjusted_max_tokens = min(max(150, estimated_prompt_tokens), 1000)
+
+    try:
+        abstraction, usage = self._get_abstraction_from_llm(prompt, adjusted_max_tokens)
+        return abstraction, usage
+    except Exception as e:
+        raise e
+
+ def _create_prompt(self, code_block: str, language: str) -> str:
+    return f"""Generate a concise yet comprehensive abstract description of the following{language}code block. 
+    Include information about:
+    1. The purpose or functionality of the code
+    2. Input parameters and return values (if applicable)
+    3. Any important algorithms or data structures used
+    4. Key dependencies or external libraries used
+    5. Any notable design patterns or architectural choices
+    6. Potential edge cases or error handling
+
+    Code:
+    ```{language}
+{code_block}
+    ```
+    """
+
+ def _get_abstraction_from_llm(self, prompt: str, max_tokens: int) -> str:
+    return self.llm_provider.chat_completion(
+        prompt="",
+        messages=[
+{
+                "role": "system",
+                "content": "You are an expert programmer tasked with generating comprehensive and accurate abstractions of code snippets.",
+},
+{"role": "user", "content": prompt},
+        ],
+        custom_model={"max_tokens": max_tokens, "model": "small"},
+    )
+```
+
+### 7. Logging configuration should be done in the main entry point of the application, not in the module.
+📁 **File:** `kaizen/retriever/tree_sitter_utils.py:8`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Configuring logging in a module can lead to unexpected behavior if the module is imported multiple times.
+💡 **Solution:** Move the logging configuration to the main entry point of the application.
+
+**Current Code:**
+```python
+logging.basicConfig(level=logging.INFO)
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 8. Duplicate code found in test cases.
+📁 **File:** `tests/retriever/test_chunker.py:98`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Duplicate code can lead to maintenance issues and bugs.
+💡 **Solution:** Refactor the duplicate code into a helper function.
+
+**Current Code:**
+```python
+print_chunks("JavaScript", chunk_code(javascript_code, "javascript"))
+```
+
+**Suggested Code:**
+```python
+def test_chunk(language, code):
+    print_chunks(language, chunk_code(code, language))
+
+test_chunk("Python", python_code)
+test_chunk("JavaScript", javascript_code)
+test_chunk("React", react_nextjs_code)
+```
+
+### 9. New dependencies added without justification.
+📁 **File:** `pyproject.toml:27`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Adding dependencies increases the attack surface and maintenance burden.
+💡 **Solution:** Provide justification for new dependencies in the pull request description.
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (8 issues)</strong></summary>
+
+<details>
+<summary><strong>Documentation (7 issues)</strong></summary>
+
+### 1. The `install_tree_sitter_languages.sh` script lacks comments explaining the purpose of each step.
+📁 **File:** `install_tree_sitter_languages.sh:1`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Comments improve readability and maintainability, especially for complex scripts.
+💡 **Solution:** Add comments explaining the purpose of each step in the script.
+
+### 2. The Dockerfile installs build dependencies and then removes them, which is good practice but can be optimized.
+📁 **File:** `Dockerfile-postgres:4`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Optimizing Dockerfile layers can reduce image size and build time.
+💡 **Solution:** Combine RUN commands to reduce the number of layers.
+
+**Current Code:**
+```python
+RUN apt-get update && apt-get install -y \
+    build-essential \
+    git \
+    postgresql-server-dev-16
+RUN apt-get remove -y build-essential git postgresql-server-dev-16 \
+    && apt-get autoremove -y \
+    && rm -rf /var/lib/apt/lists/* /pgvector
+```
+
+**Suggested Code:**
+```python
+RUN apt-get update && apt-get install -y \
+    build-essential \
+    git \
+    postgresql-server-dev-16 \
+    && git clone https://github.com/pgvector/pgvector.git \
+    && cd pgvector \
+    && make \
+    && make install \
+    && apt-get remove -y build-essential git postgresql-server-dev-16 \
+    && apt-get autoremove -y \
+    && rm -rf /var/lib/apt/lists/* /pgvector
+```
+
+### 3. The normalization of the query embedding can be simplified for better readability.
+📁 **File:** `kaizen/retriever/custom_vector_store.py:15`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Simplifying code improves readability and maintainability.
+💡 **Solution:** Combine the normalization steps into a single line.
+
+**Current Code:**
+```python
+query_embedding_np = np.array(query_embedding)
+query_embedding_normalized = query_embedding_np / np.linalg.norm(query_embedding_np)
+```
+
+**Suggested Code:**
+```python
+query_embedding_normalized = np.array(query_embedding) / np.linalg.norm(query_embedding)
+```
+
+### 4. Missing type annotations for method parameters and return types.
+📁 **File:** `kaizen/retriever/custom_vector_store.py:13`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Type annotations improve code readability and help with static analysis.
+💡 **Solution:** Add type annotations to the method parameters and return types.
+
+**Current Code:**
+```python
+def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:
+```
+
+**Suggested Code:**
+```python
+def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[Dict[str, Any]]:
+```
+
+### 5. Logging level for parsing files should be more granular.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:70`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Using `logger.debug` for file parsing can help in better debugging without cluttering the log files.
+💡 **Solution:** Change logging level to `debug` for detailed logs during file parsing.
+
+**Current Code:**
+```python
+logger.info(f"Parsing repository:{repo_path}")
+```
+
+**Suggested Code:**
+```python
+logger.debug(f"Parsing repository:{repo_path}")
+```
+
+### 6. Public functions should have docstrings.
+📁 **File:** `kaizen/retriever/tree_sitter_utils.py:15`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Docstrings provide a convenient way of associating documentation with functions.
+💡 **Solution:** Add docstrings to public functions.
+
+**Current Code:**
+```python
+
+```
+
+**Suggested Code:**
+```python
+def load_language(language: str) -> Language:
+    """
+    Load the specified language.
+    :param language: The name of the language to load.
+    :return: The loaded Language object.
+    """
+```
+
+### 7. Version bump should be accompanied by a changelog update.
+📁 **File:** `pyproject.toml:3`
+⚖️ **Severity:** 3/10
+🔍 **Description:** A changelog helps track changes and improvements in the project.
+💡 **Solution:** Update the changelog to reflect the changes made in this version.
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-2024-05-13)
+{"prompt_tokens": 21545, "completion_tokens": 5255, "total_tokens": 26800}
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_335/comments.json b/.experiments/code_review/gpt-4o/no_eval/pr_335/comments.json
new file mode 100644
index 00000000..7ab211a4
--- /dev/null
+++ b/.experiments/code_review/gpt-4o/no_eval/pr_335/comments.json
@@ -0,0 +1,32 @@
+[
+  {
+    "topic": "Function Parameters",
+    "comment": "Removed `reeval_response` parameter from multiple functions. Ensure that this parameter is no longer needed and does not affect the functionality.",
+    "confidence": "critical",
+    "reason": "Removing a parameter can lead to missing functionality if the parameter was being used elsewhere in the code.",
+    "solution": "Verify that `reeval_response` is not required for the functions to operate correctly.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 43,
+    "end_line": 43,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Function Implementation",
+    "comment": "Changed the implementation of `_process_full_diff` to remove `reeval_response` logic. Ensure that the re-evaluation logic is no longer needed.",
+    "confidence": "critical",
+    "reason": "Removing logic can lead to missing functionality if the logic was essential.",
+    "solution": "Verify that the re-evaluation logic is not required for the function to operate correctly.",
+    "actual_code": "resp, usage = self.provider.chat_completion(prompt, user=user)\ndesc = parser.extract_code_from_markdown(resp)\nreturn desc",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 79,
+    "end_line": 83,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 8
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_335/issues.json b/.experiments/code_review/gpt-4o/no_eval/pr_335/issues.json
new file mode 100644
index 00000000..49da45e6
--- /dev/null
+++ b/.experiments/code_review/gpt-4o/no_eval/pr_335/issues.json
@@ -0,0 +1,92 @@
+[
+  {
+    "topic": "Imports",
+    "comment": "The import statement was changed to import from a different module. Ensure that the new module contains the required constants.",
+    "confidence": "important",
+    "reason": "Changing import paths can lead to runtime errors if the new module does not contain the expected constants.",
+    "solution": "Verify that `PR_DESCRIPTION_SYSTEM_PROMPT` exists in `pr_desc_prompts` and is correctly defined.",
+    "actual_code": "from kaizen.llms.prompts.pr_desc_prompts import (",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 8,
+    "end_line": 8,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "Initialization",
+    "comment": "The system prompt was changed from `CODE_REVIEW_SYSTEM_PROMPT` to `PR_DESCRIPTION_SYSTEM_PROMPT`. Ensure this change aligns with the intended functionality.",
+    "confidence": "important",
+    "reason": "Changing the system prompt can alter the behavior of the LLMProvider, potentially affecting the output.",
+    "solution": "Confirm that `PR_DESCRIPTION_SYSTEM_PROMPT` is the correct prompt for the intended functionality.",
+    "actual_code": "self.provider.system_prompt = PR_DESCRIPTION_SYSTEM_PROMPT",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 28,
+    "end_line": 28,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "Function Parameters",
+    "comment": "Removed `reeval_response` parameter from multiple functions. Ensure that this parameter is no longer needed and does not affect the functionality.",
+    "confidence": "critical",
+    "reason": "Removing a parameter can lead to missing functionality if the parameter was being used elsewhere in the code.",
+    "solution": "Verify that `reeval_response` is not required for the functions to operate correctly.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 43,
+    "end_line": 43,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Function Calls",
+    "comment": "Changed function calls to remove `reeval_response` parameter. Ensure that the new function signatures match the updated calls.",
+    "confidence": "important",
+    "reason": "Mismatch in function signatures can lead to runtime errors.",
+    "solution": "Ensure that `_process_full_diff` and `_process_files` functions do not require `reeval_response` parameter.",
+    "actual_code": "desc = self._process_full_diff(prompt, user)",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 52,
+    "end_line": 52,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "Function Implementation",
+    "comment": "Changed the implementation of `_process_full_diff` to remove `reeval_response` logic. Ensure that the re-evaluation logic is no longer needed.",
+    "confidence": "critical",
+    "reason": "Removing logic can lead to missing functionality if the logic was essential.",
+    "solution": "Verify that the re-evaluation logic is not required for the function to operate correctly.",
+    "actual_code": "resp, usage = self.provider.chat_completion(prompt, user=user)\ndesc = parser.extract_code_from_markdown(resp)\nreturn desc",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 79,
+    "end_line": 83,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Prompt Definitions",
+    "comment": "Added new prompt definitions in `pr_desc_prompts.py`. Ensure that these new prompts are correctly formatted and used in the code.",
+    "confidence": "moderate",
+    "reason": "New prompt definitions need to be correctly formatted and integrated into the existing codebase.",
+    "solution": "Review the new prompt definitions for correctness and ensure they are used appropriately.",
+    "actual_code": "PR_DESCRIPTION_SYSTEM_PROMPT = \"\"\"\nAs a senior software developer reviewing code submissions, provide thorough, constructive feedback and suggestions for improvements. Consider best practices, error handling, performance, readability, and maintainability. Offer objective and respectful reviews that help developers enhance their skills and code quality. Use your expertise to provide comprehensive feedback without asking clarifying questions.\n\"\"\"",
+    "fixed_code": "",
+    "file_name": "kaizen/llms/prompts/pr_desc_prompts.py",
+    "start_line": 1,
+    "end_line": 3,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_335/review.md b/.experiments/code_review/gpt-4o/no_eval/pr_335/review.md
new file mode 100644
index 00000000..e13a98d6
--- /dev/null
+++ b/.experiments/code_review/gpt-4o/no_eval/pr_335/review.md
@@ -0,0 +1,151 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/335
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 6
+- Critical: 2
+- Important: 3
+- Minor: 1
+- Files Affected: 2
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Function Parameters (2 issues)</strong></summary>
+
+### 1. Removed `reeval_response` parameter from multiple functions. Ensure that this parameter is no longer needed and does not affect the functionality.
+📁 **File:** `kaizen/generator/pr_description.py:43`
+⚖️ **Severity:** 8/10
+🔍 **Description:** Removing a parameter can lead to missing functionality if the parameter was being used elsewhere in the code.
+💡 **Solution:** Verify that `reeval_response` is not required for the functions to operate correctly.
+
+### 2. Changed the implementation of `_process_full_diff` to remove `reeval_response` logic. Ensure that the re-evaluation logic is no longer needed.
+📁 **File:** `kaizen/generator/pr_description.py:79`
+⚖️ **Severity:** 8/10
+🔍 **Description:** Removing logic can lead to missing functionality if the logic was essential.
+💡 **Solution:** Verify that the re-evaluation logic is not required for the function to operate correctly.
+
+**Current Code:**
+```python
+resp, usage = self.provider.chat_completion(prompt, user=user)
+desc = parser.extract_code_from_markdown(resp)
+return desc
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Imports (3 issues)</strong></summary>
+
+### 1. The import statement was changed to import from a different module. Ensure that the new module contains the required constants.
+📁 **File:** `kaizen/generator/pr_description.py:8`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Changing import paths can lead to runtime errors if the new module does not contain the expected constants.
+💡 **Solution:** Verify that `PR_DESCRIPTION_SYSTEM_PROMPT` exists in `pr_desc_prompts` and is correctly defined.
+
+**Current Code:**
+```python
+from kaizen.llms.prompts.pr_desc_prompts import (
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 2. The system prompt was changed from `CODE_REVIEW_SYSTEM_PROMPT` to `PR_DESCRIPTION_SYSTEM_PROMPT`. Ensure this change aligns with the intended functionality.
+📁 **File:** `kaizen/generator/pr_description.py:28`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Changing the system prompt can alter the behavior of the LLMProvider, potentially affecting the output.
+💡 **Solution:** Confirm that `PR_DESCRIPTION_SYSTEM_PROMPT` is the correct prompt for the intended functionality.
+
+**Current Code:**
+```python
+self.provider.system_prompt = PR_DESCRIPTION_SYSTEM_PROMPT
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 3. Changed function calls to remove `reeval_response` parameter. Ensure that the new function signatures match the updated calls.
+📁 **File:** `kaizen/generator/pr_description.py:52`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Mismatch in function signatures can lead to runtime errors.
+💡 **Solution:** Ensure that `_process_full_diff` and `_process_files` functions do not require `reeval_response` parameter.
+
+**Current Code:**
+```python
+desc = self._process_full_diff(prompt, user)
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (1 issues)</strong></summary>
+
+<details>
+<summary><strong>Prompt Definitions (1 issues)</strong></summary>
+
+### 1. Added new prompt definitions in `pr_desc_prompts.py`. Ensure that these new prompts are correctly formatted and used in the code.
+📁 **File:** `kaizen/llms/prompts/pr_desc_prompts.py:1`
+⚖️ **Severity:** 4/10
+🔍 **Description:** New prompt definitions need to be correctly formatted and integrated into the existing codebase.
+💡 **Solution:** Review the new prompt definitions for correctness and ensure they are used appropriately.
+
+**Current Code:**
+```python
+PR_DESCRIPTION_SYSTEM_PROMPT = """
+As a senior software developer reviewing code submissions, provide thorough, constructive feedback and suggestions for improvements. Consider best practices, error handling, performance, readability, and maintainability. Offer objective and respectful reviews that help developers enhance their skills and code quality. Use your expertise to provide comprehensive feedback without asking clarifying questions.
+"""
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-2024-05-13)
+{"prompt_tokens": 6751, "completion_tokens": 1126, "total_tokens": 7877}
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_440/comments.json b/.experiments/code_review/gpt-4o/no_eval/pr_440/comments.json
new file mode 100644
index 00000000..0637a088
--- /dev/null
+++ b/.experiments/code_review/gpt-4o/no_eval/pr_440/comments.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_440/issues.json b/.experiments/code_review/gpt-4o/no_eval/pr_440/issues.json
new file mode 100644
index 00000000..ccaaf465
--- /dev/null
+++ b/.experiments/code_review/gpt-4o/no_eval/pr_440/issues.json
@@ -0,0 +1,47 @@
+[
+  {
+    "topic": "Comment Clarity",
+    "comment": "The comment 'TODO: DONT PUSH DUPLICATE' is ambiguous and lacks context.",
+    "confidence": "moderate",
+    "reason": "Comments should provide clear guidance or context for future developers.",
+    "solution": "Provide a more descriptive comment explaining what needs to be done to avoid pushing duplicates.",
+    "actual_code": "# TODO: DONT PUSH DUPLICATE",
+    "fixed_code": "# TODO: Ensure that duplicate embeddings are not pushed to the database.",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 157,
+    "end_line": 157,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 3
+  },
+  {
+    "topic": "Code Removal Impact",
+    "comment": "The removal of the llama-index-llms-openai dependency might cause issues if any part of the codebase relies on it.",
+    "confidence": "important",
+    "reason": "Removing dependencies without ensuring they are not used elsewhere can lead to runtime errors.",
+    "solution": "Verify that no part of the codebase uses llama-index-llms-openai before removing it.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "pyproject.toml",
+    "start_line": 27,
+    "end_line": 28,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Hardcoded Paths",
+    "comment": "The path './github_app/' is hardcoded, which can cause issues in different environments.",
+    "confidence": "important",
+    "reason": "Hardcoded paths can lead to errors when the code is run in different environments or directories.",
+    "solution": "Use a configuration file or environment variables to manage paths.",
+    "actual_code": "analyzer.setup_repository(\"./github_app/\")",
+    "fixed_code": "analyzer.setup_repository(os.getenv('GITHUB_APP_PATH', './github_app/'))",
+    "file_name": "examples/ragify_codebase/main.py",
+    "start_line": 7,
+    "end_line": 7,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_440/review.md b/.experiments/code_review/gpt-4o/no_eval/pr_440/review.md
new file mode 100644
index 00000000..ab3069eb
--- /dev/null
+++ b/.experiments/code_review/gpt-4o/no_eval/pr_440/review.md
@@ -0,0 +1,92 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/440
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 3
+- Critical: 0
+- Important: 2
+- Minor: 1
+- Files Affected: 3
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Code Removal Impact (2 issues)</strong></summary>
+
+### 1. The removal of the llama-index-llms-openai dependency might cause issues if any part of the codebase relies on it.
+📁 **File:** `pyproject.toml:27`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Removing dependencies without ensuring they are not used elsewhere can lead to runtime errors.
+💡 **Solution:** Verify that no part of the codebase uses llama-index-llms-openai before removing it.
+
+### 2. The path './github_app/' is hardcoded, which can cause issues in different environments.
+📁 **File:** `examples/ragify_codebase/main.py:7`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Hardcoded paths can lead to errors when the code is run in different environments or directories.
+💡 **Solution:** Use a configuration file or environment variables to manage paths.
+
+**Current Code:**
+```python
+analyzer.setup_repository("./github_app/")
+```
+
+**Suggested Code:**
+```python
+analyzer.setup_repository(os.getenv('GITHUB_APP_PATH', './github_app/'))
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (1 issues)</strong></summary>
+
+<details>
+<summary><strong>Comment Clarity (1 issues)</strong></summary>
+
+### 1. The comment 'TODO: DONT PUSH DUPLICATE' is ambiguous and lacks context.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:157`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Comments should provide clear guidance or context for future developers.
+💡 **Solution:** Provide a more descriptive comment explaining what needs to be done to avoid pushing duplicates.
+
+**Current Code:**
+```python
+# TODO: DONT PUSH DUPLICATE
+```
+
+**Suggested Code:**
+```python
+# TODO: Ensure that duplicate embeddings are not pushed to the database.
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-2024-05-13)
+{"prompt_tokens": 1364, "completion_tokens": 531, "total_tokens": 1895}
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_476/comments.json b/.experiments/code_review/gpt-4o/no_eval/pr_476/comments.json
new file mode 100644
index 00000000..f9fab36c
--- /dev/null
+++ b/.experiments/code_review/gpt-4o/no_eval/pr_476/comments.json
@@ -0,0 +1,31 @@
+[
+  {
+    "topic": "Error Handling",
+    "comment": "Generic exception handling without logging specific error details.",
+    "confidence": "critical",
+    "reason": "Using a generic `except Exception` block without logging the specific error details can make debugging difficult.",
+    "solution": "Log the specific error message in the exception block.",
+    "actual_code": "except Exception:\n    print(\"Error\")",
+    "fixed_code": "except Exception as e:\n    print(f\"Error:{e}\")",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 140,
+    "end_line": 141,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_476/issues.json b/.experiments/code_review/gpt-4o/no_eval/pr_476/issues.json
new file mode 100644
index 00000000..e5c492a5
--- /dev/null
+++ b/.experiments/code_review/gpt-4o/no_eval/pr_476/issues.json
@@ -0,0 +1,91 @@
+[
+  {
+    "topic": "Error Handling",
+    "comment": "Generic exception handling without logging specific error details.",
+    "confidence": "critical",
+    "reason": "Using a generic `except Exception` block without logging the specific error details can make debugging difficult.",
+    "solution": "Log the specific error message in the exception block.",
+    "actual_code": "except Exception:\n    print(\"Error\")",
+    "fixed_code": "except Exception as e:\n    print(f\"Error:{e}\")",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 140,
+    "end_line": 141,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "Code Readability",
+    "comment": "Unnecessary print statements left in the code.",
+    "confidence": "important",
+    "reason": "Leaving print statements in production code can clutter the output and is generally not recommended.",
+    "solution": "Remove or replace print statements with proper logging.",
+    "actual_code": "print(\"diff: \", diff_text)\nprint(\"pr_files\", pr_files)",
+    "fixed_code": "",
+    "file_name": "examples/code_review/main.py",
+    "start_line": 21,
+    "end_line": 22,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Function Signature",
+    "comment": "Modified function signature without updating all references.",
+    "confidence": "important",
+    "reason": "Changing a function signature without updating all references can lead to runtime errors.",
+    "solution": "Ensure all references to `post_pull_request` are updated to include the new `tests` parameter.",
+    "actual_code": "def post_pull_request(url, data, installation_id, tests=None):",
+    "fixed_code": "",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 107,
+    "end_line": 107,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 7
+  },
+  {
+    "topic": "Code Maintainability",
+    "comment": "Redundant code for sorting files.",
+    "confidence": "moderate",
+    "reason": "The custom sorting logic can be replaced with Python's built-in sorting functions for better readability and maintainability.",
+    "solution": "Use Python's `sorted` function with a key parameter.",
+    "actual_code": "sorted_files =[]\nfor file in files:\n    min_index = len(sorted_files)\n    file_name = file[\"filename\"]\n    for i, sorted_file in enumerate(sorted_files):\n        if file_name < sorted_file[\"filename\"]:\n            min_index = i\n            break\n    sorted_files.insert(min_index, file)\nreturn sorted_files",
+    "fixed_code": "sorted_files = sorted(files, key=lambda x: x[\"filename\"])\nreturn sorted_files",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 185,
+    "end_line": 194,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 5
+  },
+  {
+    "topic": "Code Quality",
+    "comment": "Unnecessary variable assignment.",
+    "confidence": "low",
+    "reason": "Assigning `issues` in the loop is unnecessary and can be removed.",
+    "solution": "Remove the assignment of `issues` within the loop.",
+    "actual_code": "issues = review",
+    "fixed_code": "",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 153,
+    "end_line": 153,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 3
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_476/review.md b/.experiments/code_review/gpt-4o/no_eval/pr_476/review.md
new file mode 100644
index 00000000..06252bed
--- /dev/null
+++ b/.experiments/code_review/gpt-4o/no_eval/pr_476/review.md
@@ -0,0 +1,144 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/476
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 6
+- Critical: 2
+- Important: 2
+- Minor: 1
+- Files Affected: 3
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Error Handling (2 issues)</strong></summary>
+
+### 1. Generic exception handling without logging specific error details.
+📁 **File:** `github_app/github_helper/pull_requests.py:140`
+⚖️ **Severity:** 9/10
+🔍 **Description:** Using a generic `except Exception` block without logging the specific error details can make debugging difficult.
+💡 **Solution:** Log the specific error message in the exception block.
+
+**Current Code:**
+```python
+except Exception:
+    print("Error")
+```
+
+**Suggested Code:**
+```python
+except Exception as e:
+    print(f"Error:{e}")
+```
+
+### 2. Changes made to sensitive file
+📁 **File:** `config.json:1`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to config.json, which needs review
+💡 **Solution:** NA
+
+</details>
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Code Readability (2 issues)</strong></summary>
+
+### 1. Unnecessary print statements left in the code.
+📁 **File:** `examples/code_review/main.py:21`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Leaving print statements in production code can clutter the output and is generally not recommended.
+💡 **Solution:** Remove or replace print statements with proper logging.
+
+**Current Code:**
+```python
+print("diff: ", diff_text)
+print("pr_files", pr_files)
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 2. Modified function signature without updating all references.
+📁 **File:** `github_app/github_helper/pull_requests.py:107`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Changing a function signature without updating all references can lead to runtime errors.
+💡 **Solution:** Ensure all references to `post_pull_request` are updated to include the new `tests` parameter.
+
+**Current Code:**
+```python
+def post_pull_request(url, data, installation_id, tests=None):
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (2 issues)</strong></summary>
+
+<details>
+<summary><strong>Code Maintainability (1 issues)</strong></summary>
+
+### 1. Redundant code for sorting files.
+📁 **File:** `github_app/github_helper/pull_requests.py:185`
+⚖️ **Severity:** 5/10
+🔍 **Description:** The custom sorting logic can be replaced with Python's built-in sorting functions for better readability and maintainability.
+💡 **Solution:** Use Python's `sorted` function with a key parameter.
+
+**Current Code:**
+```python
+sorted_files =[]
+for file in files:
+    min_index = len(sorted_files)
+    file_name = file["filename"]
+    for i, sorted_file in enumerate(sorted_files):
+        if file_name < sorted_file["filename"]:
+            min_index = i
+            break
+    sorted_files.insert(min_index, file)
+return sorted_files
+```
+
+**Suggested Code:**
+```python
+sorted_files = sorted(files, key=lambda x: x["filename"])
+return sorted_files
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-2024-05-13)
+{"prompt_tokens": 4007, "completion_tokens": 873, "total_tokens": 4880}
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_5/comments.json b/.experiments/code_review/gpt-4o/no_eval/pr_5/comments.json
new file mode 100644
index 00000000..b67c7510
--- /dev/null
+++ b/.experiments/code_review/gpt-4o/no_eval/pr_5/comments.json
@@ -0,0 +1,47 @@
+[
+  {
+    "topic": "API Call Error Handling",
+    "comment": "The API call to 'completion' lacks a retry mechanism.",
+    "confidence": "critical",
+    "reason": "API calls can fail due to network issues or server errors, and without a retry mechanism, the function may fail unexpectedly.",
+    "solution": "Implement a retry mechanism with exponential backoff for the API call.",
+    "actual_code": "response = completion(\n    model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n)",
+    "fixed_code": "import time\n\nfor attempt in range(3):\n    try:\n        response = completion(\n            model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n        )\n        break\n    except Exception as e:\n        if attempt < 2:\n            time.sleep(2 ** attempt)\n        else:\n            raise e",
+    "file_name": "main.py",
+    "start_line": 66,
+    "end_line": 68,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "Silent Failure",
+    "comment": "The exception handling for JSON decoding fails silently without logging.",
+    "confidence": "critical",
+    "reason": "Silent failures make it difficult to diagnose issues when they occur.",
+    "solution": "Add logging to capture the exception details.",
+    "actual_code": "except json.JSONDecodeError:\n    result ={",
+    "fixed_code": "except json.JSONDecodeError as e:\n    print(f\"Failed to parse content for applicant:{e}\")\n    result ={",
+    "file_name": "main.py",
+    "start_line": 82,
+    "end_line": 84,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Division by Zero",
+    "comment": "Potential division by zero when calculating total tokens.",
+    "confidence": "critical",
+    "reason": "If 'total_tokens' is zero, it will cause a division by zero error.",
+    "solution": "Add a check to ensure 'total_tokens' is not zero before performing the division.",
+    "actual_code": "total_tokens = total_input_tokens + total_output_tokens",
+    "fixed_code": "total_tokens = total_input_tokens + total_output_tokens\nif total_tokens == 0:\n    print(\"No tokens were used.\")\n    return",
+    "file_name": "main.py",
+    "start_line": 156,
+    "end_line": 158,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_5/issues.json b/.experiments/code_review/gpt-4o/no_eval/pr_5/issues.json
new file mode 100644
index 00000000..0847dab5
--- /dev/null
+++ b/.experiments/code_review/gpt-4o/no_eval/pr_5/issues.json
@@ -0,0 +1,107 @@
+[
+  {
+    "topic": "Unused Import",
+    "comment": "The import statement for the 'random' module is unnecessary.",
+    "confidence": "trivial",
+    "reason": "The 'random' module is imported but never used in the code.",
+    "solution": "Remove the import statement for 'random'.",
+    "actual_code": "import random  # Unused import",
+    "fixed_code": "",
+    "file_name": "main.py",
+    "start_line": 8,
+    "end_line": 8,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 1
+  },
+  {
+    "topic": "API Call Error Handling",
+    "comment": "The API call to 'completion' lacks a retry mechanism.",
+    "confidence": "critical",
+    "reason": "API calls can fail due to network issues or server errors, and without a retry mechanism, the function may fail unexpectedly.",
+    "solution": "Implement a retry mechanism with exponential backoff for the API call.",
+    "actual_code": "response = completion(\n    model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n)",
+    "fixed_code": "import time\n\nfor attempt in range(3):\n    try:\n        response = completion(\n            model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n        )\n        break\n    except Exception as e:\n        if attempt < 2:\n            time.sleep(2 ** attempt)\n        else:\n            raise e",
+    "file_name": "main.py",
+    "start_line": 66,
+    "end_line": 68,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "Silent Failure",
+    "comment": "The exception handling for JSON decoding fails silently without logging.",
+    "confidence": "critical",
+    "reason": "Silent failures make it difficult to diagnose issues when they occur.",
+    "solution": "Add logging to capture the exception details.",
+    "actual_code": "except json.JSONDecodeError:\n    result ={",
+    "fixed_code": "except json.JSONDecodeError as e:\n    print(f\"Failed to parse content for applicant:{e}\")\n    result ={",
+    "file_name": "main.py",
+    "start_line": 82,
+    "end_line": 84,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Inefficient Progress Printing",
+    "comment": "The progress printing method is inefficient.",
+    "confidence": "important",
+    "reason": "Printing progress in this manner can be slow and resource-intensive.",
+    "solution": "Use a more efficient method for printing progress, such as updating the progress less frequently.",
+    "actual_code": "print(f\"\\rProgress:[{'=' * int(50 * progress):<50}]{progress:.0%}\", end=\"\", flush=True)",
+    "fixed_code": "if index % 10 == 0 or index == total - 1:\n    print(f\"\\rProgress:[{'=' * int(50 * progress):<50}]{progress:.0%}\", end=\"\", flush=True)",
+    "file_name": "main.py",
+    "start_line": 121,
+    "end_line": 122,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Redundant Code",
+    "comment": "The check for an empty DataFrame is redundant.",
+    "confidence": "moderate",
+    "reason": "The code already handles an empty DataFrame gracefully, so this check is unnecessary.",
+    "solution": "Remove the redundant check for an empty DataFrame.",
+    "actual_code": "if len(df) == 0:\n    return",
+    "fixed_code": "",
+    "file_name": "main.py",
+    "start_line": 142,
+    "end_line": 143,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 3
+  },
+  {
+    "topic": "Division by Zero",
+    "comment": "Potential division by zero when calculating total tokens.",
+    "confidence": "critical",
+    "reason": "If 'total_tokens' is zero, it will cause a division by zero error.",
+    "solution": "Add a check to ensure 'total_tokens' is not zero before performing the division.",
+    "actual_code": "total_tokens = total_input_tokens + total_output_tokens",
+    "fixed_code": "total_tokens = total_input_tokens + total_output_tokens\nif total_tokens == 0:\n    print(\"No tokens were used.\")\n    return",
+    "file_name": "main.py",
+    "start_line": 156,
+    "end_line": 158,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "File Not Found Handling",
+    "comment": "No error handling for file not found.",
+    "confidence": "important",
+    "reason": "If the specified file does not exist, the program will crash.",
+    "solution": "Add error handling to check if the file exists before processing.",
+    "actual_code": "main(input_file)",
+    "fixed_code": "if not os.path.isfile(input_file):\n    print(f\"File not found:{input_file}\")\n    return\nmain(input_file)",
+    "file_name": "main.py",
+    "start_line": 174,
+    "end_line": 175,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_5/review.md b/.experiments/code_review/gpt-4o/no_eval/pr_5/review.md
new file mode 100644
index 00000000..f36368e6
--- /dev/null
+++ b/.experiments/code_review/gpt-4o/no_eval/pr_5/review.md
@@ -0,0 +1,182 @@
+PR URL: https://github.com/sauravpanda/applicant-screening/pull/5
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 7
+- Critical: 3
+- Important: 2
+- Minor: 1
+- Files Affected: 1
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>API Call Error Handling (3 issues)</strong></summary>
+
+### 1. The API call to 'completion' lacks a retry mechanism.
+📁 **File:** `main.py:66`
+⚖️ **Severity:** 9/10
+🔍 **Description:** API calls can fail due to network issues or server errors, and without a retry mechanism, the function may fail unexpectedly.
+💡 **Solution:** Implement a retry mechanism with exponential backoff for the API call.
+
+**Current Code:**
+```python
+response = completion(
+    model=os.environ.get("model", "anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1"), messages=messages
+)
+```
+
+**Suggested Code:**
+```python
+import time
+
+for attempt in range(3):
+    try:
+        response = completion(
+            model=os.environ.get("model", "anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1"), messages=messages
+        )
+        break
+    except Exception as e:
+        if attempt < 2:
+            time.sleep(2 ** attempt)
+        else:
+            raise e
+```
+
+### 2. The exception handling for JSON decoding fails silently without logging.
+📁 **File:** `main.py:82`
+⚖️ **Severity:** 8/10
+🔍 **Description:** Silent failures make it difficult to diagnose issues when they occur.
+💡 **Solution:** Add logging to capture the exception details.
+
+**Current Code:**
+```python
+except json.JSONDecodeError:
+    result ={
+```
+
+**Suggested Code:**
+```python
+except json.JSONDecodeError as e:
+    print(f"Failed to parse content for applicant:{e}")
+    result ={
+```
+
+### 3. Potential division by zero when calculating total tokens.
+📁 **File:** `main.py:156`
+⚖️ **Severity:** 7/10
+🔍 **Description:** If 'total_tokens' is zero, it will cause a division by zero error.
+💡 **Solution:** Add a check to ensure 'total_tokens' is not zero before performing the division.
+
+**Current Code:**
+```python
+total_tokens = total_input_tokens + total_output_tokens
+```
+
+**Suggested Code:**
+```python
+total_tokens = total_input_tokens + total_output_tokens
+if total_tokens == 0:
+    print("No tokens were used.")
+    return
+```
+
+</details>
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Inefficient Progress Printing (2 issues)</strong></summary>
+
+### 1. The progress printing method is inefficient.
+📁 **File:** `main.py:121`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Printing progress in this manner can be slow and resource-intensive.
+💡 **Solution:** Use a more efficient method for printing progress, such as updating the progress less frequently.
+
+**Current Code:**
+```python
+print(f"\rProgress:[{'=' * int(50 * progress):<50}]{progress:.0%}", end="", flush=True)
+```
+
+**Suggested Code:**
+```python
+if index % 10 == 0 or index == total - 1:
+    print(f"\rProgress:[{'=' * int(50 * progress):<50}]{progress:.0%}", end="", flush=True)
+```
+
+### 2. No error handling for file not found.
+📁 **File:** `main.py:174`
+⚖️ **Severity:** 6/10
+🔍 **Description:** If the specified file does not exist, the program will crash.
+💡 **Solution:** Add error handling to check if the file exists before processing.
+
+**Current Code:**
+```python
+main(input_file)
+```
+
+**Suggested Code:**
+```python
+if not os.path.isfile(input_file):
+    print(f"File not found:{input_file}")
+    return
+main(input_file)
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (2 issues)</strong></summary>
+
+<details>
+<summary><strong>Redundant Code (1 issues)</strong></summary>
+
+### 1. The check for an empty DataFrame is redundant.
+📁 **File:** `main.py:142`
+⚖️ **Severity:** 3/10
+🔍 **Description:** The code already handles an empty DataFrame gracefully, so this check is unnecessary.
+💡 **Solution:** Remove the redundant check for an empty DataFrame.
+
+**Current Code:**
+```python
+if len(df) == 0:
+    return
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-2024-05-13)
+{"prompt_tokens": 6154, "completion_tokens": 1315, "total_tokens": 7469}
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_222/comments.json b/.experiments/code_review/haiku/no_eval/pr_222/comments.json
new file mode 100644
index 00000000..7edfffbe
--- /dev/null
+++ b/.experiments/code_review/haiku/no_eval/pr_222/comments.json
@@ -0,0 +1,72 @@
+[
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Docker",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to Dockerfile, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_name": "Dockerfile",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Docker",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to docker-compose.yml, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "15",
+    "end_line": "15",
+    "side": "RIGHT",
+    "file_name": "docker-compose.yml",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Version Control",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to .gitignore, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "164",
+    "end_line": "164",
+    "side": "RIGHT",
+    "file_name": ".gitignore",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Database",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to db_setup/init.sql, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "db_setup/init.sql",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_222/issues.json b/.experiments/code_review/haiku/no_eval/pr_222/issues.json
new file mode 100644
index 00000000..b3bb8f78
--- /dev/null
+++ b/.experiments/code_review/haiku/no_eval/pr_222/issues.json
@@ -0,0 +1,312 @@
+[
+  {
+    "topic": "Dockerfile",
+    "comment": "The Dockerfile should install system dependencies before installing Poetry.",
+    "confidence": "important",
+    "reason": "Installing system dependencies before Poetry ensures that the necessary build tools are available for the Poetry installation process.",
+    "solution": "Move the system dependency installation block before the Poetry installation step.",
+    "actual_code": "",
+    "fixed_code": "# Install system dependencies\nRUN apt-get update && apt-get install -y \\\n    git \\\n    build-essential \\\n    && rm -rf /var/lib/apt/lists/*\n\n# Install Poetry\nRUN pip install --no-cache-dir poetry",
+    "file_name": "Dockerfile",
+    "start_line": 7,
+    "end_line": 11,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Dockerfile",
+    "comment": "The Dockerfile should make the Tree-sitter language installation script executable before running it.",
+    "confidence": "important",
+    "reason": "Making the script executable ensures that it can be properly executed during the build process.",
+    "solution": "Add a step to make the script executable before running it.",
+    "actual_code": "",
+    "fixed_code": "# Make the installation script executable\nRUN chmod +x install_tree_sitter_languages.sh\n\n# Run the Tree-sitter language installation script\nRUN ./install_tree_sitter_languages.sh",
+    "file_name": "Dockerfile",
+    "start_line": 25,
+    "end_line": 29,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "config.json",
+    "comment": "The config.json file should use environment variables for sensitive information like API keys.",
+    "confidence": "important",
+    "reason": "Using environment variables instead of hardcoding sensitive information in the config file improves security and makes the configuration more flexible.",
+    "solution": "Replace the API key and API base values with environment variable references, e.g., `os.environ['AZURE_API_KEY']`.",
+    "actual_code": "\"api_key\": \"azure/text-embedding-small\",\n\"api_base\": \"azure/gpt-4o-mini\"",
+    "fixed_code": "\"api_key\": \"os.environ['AZURE_API_KEY']\",\n\"api_base\": \"os.environ['AZURE_API_BASE']\"",
+    "file_name": "config.json",
+    "start_line": 13,
+    "end_line": 14,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "docker-compose.yml",
+    "comment": "The docker-compose.yml file should use the same Postgres image as the Dockerfile-postgres file.",
+    "confidence": "important",
+    "reason": "Using the same Postgres image ensures consistency and reduces potential issues with different versions or configurations.",
+    "solution": "Replace the Postgres image in the docker-compose.yml file with the one used in the Dockerfile-postgres file.",
+    "actual_code": "image: postgres:16-bullseye",
+    "fixed_code": "build:\n  context: .\n  dockerfile: Dockerfile-postgres",
+    "file_name": "docker-compose.yml",
+    "start_line": 18,
+    "end_line": 26,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "examples/ragify_codebase/main.py",
+    "comment": "The main.py file should provide more context and examples for using the RepositoryAnalyzer.",
+    "confidence": "moderate",
+    "reason": "Adding more context and examples will help users understand how to effectively use the RepositoryAnalyzer in their own projects.",
+    "solution": "Expand the example code to include more comments and explanations, such as how to set up the repository, how to perform different types of queries, and how to handle the results.",
+    "actual_code": "",
+    "fixed_code": "# Initialize the analyzer\nanalyzer = RepositoryAnalyzer()\n\n# Set up the repository (do this when you first analyze a repo or when you want to update it)\nanalyzer.setup_repository(\"./github_app/\")\n\n# Perform queries (you can do this as many times as you want without calling setup_repository again)\nresults = analyzer.query(\"Find functions that handle authentication\")\nfor result in results:\n    print(f\"File:{result['file_path']}\")\n    print(f\"Abstraction:{result['abstraction']}\")\n    print(f\"result:\\n{result}\")\n    print(f\"Relevance Score:{result['relevance_score']}\")\n    print(\"---\")\n\n# If you make changes to the repository and want to update the analysis:\nanalyzer.setup_repository(\"/path/to/your/repo\")\n\n# Then you can query again with the updated data\nresults = analyzer.query(\"authentication\")",
+    "file_name": "examples/ragify_codebase/main.py",
+    "start_line": 1,
+    "end_line": 22,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 5
+  },
+  {
+    "topic": "Normalization of query embedding",
+    "comment": "The query embedding is normalized correctly by dividing it by its L2 norm. This ensures that the query embedding has a unit length, which is important for cosine similarity calculations.",
+    "confidence": "positive",
+    "reason": "Normalizing the query embedding is a common best practice in vector similarity search to ensure that the magnitude of the vector does not affect the similarity calculation.",
+    "solution": "The current implementation of normalizing the query embedding is appropriate and does not require any changes.",
+    "actual_code": "query_embedding_normalized = query_embedding_np / np.linalg.norm(query_embedding_np)",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 15,
+    "end_line": 16,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 5
+  },
+  {
+    "topic": "SQL query construction",
+    "comment": "The SQL query is well-constructed and includes the necessary filters and ordering to retrieve the top-k most similar results based on the cosine similarity.",
+    "confidence": "positive",
+    "reason": "The query includes a join between the `embeddings` table and the `function_abstractions` and `files` tables to filter the results by the repository ID. The `ORDER BY` and `LIMIT` clauses ensure that only the top-k most similar results are returned.",
+    "solution": "The current implementation of the SQL query is appropriate and does not require any changes.",
+    "actual_code": "```sql\nSELECT \n    e.node_id,\n    e.text,\n    e.metadata,\n    1 - (e.embedding <=> %s::vector) as similarity\nFROM \n{self.table_name}e\nJOIN \n    function_abstractions fa ON e.node_id = fa.function_id::text\nJOIN \n    files f ON fa.file_id = f.file_id\nWHERE \n    f.repo_id = %s\nORDER BY \n    similarity DESC\nLIMIT \n    %s\n```",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 19,
+    "end_line": 36,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 5
+  },
+  {
+    "topic": "Metadata handling",
+    "comment": "The code correctly handles the metadata column, converting it to a dictionary if it is not already in that format.",
+    "confidence": "positive",
+    "reason": "The `row[2]` value is checked to see if it is already a dictionary, and if not, it is converted to a dictionary using the `Json` class from `psycopg2.extras`.",
+    "solution": "The current implementation of metadata handling is appropriate and does not require any changes.",
+    "actual_code": "\"metadata\": row[2] if isinstance(row[2], dict) else Json(row[2]),",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 48,
+    "end_line": 48,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Feedback system implementation",
+    "comment": "The `AbstractionFeedback` class provides a simple and effective way to store and retrieve feedback for code abstractions.",
+    "confidence": "positive",
+    "reason": "The class uses a dictionary to store the feedback, with the code ID as the key and the feedback details as the value. The `add_feedback` and `get_feedback` methods provide a clear interface for managing the feedback.",
+    "solution": "The current implementation of the `AbstractionFeedback` class is appropriate and does not require any changes.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/feedback_system.py",
+    "start_line": 0,
+    "end_line": 0,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Unused Imports",
+    "comment": "The following imports are not used in the code and can be removed: `from llama_index.core.schema import TextNode`, `from concurrent.futures import ThreadPoolExecutor, as_completed`, `from llama_index.embeddings.litellm import LiteLLMEmbedding`, `from llama_index.core import QueryBundle`.",
+    "confidence": "moderate",
+    "reason": "Unused imports can clutter the codebase and make it harder to maintain.",
+    "solution": "Remove the unused imports to improve code readability and maintainability.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 7,
+    "end_line": 19,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 3
+  },
+  {
+    "topic": "Logging Configuration",
+    "comment": "The logging configuration is set up in the global scope, which can lead to issues if the module is imported in multiple places. Consider moving the logging setup to a function or class initialization to ensure it's only configured once.",
+    "confidence": "moderate",
+    "reason": "Global logging configuration can cause conflicts if the module is used in multiple places.",
+    "solution": "Move the logging setup to a function or class initialization to ensure it's only configured once.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 22,
+    "end_line": 26,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 4
+  },
+  {
+    "topic": "Tokenizer Initialization",
+    "comment": "The tokenizer is initialized in the global scope, which can lead to issues if the module is imported in multiple places. Consider moving the tokenizer initialization to a function or class initialization to ensure it's only initialized once.",
+    "confidence": "moderate",
+    "reason": "Global tokenizer initialization can cause conflicts if the module is used in multiple places.",
+    "solution": "Move the tokenizer initialization to a function or class initialization to ensure it's only initialized once.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 28,
+    "end_line": 29,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 4
+  },
+  {
+    "topic": "Unused Code",
+    "comment": "The commented-out code block starting from line 315 appears to be an unused query method. Consider removing this code if it's no longer needed.",
+    "confidence": "important",
+    "reason": "Unused code can make the codebase harder to maintain and understand.",
+    "solution": "Remove the commented-out code block if it's no longer needed.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 315,
+    "end_line": 335,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 5
+  },
+  {
+    "topic": "Potential Performance Optimization",
+    "comment": "The `query` method retrieves the function details from the database for each result node. Consider optimizing this by fetching all the required information in a single database query.",
+    "confidence": "important",
+    "reason": "Fetching data from the database for each result node can be inefficient, especially for larger result sets.",
+    "solution": "Modify the `query` method to fetch all the required information in a single database query to improve performance.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 337,
+    "end_line": 390,
+    "side": "LEFT",
+    "sentiment": "positive",
+    "severity_level": 6
+  },
+  {
+    "topic": "Dependencies",
+    "comment": "The project dependencies have been updated to use newer versions of some libraries, such as Python 3.9 and various tree-sitter language parsers.",
+    "confidence": "important",
+    "reason": "Keeping dependencies up-to-date is important for security, performance, and access to new features.",
+    "solution": "The changes look good and should help improve the project's overall maintainability.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "pyproject.toml",
+    "start_line": 3,
+    "end_line": 13,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 7
+  },
+  {
+    "topic": "Code Chunking",
+    "comment": "The new `chunk_code` function in `test_chunker.py` looks like a useful utility for testing the code chunking functionality.",
+    "confidence": "moderate",
+    "reason": "The function provides a clear way to test the code chunking behavior for different programming languages.",
+    "solution": "Consider adding more test cases to cover edge cases and ensure the chunking works as expected for a variety of code samples.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "tests/retriever/test_chunker.py",
+    "start_line": 1,
+    "end_line": 101,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 6
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Docker",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to Dockerfile, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_name": "Dockerfile",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Docker",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to docker-compose.yml, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "15",
+    "end_line": "15",
+    "side": "RIGHT",
+    "file_name": "docker-compose.yml",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Version Control",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to .gitignore, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "164",
+    "end_line": "164",
+    "side": "RIGHT",
+    "file_name": ".gitignore",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Database",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to db_setup/init.sql, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "db_setup/init.sql",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_222/review.md b/.experiments/code_review/haiku/no_eval/pr_222/review.md
new file mode 100644
index 00000000..de7eb626
--- /dev/null
+++ b/.experiments/code_review/haiku/no_eval/pr_222/review.md
@@ -0,0 +1,247 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/222
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 21
+- Critical: 5
+- Important: 7
+- Minor: 5
+- Files Affected: 11
+## 🏆 Code Quality
+[██████████████████░░] 90% (Excellent)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Configuration (5 issues)</strong></summary>
+
+### 1. Changes made to sensitive file
+📁 **File:** `config.json:4`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to config.json, which needs review
+💡 **Solution:** NA
+
+### 2. Changes made to sensitive file
+📁 **File:** `Dockerfile:4`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to Dockerfile, which needs review
+💡 **Solution:** NA
+
+### 3. Changes made to sensitive file
+📁 **File:** `docker-compose.yml:15`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to docker-compose.yml, which needs review
+💡 **Solution:** NA
+
+### 4. Changes made to sensitive file
+📁 **File:** `.gitignore:164`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to .gitignore, which needs review
+💡 **Solution:** NA
+
+### 5. Changes made to sensitive file
+📁 **File:** `db_setup/init.sql:1`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to db_setup/init.sql, which needs review
+💡 **Solution:** NA
+
+</details>
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Dockerfile (7 issues)</strong></summary>
+
+### 1. The Dockerfile should install system dependencies before installing Poetry.
+📁 **File:** `Dockerfile:7`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Installing system dependencies before Poetry ensures that the necessary build tools are available for the Poetry installation process.
+💡 **Solution:** Move the system dependency installation block before the Poetry installation step.
+
+**Current Code:**
+```python
+
+```
+
+**Suggested Code:**
+```python
+# Install system dependencies
+RUN apt-get update && apt-get install -y \
+    git \
+    build-essential \
+    && rm -rf /var/lib/apt/lists/*
+
+# Install Poetry
+RUN pip install --no-cache-dir poetry
+```
+
+### 2. The Dockerfile should make the Tree-sitter language installation script executable before running it.
+📁 **File:** `Dockerfile:25`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Making the script executable ensures that it can be properly executed during the build process.
+💡 **Solution:** Add a step to make the script executable before running it.
+
+**Current Code:**
+```python
+
+```
+
+**Suggested Code:**
+```python
+# Make the installation script executable
+RUN chmod +x install_tree_sitter_languages.sh
+
+# Run the Tree-sitter language installation script
+RUN ./install_tree_sitter_languages.sh
+```
+
+### 3. The config.json file should use environment variables for sensitive information like API keys.
+📁 **File:** `config.json:13`
+⚖️ **Severity:** 8/10
+🔍 **Description:** Using environment variables instead of hardcoding sensitive information in the config file improves security and makes the configuration more flexible.
+💡 **Solution:** Replace the API key and API base values with environment variable references, e.g., `os.environ['AZURE_API_KEY']`.
+
+**Current Code:**
+```python
+"api_key": "azure/text-embedding-small",
+"api_base": "azure/gpt-4o-mini"
+```
+
+**Suggested Code:**
+```python
+"api_key": "os.environ['AZURE_API_KEY']",
+"api_base": "os.environ['AZURE_API_BASE']"
+```
+
+### 4. The docker-compose.yml file should use the same Postgres image as the Dockerfile-postgres file.
+📁 **File:** `docker-compose.yml:18`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Using the same Postgres image ensures consistency and reduces potential issues with different versions or configurations.
+💡 **Solution:** Replace the Postgres image in the docker-compose.yml file with the one used in the Dockerfile-postgres file.
+
+**Current Code:**
+```python
+image: postgres:16-bullseye
+```
+
+**Suggested Code:**
+```python
+build:
+  context: .
+  dockerfile: Dockerfile-postgres
+```
+
+### 5. The commented-out code block starting from line 315 appears to be an unused query method. Consider removing this code if it's no longer needed.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:315`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Unused code can make the codebase harder to maintain and understand.
+💡 **Solution:** Remove the commented-out code block if it's no longer needed.
+
+### 6. The `query` method retrieves the function details from the database for each result node. Consider optimizing this by fetching all the required information in a single database query.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:337`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Fetching data from the database for each result node can be inefficient, especially for larger result sets.
+💡 **Solution:** Modify the `query` method to fetch all the required information in a single database query to improve performance.
+
+### 7. The project dependencies have been updated to use newer versions of some libraries, such as Python 3.9 and various tree-sitter language parsers.
+📁 **File:** `pyproject.toml:3`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Keeping dependencies up-to-date is important for security, performance, and access to new features.
+💡 **Solution:** The changes look good and should help improve the project's overall maintainability.
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (5 issues)</strong></summary>
+
+<details>
+<summary><strong>examples/ragify_codebase/main.py (5 issues)</strong></summary>
+
+### 1. The main.py file should provide more context and examples for using the RepositoryAnalyzer.
+📁 **File:** `examples/ragify_codebase/main.py:1`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Adding more context and examples will help users understand how to effectively use the RepositoryAnalyzer in their own projects.
+💡 **Solution:** Expand the example code to include more comments and explanations, such as how to set up the repository, how to perform different types of queries, and how to handle the results.
+
+**Current Code:**
+```python
+
+```
+
+**Suggested Code:**
+```python
+# Initialize the analyzer
+analyzer = RepositoryAnalyzer()
+
+# Set up the repository (do this when you first analyze a repo or when you want to update it)
+analyzer.setup_repository("./github_app/")
+
+# Perform queries (you can do this as many times as you want without calling setup_repository again)
+results = analyzer.query("Find functions that handle authentication")
+for result in results:
+    print(f"File:{result['file_path']}")
+    print(f"Abstraction:{result['abstraction']}")
+    print(f"result:\n{result}")
+    print(f"Relevance Score:{result['relevance_score']}")
+    print("---")
+
+# If you make changes to the repository and want to update the analysis:
+analyzer.setup_repository("/path/to/your/repo")
+
+# Then you can query again with the updated data
+results = analyzer.query("authentication")
+```
+
+### 2. The following imports are not used in the code and can be removed: `from llama_index.core.schema import TextNode`, `from concurrent.futures import ThreadPoolExecutor, as_completed`, `from llama_index.embeddings.litellm import LiteLLMEmbedding`, `from llama_index.core import QueryBundle`.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:7`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Unused imports can clutter the codebase and make it harder to maintain.
+💡 **Solution:** Remove the unused imports to improve code readability and maintainability.
+
+### 3. The logging configuration is set up in the global scope, which can lead to issues if the module is imported in multiple places. Consider moving the logging setup to a function or class initialization to ensure it's only configured once.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:22`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Global logging configuration can cause conflicts if the module is used in multiple places.
+💡 **Solution:** Move the logging setup to a function or class initialization to ensure it's only configured once.
+
+### 4. The tokenizer is initialized in the global scope, which can lead to issues if the module is imported in multiple places. Consider moving the tokenizer initialization to a function or class initialization to ensure it's only initialized once.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:28`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Global tokenizer initialization can cause conflicts if the module is used in multiple places.
+💡 **Solution:** Move the tokenizer initialization to a function or class initialization to ensure it's only initialized once.
+
+### 5. The new `chunk_code` function in `test_chunker.py` looks like a useful utility for testing the code chunking functionality.
+📁 **File:** `tests/retriever/test_chunker.py:1`
+⚖️ **Severity:** 6/10
+🔍 **Description:** The function provides a clear way to test the code chunking behavior for different programming languages.
+💡 **Solution:** Consider adding more test cases to cover edge cases and ensure the chunking works as expected for a variety of code samples.
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (anthropic.claude-3-haiku-20240307-v1:0)
+{"prompt_tokens": 24844, "completion_tokens": 4069, "total_tokens": 28913}
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_252/comments.json b/.experiments/code_review/haiku/no_eval/pr_252/comments.json
new file mode 100644
index 00000000..0637a088
--- /dev/null
+++ b/.experiments/code_review/haiku/no_eval/pr_252/comments.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_252/issues.json b/.experiments/code_review/haiku/no_eval/pr_252/issues.json
new file mode 100644
index 00000000..38344bd0
--- /dev/null
+++ b/.experiments/code_review/haiku/no_eval/pr_252/issues.json
@@ -0,0 +1,32 @@
+[
+  {
+    "topic": "Linkedin Post Generation",
+    "comment": "The LinkedIn post generation code is not formatted correctly.",
+    "confidence": "moderate",
+    "reason": "The LinkedIn post generation code is spread across multiple lines, making it less readable and maintainable.",
+    "solution": "Condense the LinkedIn post generation code into a single line, similar to the Twitter post generation.",
+    "actual_code": "linkedin_post = work_summary_generator.generate_linkedin_post(\n    summary, user=\"oss_example\"\n)",
+    "fixed_code": "linkedin_post = work_summary_generator.generate_linkedin_post(summary, user=\"oss_example\")",
+    "file_name": "examples/work_summarizer/main.py",
+    "start_line": 60,
+    "end_line": 62,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 3
+  },
+  {
+    "topic": "Prompt Formatting",
+    "comment": "The TWITTER_POST_PROMPT and LINKEDIN_POST_PROMPT could be improved for better readability.",
+    "confidence": "moderate",
+    "reason": "The prompts are currently formatted as a single long string, making them difficult to read and maintain.",
+    "solution": "Consider breaking the prompts into multiple lines, using string formatting, and adding comments to explain the different sections.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/llms/prompts/work_summary_prompts.py",
+    "start_line": 44,
+    "end_line": 65,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_252/review.md b/.experiments/code_review/haiku/no_eval/pr_252/review.md
new file mode 100644
index 00000000..e28ec0f5
--- /dev/null
+++ b/.experiments/code_review/haiku/no_eval/pr_252/review.md
@@ -0,0 +1,70 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/252
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 2
+- Critical: 0
+- Important: 0
+- Minor: 2
+- Files Affected: 2
+## 🏆 Code Quality
+[██████████████████░░] 90% (Excellent)
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (2 issues)</strong></summary>
+
+<details>
+<summary><strong>Linkedin Post Generation (2 issues)</strong></summary>
+
+### 1. The LinkedIn post generation code is not formatted correctly.
+📁 **File:** `examples/work_summarizer/main.py:60`
+⚖️ **Severity:** 3/10
+🔍 **Description:** The LinkedIn post generation code is spread across multiple lines, making it less readable and maintainable.
+💡 **Solution:** Condense the LinkedIn post generation code into a single line, similar to the Twitter post generation.
+
+**Current Code:**
+```python
+linkedin_post = work_summary_generator.generate_linkedin_post(
+    summary, user="oss_example"
+)
+```
+
+**Suggested Code:**
+```python
+linkedin_post = work_summary_generator.generate_linkedin_post(summary, user="oss_example")
+```
+
+### 2. The TWITTER_POST_PROMPT and LINKEDIN_POST_PROMPT could be improved for better readability.
+📁 **File:** `kaizen/llms/prompts/work_summary_prompts.py:44`
+⚖️ **Severity:** 4/10
+🔍 **Description:** The prompts are currently formatted as a single long string, making them difficult to read and maintain.
+💡 **Solution:** Consider breaking the prompts into multiple lines, using string formatting, and adding comments to explain the different sections.
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (anthropic.claude-3-haiku-20240307-v1:0)
+{"prompt_tokens": 4436, "completion_tokens": 465, "total_tokens": 4901}
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_335/comments.json b/.experiments/code_review/haiku/no_eval/pr_335/comments.json
new file mode 100644
index 00000000..0637a088
--- /dev/null
+++ b/.experiments/code_review/haiku/no_eval/pr_335/comments.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_335/issues.json b/.experiments/code_review/haiku/no_eval/pr_335/issues.json
new file mode 100644
index 00000000..d941a81f
--- /dev/null
+++ b/.experiments/code_review/haiku/no_eval/pr_335/issues.json
@@ -0,0 +1,62 @@
+[
+  {
+    "topic": "Prompt Changes",
+    "comment": "The `PR_DESCRIPTION_PROMPT` and `PR_FILE_DESCRIPTION_PROMPT` have been updated to use a more concise and structured format for the output.",
+    "confidence": "important",
+    "reason": "The new prompts provide a clearer and more organized structure for the generated pull request description, making it easier to read and understand.",
+    "solution": "The changes look good and should improve the overall quality and readability of the generated pull request descriptions.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/llms/prompts/pr_desc_prompts.py",
+    "start_line": 1,
+    "end_line": 92,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 7
+  },
+  {
+    "topic": "Removal of `PR_DESC_EVALUATION_PROMPT`",
+    "comment": "The `PR_DESC_EVALUATION_PROMPT` has been removed from the `code_review_prompts.py` file.",
+    "confidence": "moderate",
+    "reason": "The `PR_DESC_EVALUATION_PROMPT` was previously used for re-evaluating the generated pull request description, but it has been removed in the current changes.",
+    "solution": "Ensure that the functionality for re-evaluating the pull request description is still maintained, either through a different mechanism or by incorporating the evaluation logic directly into the main description generation process.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/llms/prompts/code_review_prompts.py",
+    "start_line": 190,
+    "end_line": 201,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Removal of `reeval_response` parameter",
+    "comment": "The `reeval_response` parameter has been removed from several method calls in the `pr_description.py` file.",
+    "confidence": "moderate",
+    "reason": "The `reeval_response` parameter was previously used for re-evaluating the generated pull request description, but it has been removed in the current changes.",
+    "solution": "Ensure that the functionality for re-evaluating the pull request description is still maintained, either through a different mechanism or by incorporating the evaluation logic directly into the main description generation process.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 43,
+    "end_line": 96,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Removal of `CODE_REVIEW_SYSTEM_PROMPT`",
+    "comment": "The `CODE_REVIEW_SYSTEM_PROMPT` has been removed and replaced with `PR_DESCRIPTION_SYSTEM_PROMPT` in the `pr_description.py` file.",
+    "confidence": "important",
+    "reason": "The `CODE_REVIEW_SYSTEM_PROMPT` was previously used as the system prompt for the code review process, but it has been replaced with a new prompt specifically for generating pull request descriptions.",
+    "solution": "The changes look good and should help align the system prompt with the updated pull request description generation process.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 29,
+    "end_line": 29,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 7
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_335/review.md b/.experiments/code_review/haiku/no_eval/pr_335/review.md
new file mode 100644
index 00000000..c4f147c7
--- /dev/null
+++ b/.experiments/code_review/haiku/no_eval/pr_335/review.md
@@ -0,0 +1,78 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/335
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 4
+- Critical: 0
+- Important: 2
+- Minor: 2
+- Files Affected: 3
+## 🏆 Code Quality
+[██████████████████░░] 90% (Excellent)
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Prompt Changes (2 issues)</strong></summary>
+
+### 1. The `PR_DESCRIPTION_PROMPT` and `PR_FILE_DESCRIPTION_PROMPT` have been updated to use a more concise and structured format for the output.
+📁 **File:** `kaizen/llms/prompts/pr_desc_prompts.py:1`
+⚖️ **Severity:** 7/10
+🔍 **Description:** The new prompts provide a clearer and more organized structure for the generated pull request description, making it easier to read and understand.
+💡 **Solution:** The changes look good and should improve the overall quality and readability of the generated pull request descriptions.
+
+### 2. The `CODE_REVIEW_SYSTEM_PROMPT` has been removed and replaced with `PR_DESCRIPTION_SYSTEM_PROMPT` in the `pr_description.py` file.
+📁 **File:** `kaizen/generator/pr_description.py:29`
+⚖️ **Severity:** 7/10
+🔍 **Description:** The `CODE_REVIEW_SYSTEM_PROMPT` was previously used as the system prompt for the code review process, but it has been replaced with a new prompt specifically for generating pull request descriptions.
+💡 **Solution:** The changes look good and should help align the system prompt with the updated pull request description generation process.
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (2 issues)</strong></summary>
+
+<details>
+<summary><strong>Removal of `PR_DESC_EVALUATION_PROMPT` (2 issues)</strong></summary>
+
+### 1. The `PR_DESC_EVALUATION_PROMPT` has been removed from the `code_review_prompts.py` file.
+📁 **File:** `kaizen/llms/prompts/code_review_prompts.py:190`
+⚖️ **Severity:** 5/10
+🔍 **Description:** The `PR_DESC_EVALUATION_PROMPT` was previously used for re-evaluating the generated pull request description, but it has been removed in the current changes.
+💡 **Solution:** Ensure that the functionality for re-evaluating the pull request description is still maintained, either through a different mechanism or by incorporating the evaluation logic directly into the main description generation process.
+
+### 2. The `reeval_response` parameter has been removed from several method calls in the `pr_description.py` file.
+📁 **File:** `kaizen/generator/pr_description.py:43`
+⚖️ **Severity:** 5/10
+🔍 **Description:** The `reeval_response` parameter was previously used for re-evaluating the generated pull request description, but it has been removed in the current changes.
+💡 **Solution:** Ensure that the functionality for re-evaluating the pull request description is still maintained, either through a different mechanism or by incorporating the evaluation logic directly into the main description generation process.
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (anthropic.claude-3-haiku-20240307-v1:0)
+{"prompt_tokens": 7614, "completion_tokens": 945, "total_tokens": 8559}
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_400/comments.json b/.experiments/code_review/haiku/no_eval/pr_400/comments.json
new file mode 100644
index 00000000..33197e40
--- /dev/null
+++ b/.experiments/code_review/haiku/no_eval/pr_400/comments.json
@@ -0,0 +1,16 @@
+[
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "11",
+    "end_line": "11",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_400/issues.json b/.experiments/code_review/haiku/no_eval/pr_400/issues.json
new file mode 100644
index 00000000..03fa0f89
--- /dev/null
+++ b/.experiments/code_review/haiku/no_eval/pr_400/issues.json
@@ -0,0 +1,406 @@
+[
+  {
+    "topic": "Boundary Conditions",
+    "comment": "The test cases for boundary conditions (very long descriptions) look good. The execution time is also printed, which is a nice addition.",
+    "confidence": "important",
+    "reason": "Handling large inputs is an important aspect of the function's robustness.",
+    "solution": "No changes needed.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
+    "start_line": 45,
+    "end_line": 61,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 8
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "The error handling tests cover various invalid input scenarios, which is good.",
+    "confidence": "important",
+    "reason": "Proper error handling is crucial for the function's reliability.",
+    "solution": "No changes needed.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
+    "start_line": 31,
+    "end_line": 43,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 8
+  },
+  {
+    "topic": "Collapsible Template",
+    "comment": "The collapsible template for the original description has been improved to include newlines for better readability.",
+    "confidence": "moderate",
+    "reason": "The previous template did not have proper newline formatting.",
+    "solution": "No changes needed.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
+    "start_line": 5,
+    "end_line": 6,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 5
+  },
+  {
+    "topic": "Readability",
+    "comment": "The updated PR_COLLAPSIBLE_TEMPLATE is more readable and follows better formatting.",
+    "confidence": "high",
+    "reason": "The previous template used a mix of string concatenation and formatting, which made it less readable and maintainable. The new template uses a multi-line string with proper indentation and formatting.",
+    "solution": "Keep the updated PR_COLLAPSIBLE_TEMPLATE, as it improves the overall readability of the code.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 4,
+    "end_line": 16,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Test Coverage",
+    "comment": "The new test cases cover a wider range of scenarios, including empty topics, single topic with single review, and multiple topics with multiple reviews.",
+    "confidence": "high",
+    "reason": "The additional test cases ensure the `create_pr_review_text` function handles different input scenarios correctly.",
+    "solution": "Keep the new test cases, as they improve the overall test coverage and ensure the function's robustness.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 74,
+    "end_line": 270,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "The new test cases cover scenarios where the review data is missing required fields, such as 'solution', 'reason', 'confidence', and 'severity_level'.",
+    "confidence": "high",
+    "reason": "Handling missing fields is important to ensure the function can gracefully handle incomplete review data.",
+    "solution": "Keep the new test cases, as they ensure the function can handle missing fields in the review data without crashing or producing unexpected output.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 142,
+    "end_line": 234,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Maintainability",
+    "comment": "The new test case for 'reviews_with_missing_comment' ensures the function can handle missing 'comment' field in the review data.",
+    "confidence": "high",
+    "reason": "Handling missing fields, such as 'comment', is important for the function's robustness and maintainability.",
+    "solution": "Keep the new test case, as it ensures the function can handle missing 'comment' field without crashing or producing unexpected output.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 238,
+    "end_line": 269,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Efficiency",
+    "comment": "The new test case for 'empty_list_in_topics' ensures the function can handle an empty list of reviews for a given topic.",
+    "confidence": "high",
+    "reason": "Handling empty lists of reviews is important for the function's efficiency and edge case handling.",
+    "solution": "Keep the new test case, as it ensures the function can handle empty lists of reviews without producing unexpected output.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 271,
+    "end_line": 276,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Unused imports",
+    "comment": "The imports `mock` and `pytest` are not used in the updated tests.",
+    "confidence": "moderate",
+    "reason": "Unused imports can make the code harder to read and maintain.",
+    "solution": "Remove the unused imports.",
+    "actual_code": "import os\nimport pytest\nfrom unittest import mock\nfrom kaizen.helpers.output import get_parent_folder",
+    "fixed_code": "import os\nfrom kaizen.helpers.output import get_parent_folder",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py",
+    "start_line": 1,
+    "end_line": 6,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 3
+  },
+  {
+    "topic": "Unnecessary mocking",
+    "comment": "The `test_get_parent_folder_normal()` function does not need to mock `os.getcwd()` as the actual implementation can be used.",
+    "confidence": "moderate",
+    "reason": "Mocking should be used only when necessary, as it can make the tests more complex and harder to maintain.",
+    "solution": "Remove the mocking in `test_get_parent_folder_normal()` and use the actual implementation of `get_parent_folder()`.",
+    "actual_code": "    with mock.patch('os.getcwd', return_value='/home/user/project'):\n        expected = '/home/user/project'\n        result = get_parent_folder()\n        assert result == expected, f\"Expected{expected}, but got{result}\"",
+    "fixed_code": "    expected = os.path.dirname(os.getcwd())\n    result = get_parent_folder()\n    assert result == expected, f\"Expected{expected}, but got{result}\"",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py",
+    "start_line": 13,
+    "end_line": 16,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 4
+  },
+  {
+    "topic": "Comprehensive error handling",
+    "comment": "The updated `test_get_parent_folder_error_handling()` function covers more error scenarios, including a generic `Exception` case.",
+    "confidence": "positive",
+    "reason": "Thorough error handling is important to ensure the function behaves correctly in various exceptional situations.",
+    "solution": "No changes needed, the updated error handling tests are comprehensive.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py",
+    "start_line": 20,
+    "end_line": 26,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 5
+  },
+  {
+    "topic": "Nested directory structure test",
+    "comment": "The updated `test_get_parent_folder_nested()` function tests the behavior of `get_parent_folder()` in a nested directory structure.",
+    "confidence": "positive",
+    "reason": "Testing the function in a nested directory structure is important to ensure it works correctly in different scenarios.",
+    "solution": "No changes needed, the nested directory structure test is a good addition.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py",
+    "start_line": 28,
+    "end_line": 33,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 5
+  },
+  {
+    "topic": "Boundary condition test",
+    "comment": "The previous `test_get_parent_folder_boundary_condition_long_path()` function testing a long path has been removed, which is a reasonable decision as the current implementation of `get_parent_folder()` should handle long paths without issues.",
+    "confidence": "positive",
+    "reason": "The previous test case was not necessary as the current implementation should handle long paths correctly.",
+    "solution": "No changes needed, the removal of the unnecessary boundary condition test is appropriate.",
+    "actual_code": "def test_get_parent_folder_boundary_condition_long_path():\n    long_path = \"/\" + \"a\" * 255\n    with mock.patch(\"os.getcwd\", return_value=long_path):\n        assert get_parent_folder() == long_path",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py",
+    "start_line": 29,
+    "end_line": 32,
+    "side": "LEFT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Unused imports",
+    "comment": "The imports `asyncio` and `nest_asyncio` are not used in the original test cases. Consider removing them if they are not required.",
+    "confidence": "moderate",
+    "reason": "Unused imports can clutter the code and make it less readable.",
+    "solution": "Remove the unused imports `asyncio` and `nest_asyncio` from the test file.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 4,
+    "end_line": 5,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 3
+  },
+  {
+    "topic": "Parameterized test",
+    "comment": "The new test `test_get_web_html_normal_cases` uses a parameterized approach, which is a good practice. It covers multiple test cases in a concise manner.",
+    "confidence": "positive",
+    "reason": "Parameterized tests improve code readability and maintainability by reducing duplication.",
+    "solution": "Keep the parameterized test approach.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 20,
+    "end_line": 75,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 5
+  },
+  {
+    "topic": "Handling empty HTML content",
+    "comment": "The new test `test_get_web_html_empty_html` ensures that the function handles empty HTML content correctly.",
+    "confidence": "positive",
+    "reason": "Handling edge cases like empty input is important for robust error handling.",
+    "solution": "Keep the test case for handling empty HTML content.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 85,
+    "end_line": 92,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Handling network errors",
+    "comment": "The new test `test_get_web_html_invalid_url` ensures that the function handles network errors correctly.",
+    "confidence": "positive",
+    "reason": "Handling network errors is important for a robust web scraping implementation.",
+    "solution": "Keep the test case for handling network errors.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 85,
+    "end_line": 91,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Handling large HTML content",
+    "comment": "The new test `test_get_web_html_large_content` ensures that the function can handle large HTML content without performance issues.",
+    "confidence": "positive",
+    "reason": "Testing the function's ability to handle large inputs is important for ensuring its scalability.",
+    "solution": "Keep the test case for handling large HTML content.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 93,
+    "end_line": 97,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Handling invalid HTML",
+    "comment": "The previous test `test_get_web_html_invalid_html` has been removed. Consider adding a new test case to ensure the function can handle invalid HTML content gracefully.",
+    "confidence": "moderate",
+    "reason": "Handling invalid HTML is important for a robust web scraping implementation.",
+    "solution": "Add a new test case to ensure the function can handle invalid HTML content without raising unexpected exceptions.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 98,
+    "end_line": 107,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 4
+  },
+  {
+    "topic": "Enabling critique mode",
+    "comment": "The `generate_tests` function call in the `examples/unittest/main.py` file has been updated to enable the critique mode and verbose output.",
+    "confidence": "positive",
+    "reason": "Enabling the critique mode and verbose output can provide more detailed feedback and insights during the testing process.",
+    "solution": "Keep the changes to enable critique mode and verbose output in the `generate_tests` function call.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "examples/unittest/main.py",
+    "start_line": 35,
+    "end_line": 37,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Refactoring",
+    "comment": "The `generate_tests` method has become quite long and complex. Consider breaking it down into smaller, more focused methods to improve readability and maintainability.",
+    "confidence": "important",
+    "reason": "Large methods can be difficult to understand and maintain, especially as the codebase grows.",
+    "solution": "Refactor the `generate_tests` method by extracting smaller, more focused methods for specific tasks, such as preparing the test file path, generating the AI tests, and writing the test file.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/unit_test.py",
+    "start_line": 0,
+    "end_line": 0,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Logging",
+    "comment": "The logging implementation could be improved by using a more structured approach, such as using the built-in `logging` module with appropriate log levels.",
+    "confidence": "moderate",
+    "reason": "The current logging implementation uses print statements, which can be less flexible and harder to manage than a structured logging approach.",
+    "solution": "Refactor the logging implementation to use the `logging` module, with appropriate log levels (e.g., DEBUG, INFO, WARNING, ERROR) and log messages that provide more context and details.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/unit_test.py",
+    "start_line": 0,
+    "end_line": 0,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 5
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "The `generate_tests_from_dir` method could benefit from more robust error handling, such as catching and handling specific exceptions.",
+    "confidence": "moderate",
+    "reason": "Catching and handling specific exceptions can help provide more informative error messages and improve the overall robustness of the application.",
+    "solution": "Modify the `generate_tests_from_dir` method to catch and handle specific exceptions, such as `FileNotFoundError` or `ValueError`, and provide more detailed error messages to help with debugging and troubleshooting.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/unit_test.py",
+    "start_line": 0,
+    "end_line": 0,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Separation of Concerns",
+    "comment": "The `UnitTestGenerator` class is responsible for both generating and running the tests. Consider separating these concerns into two different classes or modules.",
+    "confidence": "important",
+    "reason": "Separating the concerns of test generation and test execution can improve the overall design and maintainability of the codebase.",
+    "solution": "Create a separate `UnitTestRunner` class or module that is responsible for discovering and running the generated tests, while the `UnitTestGenerator` class focuses solely on generating the tests.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/unit_test.py",
+    "start_line": 0,
+    "end_line": 0,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Dependency Injection",
+    "comment": "The `UnitTestGenerator` class has several dependencies, such as the `LLMProvider` and the various prompt templates. Consider using dependency injection to improve the testability and flexibility of the class.",
+    "confidence": "moderate",
+    "reason": "Dependency injection can make the code more modular and easier to test, as it allows for easier substitution of dependencies.",
+    "solution": "Refactor the `UnitTestGenerator` class to accept its dependencies (e.g., `LLMProvider`, prompt templates) as constructor arguments, rather than creating them internally. This will improve the testability and flexibility of the class.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/unit_test.py",
+    "start_line": 0,
+    "end_line": 0,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Logging Configuration",
+    "comment": "The code sets all loggers to the ERROR level, which may be too restrictive. Consider providing more granular control over log levels.",
+    "confidence": "moderate",
+    "reason": "Setting all loggers to ERROR level may result in losing valuable information during development and debugging. It's generally better to have more fine-grained control over log levels for different components.",
+    "solution": "Instead of setting all loggers to ERROR, consider the following:\n1. Set a default log level (e.g., INFO) for all loggers using `logging.basicConfig()`.\n2. Selectively set the log level for specific loggers (e.g., 'LiteLLM', 'LiteLLM Router', 'LiteLLM Proxy') to a more appropriate level (e.g., DEBUG, INFO, or WARNING) based on the importance and verbosity of each component.\n3. Provide a way for users to easily adjust the log level, such as through an environment variable or a configuration file.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/llms/provider.py",
+    "start_line": 13,
+    "end_line": 28,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "11",
+    "end_line": "11",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_400/review.md b/.experiments/code_review/haiku/no_eval/pr_400/review.md
new file mode 100644
index 00000000..a785248a
--- /dev/null
+++ b/.experiments/code_review/haiku/no_eval/pr_400/review.md
@@ -0,0 +1,177 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/400
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 27
+- Critical: 1
+- Important: 4
+- Minor: 9
+- Files Affected: 8
+## 🏆 Code Quality
+[██████████████████░░] 90% (Excellent)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Configuration (1 issues)</strong></summary>
+
+### 1. Changes made to sensitive file
+📁 **File:** `config.json:11`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to config.json, which needs review
+💡 **Solution:** NA
+
+</details>
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Boundary Conditions (4 issues)</strong></summary>
+
+### 1. The test cases for boundary conditions (very long descriptions) look good. The execution time is also printed, which is a nice addition.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_description.py:45`
+⚖️ **Severity:** 8/10
+🔍 **Description:** Handling large inputs is an important aspect of the function's robustness.
+💡 **Solution:** No changes needed.
+
+### 2. The error handling tests cover various invalid input scenarios, which is good.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_description.py:31`
+⚖️ **Severity:** 8/10
+🔍 **Description:** Proper error handling is crucial for the function's reliability.
+💡 **Solution:** No changes needed.
+
+### 3. The `generate_tests` method has become quite long and complex. Consider breaking it down into smaller, more focused methods to improve readability and maintainability.
+📁 **File:** `kaizen/generator/unit_test.py:0`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Large methods can be difficult to understand and maintain, especially as the codebase grows.
+💡 **Solution:** Refactor the `generate_tests` method by extracting smaller, more focused methods for specific tasks, such as preparing the test file path, generating the AI tests, and writing the test file.
+
+### 4. The `UnitTestGenerator` class is responsible for both generating and running the tests. Consider separating these concerns into two different classes or modules.
+📁 **File:** `kaizen/generator/unit_test.py:0`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Separating the concerns of test generation and test execution can improve the overall design and maintainability of the codebase.
+💡 **Solution:** Create a separate `UnitTestRunner` class or module that is responsible for discovering and running the generated tests, while the `UnitTestGenerator` class focuses solely on generating the tests.
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (9 issues)</strong></summary>
+
+<details>
+<summary><strong>Collapsible Template (9 issues)</strong></summary>
+
+### 1. The collapsible template for the original description has been improved to include newlines for better readability.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_description.py:5`
+⚖️ **Severity:** 5/10
+🔍 **Description:** The previous template did not have proper newline formatting.
+💡 **Solution:** No changes needed.
+
+### 2. The imports `mock` and `pytest` are not used in the updated tests.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py:1`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Unused imports can make the code harder to read and maintain.
+💡 **Solution:** Remove the unused imports.
+
+**Current Code:**
+```python
+import os
+import pytest
+from unittest import mock
+from kaizen.helpers.output import get_parent_folder
+```
+
+**Suggested Code:**
+```python
+import os
+from kaizen.helpers.output import get_parent_folder
+```
+
+### 3. The `test_get_parent_folder_normal()` function does not need to mock `os.getcwd()` as the actual implementation can be used.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py:13`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Mocking should be used only when necessary, as it can make the tests more complex and harder to maintain.
+💡 **Solution:** Remove the mocking in `test_get_parent_folder_normal()` and use the actual implementation of `get_parent_folder()`.
+
+**Current Code:**
+```python
+    with mock.patch('os.getcwd', return_value='/home/user/project'):
+        expected = '/home/user/project'
+        result = get_parent_folder()
+        assert result == expected, f"Expected{expected}, but got{result}"
+```
+
+**Suggested Code:**
+```python
+    expected = os.path.dirname(os.getcwd())
+    result = get_parent_folder()
+    assert result == expected, f"Expected{expected}, but got{result}"
+```
+
+### 4. The imports `asyncio` and `nest_asyncio` are not used in the original test cases. Consider removing them if they are not required.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_web_html.py:4`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Unused imports can clutter the code and make it less readable.
+💡 **Solution:** Remove the unused imports `asyncio` and `nest_asyncio` from the test file.
+
+### 5. The previous test `test_get_web_html_invalid_html` has been removed. Consider adding a new test case to ensure the function can handle invalid HTML content gracefully.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_web_html.py:98`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Handling invalid HTML is important for a robust web scraping implementation.
+💡 **Solution:** Add a new test case to ensure the function can handle invalid HTML content without raising unexpected exceptions.
+
+### 6. The logging implementation could be improved by using a more structured approach, such as using the built-in `logging` module with appropriate log levels.
+📁 **File:** `kaizen/generator/unit_test.py:0`
+⚖️ **Severity:** 5/10
+🔍 **Description:** The current logging implementation uses print statements, which can be less flexible and harder to manage than a structured logging approach.
+💡 **Solution:** Refactor the logging implementation to use the `logging` module, with appropriate log levels (e.g., DEBUG, INFO, WARNING, ERROR) and log messages that provide more context and details.
+
+### 7. The `generate_tests_from_dir` method could benefit from more robust error handling, such as catching and handling specific exceptions.
+📁 **File:** `kaizen/generator/unit_test.py:0`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Catching and handling specific exceptions can help provide more informative error messages and improve the overall robustness of the application.
+💡 **Solution:** Modify the `generate_tests_from_dir` method to catch and handle specific exceptions, such as `FileNotFoundError` or `ValueError`, and provide more detailed error messages to help with debugging and troubleshooting.
+
+### 8. The `UnitTestGenerator` class has several dependencies, such as the `LLMProvider` and the various prompt templates. Consider using dependency injection to improve the testability and flexibility of the class.
+📁 **File:** `kaizen/generator/unit_test.py:0`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Dependency injection can make the code more modular and easier to test, as it allows for easier substitution of dependencies.
+💡 **Solution:** Refactor the `UnitTestGenerator` class to accept its dependencies (e.g., `LLMProvider`, prompt templates) as constructor arguments, rather than creating them internally. This will improve the testability and flexibility of the class.
+
+### 9. The code sets all loggers to the ERROR level, which may be too restrictive. Consider providing more granular control over log levels.
+📁 **File:** `kaizen/llms/provider.py:13`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Setting all loggers to ERROR level may result in losing valuable information during development and debugging. It's generally better to have more fine-grained control over log levels for different components.
+💡 **Solution:** Instead of setting all loggers to ERROR, consider the following:
+1. Set a default log level (e.g., INFO) for all loggers using `logging.basicConfig()`.
+2. Selectively set the log level for specific loggers (e.g., 'LiteLLM', 'LiteLLM Router', 'LiteLLM Proxy') to a more appropriate level (e.g., DEBUG, INFO, or WARNING) based on the importance and verbosity of each component.
+3. Provide a way for users to easily adjust the log level, such as through an environment variable or a configuration file.
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (anthropic.claude-3-haiku-20240307-v1:0)
+{"prompt_tokens": 42996, "completion_tokens": 5792, "total_tokens": 48788}
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_440/comments.json b/.experiments/code_review/haiku/no_eval/pr_440/comments.json
new file mode 100644
index 00000000..0637a088
--- /dev/null
+++ b/.experiments/code_review/haiku/no_eval/pr_440/comments.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_440/issues.json b/.experiments/code_review/haiku/no_eval/pr_440/issues.json
new file mode 100644
index 00000000..8e24d702
--- /dev/null
+++ b/.experiments/code_review/haiku/no_eval/pr_440/issues.json
@@ -0,0 +1,62 @@
+[
+  {
+    "topic": "Unnecessary comment removal",
+    "comment": "The commented-out line `# analyzer.setup_repository(\"./github_app/\")` should be removed, as the line below it already sets up the repository.",
+    "confidence": "moderate",
+    "reason": "Removing unnecessary comments improves code readability and maintainability.",
+    "solution": "Remove the commented-out line `# analyzer.setup_repository(\"./github_app/\")` in `examples/ragify_codebase/main.py`.",
+    "actual_code": "# analyzer.setup_repository(\"./github_app/\")",
+    "fixed_code": "",
+    "file_name": "examples/ragify_codebase/main.py",
+    "start_line": 7,
+    "end_line": 7,
+    "side": "LEFT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Duplicate embedding storage",
+    "comment": "The TODO comment `# TODO: DONT PUSH DUPLICATE` suggests that the code is storing duplicate embeddings, which could lead to performance and storage issues.",
+    "confidence": "important",
+    "reason": "Storing duplicate embeddings can waste storage space and slow down the retrieval process.",
+    "solution": "Implement a mechanism to check for and avoid storing duplicate embeddings in the database.",
+    "actual_code": "# TODO: DONT PUSH DUPLICATE",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 156,
+    "end_line": 156,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Dependency version update",
+    "comment": "The `llama-index-core` dependency version has been updated from `0.10.47` to `0.10.65`. This update should be reviewed to ensure compatibility with the rest of the codebase.",
+    "confidence": "moderate",
+    "reason": "Dependency version updates can introduce breaking changes, so it's important to review the changes and ensure they don't introduce any issues.",
+    "solution": "Review the changelog and release notes for the `llama-index-core` version update to understand the changes and ensure they don't introduce any issues in the codebase.",
+    "actual_code": "llama-index-core = \"0.10.65\"",
+    "fixed_code": "",
+    "file_name": "pyproject.toml",
+    "start_line": 27,
+    "end_line": 27,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Dependency removal",
+    "comment": "The `llama-index-llms-openai` dependency has been removed. This change should be reviewed to ensure that it doesn't impact the functionality of the codebase.",
+    "confidence": "moderate",
+    "reason": "Removing dependencies can have unintended consequences, so it's important to review the impact of the change.",
+    "solution": "Review the codebase to ensure that the removal of the `llama-index-llms-openai` dependency doesn't break any functionality.",
+    "actual_code": "llama-index-llms-openai = \"^0.1.22\"",
+    "fixed_code": "",
+    "file_name": "pyproject.toml",
+    "start_line": 28,
+    "end_line": 28,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_440/review.md b/.experiments/code_review/haiku/no_eval/pr_440/review.md
new file mode 100644
index 00000000..62d5b74d
--- /dev/null
+++ b/.experiments/code_review/haiku/no_eval/pr_440/review.md
@@ -0,0 +1,118 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/440
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 4
+- Critical: 0
+- Important: 1
+- Minor: 3
+- Files Affected: 3
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Duplicate embedding storage (1 issues)</strong></summary>
+
+### 1. The TODO comment `# TODO: DONT PUSH DUPLICATE` suggests that the code is storing duplicate embeddings, which could lead to performance and storage issues.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:156`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Storing duplicate embeddings can waste storage space and slow down the retrieval process.
+💡 **Solution:** Implement a mechanism to check for and avoid storing duplicate embeddings in the database.
+
+**Current Code:**
+```python
+# TODO: DONT PUSH DUPLICATE
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (3 issues)</strong></summary>
+
+<details>
+<summary><strong>Unnecessary comment removal (3 issues)</strong></summary>
+
+### 1. The commented-out line `# analyzer.setup_repository("./github_app/")` should be removed, as the line below it already sets up the repository.
+📁 **File:** `examples/ragify_codebase/main.py:7`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Removing unnecessary comments improves code readability and maintainability.
+💡 **Solution:** Remove the commented-out line `# analyzer.setup_repository("./github_app/")` in `examples/ragify_codebase/main.py`.
+
+**Current Code:**
+```python
+# analyzer.setup_repository("./github_app/")
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 2. The `llama-index-core` dependency version has been updated from `0.10.47` to `0.10.65`. This update should be reviewed to ensure compatibility with the rest of the codebase.
+📁 **File:** `pyproject.toml:27`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Dependency version updates can introduce breaking changes, so it's important to review the changes and ensure they don't introduce any issues.
+💡 **Solution:** Review the changelog and release notes for the `llama-index-core` version update to understand the changes and ensure they don't introduce any issues in the codebase.
+
+**Current Code:**
+```python
+llama-index-core = "0.10.65"
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 3. The `llama-index-llms-openai` dependency has been removed. This change should be reviewed to ensure that it doesn't impact the functionality of the codebase.
+📁 **File:** `pyproject.toml:28`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Removing dependencies can have unintended consequences, so it's important to review the impact of the change.
+💡 **Solution:** Review the codebase to ensure that the removal of the `llama-index-llms-openai` dependency doesn't break any functionality.
+
+**Current Code:**
+```python
+llama-index-llms-openai = "^0.1.22"
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (anthropic.claude-3-haiku-20240307-v1:0)
+{"prompt_tokens": 1579, "completion_tokens": 944, "total_tokens": 2523}
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_476/comments.json b/.experiments/code_review/haiku/no_eval/pr_476/comments.json
new file mode 100644
index 00000000..a9d40eac
--- /dev/null
+++ b/.experiments/code_review/haiku/no_eval/pr_476/comments.json
@@ -0,0 +1,16 @@
+[
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_476/issues.json b/.experiments/code_review/haiku/no_eval/pr_476/issues.json
new file mode 100644
index 00000000..15365ce7
--- /dev/null
+++ b/.experiments/code_review/haiku/no_eval/pr_476/issues.json
@@ -0,0 +1,106 @@
+[
+  {
+    "topic": "Sorting PR Files",
+    "comment": "The PR files are now being sorted before passing them to the description generator. This is a good improvement for maintaining consistent file order in the review.",
+    "confidence": "important",
+    "reason": "Sorting the files ensures a consistent order in the review, making it easier for the reviewer to understand the changes.",
+    "solution": "The `sort_files` function looks good and should effectively sort the files in alphabetical order.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 184,
+    "end_line": 194,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Generating Tests",
+    "comment": "The new `generate_tests` function is a good addition, as it provides a way to generate test cases based on the changed files in the PR.",
+    "confidence": "important",
+    "reason": "Generating tests based on the PR files can help ensure the changes don't break existing functionality.",
+    "solution": "The current implementation of `generate_tests` is simple and effective. It returns a list of file names, which can be used to create test cases.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 199,
+    "end_line": 200,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Printing Diff and PR Files",
+    "comment": "The changes in the `main.py` file to print the `diff_text` and `pr_files` are useful for debugging and understanding the input data.",
+    "confidence": "moderate",
+    "reason": "Printing the diff and PR files can help developers better understand the changes being reviewed.",
+    "solution": "The changes look good and should provide helpful information during the review process.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "examples/code_review/main.py",
+    "start_line": 21,
+    "end_line": 22,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Passing Code Quality to Review Description",
+    "comment": "The change to pass the `code_quality` parameter to the `create_pr_review_text` function is a good improvement, as it allows the review description to include information about the overall code quality.",
+    "confidence": "important",
+    "reason": "Providing information about the code quality in the review description can give the developer a better understanding of the overall state of the codebase.",
+    "solution": "The change looks good and should provide valuable information in the review description.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "examples/code_review/main.py",
+    "start_line": 36,
+    "end_line": 36,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Printing Raw Issues",
+    "comment": "The change to print the `review_data.issues` instead of the `topics` variable is an improvement, as it provides more detailed information about the identified issues.",
+    "confidence": "moderate",
+    "reason": "Printing the raw issues can give the developer a better understanding of the specific problems found during the review.",
+    "solution": "The change looks good and should provide more useful information in the output.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "examples/code_review/main.py",
+    "start_line": 39,
+    "end_line": 39,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Removing Unused Configuration",
+    "comment": "The removal of the `enable_observability_logging` configuration option is a good cleanup, as it removes an unused feature from the configuration file.",
+    "confidence": "moderate",
+    "reason": "Removing unused configuration options helps keep the codebase clean and maintainable.",
+    "solution": "The change looks good and should help simplify the configuration file.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "config.json",
+    "start_line": 4,
+    "end_line": 4,
+    "side": "LEFT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_476/review.md b/.experiments/code_review/haiku/no_eval/pr_476/review.md
new file mode 100644
index 00000000..86156ba1
--- /dev/null
+++ b/.experiments/code_review/haiku/no_eval/pr_476/review.md
@@ -0,0 +1,103 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/476
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 7
+- Critical: 1
+- Important: 3
+- Minor: 3
+- Files Affected: 3
+## 🏆 Code Quality
+[██████████████████░░] 90% (Excellent)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Configuration (1 issues)</strong></summary>
+
+### 1. Changes made to sensitive file
+📁 **File:** `config.json:1`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to config.json, which needs review
+💡 **Solution:** NA
+
+</details>
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Sorting PR Files (3 issues)</strong></summary>
+
+### 1. The PR files are now being sorted before passing them to the description generator. This is a good improvement for maintaining consistent file order in the review.
+📁 **File:** `github_app/github_helper/pull_requests.py:184`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Sorting the files ensures a consistent order in the review, making it easier for the reviewer to understand the changes.
+💡 **Solution:** The `sort_files` function looks good and should effectively sort the files in alphabetical order.
+
+### 2. The new `generate_tests` function is a good addition, as it provides a way to generate test cases based on the changed files in the PR.
+📁 **File:** `github_app/github_helper/pull_requests.py:199`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Generating tests based on the PR files can help ensure the changes don't break existing functionality.
+💡 **Solution:** The current implementation of `generate_tests` is simple and effective. It returns a list of file names, which can be used to create test cases.
+
+### 3. The change to pass the `code_quality` parameter to the `create_pr_review_text` function is a good improvement, as it allows the review description to include information about the overall code quality.
+📁 **File:** `examples/code_review/main.py:36`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Providing information about the code quality in the review description can give the developer a better understanding of the overall state of the codebase.
+💡 **Solution:** The change looks good and should provide valuable information in the review description.
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (3 issues)</strong></summary>
+
+<details>
+<summary><strong>Printing Diff and PR Files (3 issues)</strong></summary>
+
+### 1. The changes in the `main.py` file to print the `diff_text` and `pr_files` are useful for debugging and understanding the input data.
+📁 **File:** `examples/code_review/main.py:21`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Printing the diff and PR files can help developers better understand the changes being reviewed.
+💡 **Solution:** The changes look good and should provide helpful information during the review process.
+
+### 2. The change to print the `review_data.issues` instead of the `topics` variable is an improvement, as it provides more detailed information about the identified issues.
+📁 **File:** `examples/code_review/main.py:39`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Printing the raw issues can give the developer a better understanding of the specific problems found during the review.
+💡 **Solution:** The change looks good and should provide more useful information in the output.
+
+### 3. The removal of the `enable_observability_logging` configuration option is a good cleanup, as it removes an unused feature from the configuration file.
+📁 **File:** `config.json:4`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Removing unused configuration options helps keep the codebase clean and maintainable.
+💡 **Solution:** The change looks good and should help simplify the configuration file.
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (anthropic.claude-3-haiku-20240307-v1:0)
+{"prompt_tokens": 4511, "completion_tokens": 1305, "total_tokens": 5816}
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_5/comments.json b/.experiments/code_review/haiku/no_eval/pr_5/comments.json
new file mode 100644
index 00000000..1130b98e
--- /dev/null
+++ b/.experiments/code_review/haiku/no_eval/pr_5/comments.json
@@ -0,0 +1,32 @@
+[
+  {
+    "topic": "Error Handling",
+    "comment": "Potential for API call to fail without retry mechanism",
+    "confidence": "critical",
+    "reason": "The `process_applicant` function makes a call to the `completion` function, which could fail for various reasons (e.g., network issues, API downtime). Without a retry mechanism, the function will silently fail, leading to incomplete or inaccurate results.",
+    "solution": "Implement a retry mechanism with exponential backoff to handle transient failures in the `completion` function call. This will improve the reliability and robustness of the application.",
+    "actual_code": "response = completion(\n    model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n)",
+    "fixed_code": "num_retries = 3\nretry_delay = 1\nfor _ in range(num_retries):\n    try:\n        response = completion(\n            model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n        )\n        break\n    except Exception as e:\n        print(f\"Error calling completion function:{e}. Retrying in{retry_delay}seconds...\")\n        time.sleep(retry_delay)\n        retry_delay *= 2\nelse:\n    print(\"Failed to call completion function after multiple retries. Skipping this applicant.\")\n    return{key: \"\" for key in[\"feedback\", \"review\", \"should_interview\", \"rating\", \"input_tokens\", \"output_tokens\"]}",
+    "file_name": "main.py",
+    "start_line": 66,
+    "end_line": 68,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Silent failure without logging",
+    "confidence": "critical",
+    "reason": "In the `process_applicant` function, when a `json.JSONDecodeError` is raised, the function silently returns a default result without any logging or error reporting. This makes it difficult to diagnose and troubleshoot issues that may occur during the processing of applicants.",
+    "solution": "Add proper error logging to the `except` block to capture the error and provide more visibility into the failure. This will help with debugging and monitoring the application's performance.",
+    "actual_code": "except json.JSONDecodeError:\n    result ={\n        key: \"\" for key in[\n            \"feedback\",\n            \"review\",\n            \"should_interview\",\n            \"rating\",\n            \"input_tokens\",\n            \"output_tokens\",\n        ]\n}",
+    "fixed_code": "except json.JSONDecodeError as e:\n    print(f\"Failed to parse content for applicant:{e}\")\n    result ={\n        key: \"\" for key in[\n            \"feedback\",\n            \"review\",\n            \"should_interview\",\n            \"rating\",\n            \"input_tokens\",\n            \"output_tokens\",\n        ]\n}",
+    "file_name": "main.py",
+    "start_line": 82,
+    "end_line": 94,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 8
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_5/issues.json b/.experiments/code_review/haiku/no_eval/pr_5/issues.json
new file mode 100644
index 00000000..5482c08e
--- /dev/null
+++ b/.experiments/code_review/haiku/no_eval/pr_5/issues.json
@@ -0,0 +1,107 @@
+[
+  {
+    "topic": "Error Handling",
+    "comment": "Potential for API call to fail without retry mechanism",
+    "confidence": "critical",
+    "reason": "The `process_applicant` function makes a call to the `completion` function, which could fail for various reasons (e.g., network issues, API downtime). Without a retry mechanism, the function will silently fail, leading to incomplete or inaccurate results.",
+    "solution": "Implement a retry mechanism with exponential backoff to handle transient failures in the `completion` function call. This will improve the reliability and robustness of the application.",
+    "actual_code": "response = completion(\n    model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n)",
+    "fixed_code": "num_retries = 3\nretry_delay = 1\nfor _ in range(num_retries):\n    try:\n        response = completion(\n            model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n        )\n        break\n    except Exception as e:\n        print(f\"Error calling completion function:{e}. Retrying in{retry_delay}seconds...\")\n        time.sleep(retry_delay)\n        retry_delay *= 2\nelse:\n    print(\"Failed to call completion function after multiple retries. Skipping this applicant.\")\n    return{key: \"\" for key in[\"feedback\", \"review\", \"should_interview\", \"rating\", \"input_tokens\", \"output_tokens\"]}",
+    "file_name": "main.py",
+    "start_line": 66,
+    "end_line": 68,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Silent failure without logging",
+    "confidence": "critical",
+    "reason": "In the `process_applicant` function, when a `json.JSONDecodeError` is raised, the function silently returns a default result without any logging or error reporting. This makes it difficult to diagnose and troubleshoot issues that may occur during the processing of applicants.",
+    "solution": "Add proper error logging to the `except` block to capture the error and provide more visibility into the failure. This will help with debugging and monitoring the application's performance.",
+    "actual_code": "except json.JSONDecodeError:\n    result ={\n        key: \"\" for key in[\n            \"feedback\",\n            \"review\",\n            \"should_interview\",\n            \"rating\",\n            \"input_tokens\",\n            \"output_tokens\",\n        ]\n}",
+    "fixed_code": "except json.JSONDecodeError as e:\n    print(f\"Failed to parse content for applicant:{e}\")\n    result ={\n        key: \"\" for key in[\n            \"feedback\",\n            \"review\",\n            \"should_interview\",\n            \"rating\",\n            \"input_tokens\",\n            \"output_tokens\",\n        ]\n}",
+    "file_name": "main.py",
+    "start_line": 82,
+    "end_line": 94,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Performance",
+    "comment": "Inefficient way to print progress",
+    "confidence": "important",
+    "reason": "The `process_applicants` function uses a print statement with carriage return (`\r`) to update the progress bar. This approach can be inefficient, especially for large datasets, as it requires continuously overwriting the same line of output.",
+    "solution": "Use a dedicated progress reporting library, such as `tqdm`, which provides a more efficient and visually appealing progress bar. This will improve the overall performance and user experience of the application.",
+    "actual_code": "progress = (index + 1) / total\nprint(f\"\\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}\", end=\"\", flush=True)",
+    "fixed_code": "if use_tqdm:\n    progress_bar = tqdm(total=total, desc=\"Processing applicants\")\n    progress_bar.update(1)\nelse:\n    progress = (index + 1) / total\n    print(f\"\\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}\", end=\"\", flush=True)",
+    "file_name": "main.py",
+    "start_line": 120,
+    "end_line": 122,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Redundant Code",
+    "comment": "Redundant code: The following line is unnecessary",
+    "confidence": "moderate",
+    "reason": "The `if len(df) == 0` check in the `main` function is redundant, as the `process_applicants` function already handles the case where the DataFrame is empty.",
+    "solution": "Remove the unnecessary `if` statement, as it does not provide any additional value to the code.",
+    "actual_code": "if len(df) == 0:\n    return",
+    "fixed_code": "",
+    "file_name": "main.py",
+    "start_line": 142,
+    "end_line": 143,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Division by zero potential if total_tokens is zero",
+    "confidence": "important",
+    "reason": "In the `main` function, the code calculates the total tokens used and prints a summary. However, if the total tokens is zero, the division operation will result in a division by zero error, which can cause the application to crash.",
+    "solution": "Add a check to ensure that the total tokens is not zero before performing the division operation. If the total tokens is zero, handle the case gracefully by printing a message or skipping the division step.",
+    "actual_code": "print(f\"Total tokens used:{total_tokens:,}\")\nprint(f\"  - Input tokens:{total_input_tokens:,}\")\nprint(f\"  - Output tokens:{total_output_tokens:,}\")",
+    "fixed_code": "if total_tokens > 0:\n    print(f\"Total tokens used:{total_tokens:,}\")\n    print(f\"  - Input tokens:{total_input_tokens:,}\")\n    print(f\"  - Output tokens:{total_output_tokens:,}\")\nelse:\n    print(\"Total tokens used: 0\")",
+    "file_name": "main.py",
+    "start_line": 158,
+    "end_line": 163,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "No error handling for file not found",
+    "confidence": "important",
+    "reason": "The `main` function does not handle the case where the input file specified by the user does not exist. This can lead to a `FileNotFoundError` being raised, which will cause the application to crash without any meaningful error message.",
+    "solution": "Add a try-except block to handle the `FileNotFoundError` and provide a user-friendly error message when the input file is not found.",
+    "actual_code": "main(input_file)",
+    "fixed_code": "try:\n    main(input_file)\nexcept FileNotFoundError:\n    print(f\"Error: The file '{input_file}' does not exist. Please check the file path and try again.\")",
+    "file_name": "main.py",
+    "start_line": 174,
+    "end_line": 175,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Unused Import",
+    "comment": "Unused import",
+    "confidence": "low",
+    "reason": "The `random` module is imported but not used in the code.",
+    "solution": "Remove the unused import statement.",
+    "actual_code": "import random  # Unused import",
+    "fixed_code": "",
+    "file_name": "main.py",
+    "start_line": 8,
+    "end_line": 8,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 1
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_5/review.md b/.experiments/code_review/haiku/no_eval/pr_5/review.md
new file mode 100644
index 00000000..5e96731d
--- /dev/null
+++ b/.experiments/code_review/haiku/no_eval/pr_5/review.md
@@ -0,0 +1,211 @@
+PR URL: https://github.com/sauravpanda/applicant-screening/pull/5
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 7
+- Critical: 2
+- Important: 3
+- Minor: 1
+- Files Affected: 1
+## 🏆 Code Quality
+[███████████████░░░░░] 75% (Fair)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Error Handling (2 issues)</strong></summary>
+
+### 1. Potential for API call to fail without retry mechanism
+📁 **File:** `main.py:66`
+⚖️ **Severity:** 9/10
+🔍 **Description:** The `process_applicant` function makes a call to the `completion` function, which could fail for various reasons (e.g., network issues, API downtime). Without a retry mechanism, the function will silently fail, leading to incomplete or inaccurate results.
+💡 **Solution:** Implement a retry mechanism with exponential backoff to handle transient failures in the `completion` function call. This will improve the reliability and robustness of the application.
+
+**Current Code:**
+```python
+response = completion(
+    model=os.environ.get("model", "anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1"), messages=messages
+)
+```
+
+**Suggested Code:**
+```python
+num_retries = 3
+retry_delay = 1
+for _ in range(num_retries):
+    try:
+        response = completion(
+            model=os.environ.get("model", "anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1"), messages=messages
+        )
+        break
+    except Exception as e:
+        print(f"Error calling completion function:{e}. Retrying in{retry_delay}seconds...")
+        time.sleep(retry_delay)
+        retry_delay *= 2
+else:
+    print("Failed to call completion function after multiple retries. Skipping this applicant.")
+    return{key: "" for key in["feedback", "review", "should_interview", "rating", "input_tokens", "output_tokens"]}
+```
+
+### 2. Silent failure without logging
+📁 **File:** `main.py:82`
+⚖️ **Severity:** 8/10
+🔍 **Description:** In the `process_applicant` function, when a `json.JSONDecodeError` is raised, the function silently returns a default result without any logging or error reporting. This makes it difficult to diagnose and troubleshoot issues that may occur during the processing of applicants.
+💡 **Solution:** Add proper error logging to the `except` block to capture the error and provide more visibility into the failure. This will help with debugging and monitoring the application's performance.
+
+**Current Code:**
+```python
+except json.JSONDecodeError:
+    result ={
+        key: "" for key in[
+            "feedback",
+            "review",
+            "should_interview",
+            "rating",
+            "input_tokens",
+            "output_tokens",
+        ]
+}
+```
+
+**Suggested Code:**
+```python
+except json.JSONDecodeError as e:
+    print(f"Failed to parse content for applicant:{e}")
+    result ={
+        key: "" for key in[
+            "feedback",
+            "review",
+            "should_interview",
+            "rating",
+            "input_tokens",
+            "output_tokens",
+        ]
+}
+```
+
+</details>
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Performance (3 issues)</strong></summary>
+
+### 1. Inefficient way to print progress
+📁 **File:** `main.py:120`
+⚖️ **Severity:** 6/10
+🔍 **Description:** The `process_applicants` function uses a print statement with carriage return (`
`) to update the progress bar. This approach can be inefficient, especially for large datasets, as it requires continuously overwriting the same line of output.
+💡 **Solution:** Use a dedicated progress reporting library, such as `tqdm`, which provides a more efficient and visually appealing progress bar. This will improve the overall performance and user experience of the application.
+
+**Current Code:**
+```python
+progress = (index + 1) / total
+print(f"\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}", end="", flush=True)
+```
+
+**Suggested Code:**
+```python
+if use_tqdm:
+    progress_bar = tqdm(total=total, desc="Processing applicants")
+    progress_bar.update(1)
+else:
+    progress = (index + 1) / total
+    print(f"\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}", end="", flush=True)
+```
+
+### 2. Division by zero potential if total_tokens is zero
+📁 **File:** `main.py:158`
+⚖️ **Severity:** 7/10
+🔍 **Description:** In the `main` function, the code calculates the total tokens used and prints a summary. However, if the total tokens is zero, the division operation will result in a division by zero error, which can cause the application to crash.
+💡 **Solution:** Add a check to ensure that the total tokens is not zero before performing the division operation. If the total tokens is zero, handle the case gracefully by printing a message or skipping the division step.
+
+**Current Code:**
+```python
+print(f"Total tokens used:{total_tokens:,}")
+print(f"  - Input tokens:{total_input_tokens:,}")
+print(f"  - Output tokens:{total_output_tokens:,}")
+```
+
+**Suggested Code:**
+```python
+if total_tokens > 0:
+    print(f"Total tokens used:{total_tokens:,}")
+    print(f"  - Input tokens:{total_input_tokens:,}")
+    print(f"  - Output tokens:{total_output_tokens:,}")
+else:
+    print("Total tokens used: 0")
+```
+
+### 3. No error handling for file not found
+📁 **File:** `main.py:174`
+⚖️ **Severity:** 7/10
+🔍 **Description:** The `main` function does not handle the case where the input file specified by the user does not exist. This can lead to a `FileNotFoundError` being raised, which will cause the application to crash without any meaningful error message.
+💡 **Solution:** Add a try-except block to handle the `FileNotFoundError` and provide a user-friendly error message when the input file is not found.
+
+**Current Code:**
+```python
+main(input_file)
+```
+
+**Suggested Code:**
+```python
+try:
+    main(input_file)
+except FileNotFoundError:
+    print(f"Error: The file '{input_file}' does not exist. Please check the file path and try again.")
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (2 issues)</strong></summary>
+
+<details>
+<summary><strong>Redundant Code (1 issues)</strong></summary>
+
+### 1. Redundant code: The following line is unnecessary
+📁 **File:** `main.py:142`
+⚖️ **Severity:** 3/10
+🔍 **Description:** The `if len(df) == 0` check in the `main` function is redundant, as the `process_applicants` function already handles the case where the DataFrame is empty.
+💡 **Solution:** Remove the unnecessary `if` statement, as it does not provide any additional value to the code.
+
+**Current Code:**
+```python
+if len(df) == 0:
+    return
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (anthropic.claude-3-haiku-20240307-v1:0)
+{"prompt_tokens": 7009, "completion_tokens": 2247, "total_tokens": 9256}
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_222/comments.json b/.experiments/code_review/llama-405b/no_eval/pr_222/comments.json
new file mode 100644
index 00000000..f054c504
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_222/comments.json
@@ -0,0 +1,117 @@
+[
+  {
+    "topic": "Error handling",
+    "comment": "Error handling is missing in some critical sections of the code.",
+    "confidence": "critical",
+    "reason": "Error handling is crucial for preventing crashes and providing useful error messages.",
+    "solution": "Add try-except blocks to handle potential errors.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "",
+    "start_line": 0,
+    "end_line": 0,
+    "side": "",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "SQL Injection",
+    "comment": "Potential SQL injection vulnerability",
+    "confidence": "critical",
+    "reason": "Using string formatting to construct SQL queries can lead to SQL injection attacks",
+    "solution": "Use parameterized queries or an ORM to prevent SQL injection",
+    "actual_code": "query = f\"\"\"SELECT ... FROM{self.table_name}e\"\"\"",
+    "fixed_code": "query = \"\"\"SELECT ... FROM %s e\"\"\" % self.table_name",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 19,
+    "end_line": 37,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "Version bump",
+    "comment": "Verify that the version bump is intentional and follows the project's versioning scheme",
+    "confidence": "critical",
+    "reason": "Inconsistent versioning can cause confusion and break dependencies",
+    "solution": "Verify the version bump and update the project's versioning scheme if necessary",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "pyproject.toml",
+    "start_line": 3,
+    "end_line": 3,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 9
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Docker",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to Dockerfile, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_name": "Dockerfile",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Docker",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to docker-compose.yml, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "15",
+    "end_line": "15",
+    "side": "RIGHT",
+    "file_name": "docker-compose.yml",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Version Control",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to .gitignore, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "164",
+    "end_line": "164",
+    "side": "RIGHT",
+    "file_name": ".gitignore",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Database",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to db_setup/init.sql, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "db_setup/init.sql",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_222/issues.json b/.experiments/code_review/llama-405b/no_eval/pr_222/issues.json
new file mode 100644
index 00000000..599c0667
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_222/issues.json
@@ -0,0 +1,282 @@
+[
+  {
+    "topic": "Code organization",
+    "comment": "The code is well-organized, but some files have too many responsibilities.",
+    "confidence": "important",
+    "reason": "Separation of concerns is crucial for maintainability.",
+    "solution": "Consider breaking down large files into smaller ones.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "",
+    "start_line": 0,
+    "end_line": 0,
+    "side": "",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Type hints",
+    "comment": "Type hints are missing in some function definitions.",
+    "confidence": "moderate",
+    "reason": "Type hints improve code readability and help catch type-related errors.",
+    "solution": "Add type hints for function parameters and return types.",
+    "actual_code": "def chunk_code(code: str, language: str) -> ParsedBody:",
+    "fixed_code": "def chunk_code(code: str, language: str) -> Dict[str, Dict[str, Any]]:",
+    "file_name": "kaizen/retriever/code_chunker.py",
+    "start_line": 7,
+    "end_line": 7,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Error handling",
+    "comment": "Error handling is missing in some critical sections of the code.",
+    "confidence": "critical",
+    "reason": "Error handling is crucial for preventing crashes and providing useful error messages.",
+    "solution": "Add try-except blocks to handle potential errors.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "",
+    "start_line": 0,
+    "end_line": 0,
+    "side": "",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Code duplication",
+    "comment": "Some code is duplicated across multiple files.",
+    "confidence": "moderate",
+    "reason": "Code duplication makes maintenance harder and increases the chance of bugs.",
+    "solution": "Extract duplicated code into reusable functions or classes.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "",
+    "start_line": 0,
+    "end_line": 0,
+    "side": "",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Type Hints",
+    "comment": "Missing type hints for function return types",
+    "confidence": "moderate",
+    "reason": "Type hints improve code readability and help catch type-related errors",
+    "solution": "Add type hints for function return types",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 13,
+    "end_line": 52,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "SQL Injection",
+    "comment": "Potential SQL injection vulnerability",
+    "confidence": "critical",
+    "reason": "Using string formatting to construct SQL queries can lead to SQL injection attacks",
+    "solution": "Use parameterized queries or an ORM to prevent SQL injection",
+    "actual_code": "query = f\"\"\"SELECT ... FROM{self.table_name}e\"\"\"",
+    "fixed_code": "query = \"\"\"SELECT ... FROM %s e\"\"\" % self.table_name",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 19,
+    "end_line": 37,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Missing error handling for database operations",
+    "confidence": "important",
+    "reason": "Database operations can fail due to various reasons, and error handling is necessary to prevent crashes",
+    "solution": "Add try-except blocks to handle database operation errors",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 39,
+    "end_line": 52,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "Code Organization",
+    "comment": "AbstractionFeedback class has a single responsibility, but its methods are not well-organized",
+    "confidence": "moderate",
+    "reason": "Well-organized code is easier to read and maintain",
+    "solution": "Consider reorganizing the methods into separate classes or modules",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/feedback_system.py",
+    "start_line": 4,
+    "end_line": 18,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "The code does not handle potential errors that may occur when connecting to the database or executing queries.",
+    "confidence": "important",
+    "reason": "Error handling is crucial to prevent the program from crashing and to provide meaningful error messages instead.",
+    "solution": "Add try-except blocks to handle potential errors when connecting to the database or executing queries.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 250,
+    "end_line": 270,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Code Duplication",
+    "comment": "The code has duplicated logic for storing code in the database and storing function relationships.",
+    "confidence": "moderate",
+    "reason": "Code duplication makes the code harder to maintain and modify.",
+    "solution": "Extract the duplicated logic into a separate function.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 295,
+    "end_line": 313,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 4
+  },
+  {
+    "topic": "Code Readability",
+    "comment": "The code has long and complex functions that are hard to read and understand.",
+    "confidence": "moderate",
+    "reason": "Long and complex functions make the code harder to maintain and modify.",
+    "solution": "Break down the long and complex functions into smaller and simpler functions.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 100,
+    "end_line": 200,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 5
+  },
+  {
+    "topic": "Unused imports",
+    "comment": "Remove unused imports in tree_sitter_utils.py and test_chunker.py",
+    "confidence": "moderate",
+    "reason": "Unused imports can clutter the codebase and make it harder to maintain",
+    "solution": "Remove the unnecessary imports",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 1,
+    "end_line": 1,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Error handling",
+    "comment": "Catch specific exceptions in LanguageLoader.load_language",
+    "confidence": "important",
+    "reason": "Broad exception catching can mask bugs and make debugging harder",
+    "solution": "Catch specific exceptions, such as ImportError or ModuleNotFoundError",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 15,
+    "end_line": 15,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 7
+  },
+  {
+    "topic": "Version bump",
+    "comment": "Verify that the version bump is intentional and follows the project's versioning scheme",
+    "confidence": "critical",
+    "reason": "Inconsistent versioning can cause confusion and break dependencies",
+    "solution": "Verify the version bump and update the project's versioning scheme if necessary",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "pyproject.toml",
+    "start_line": 3,
+    "end_line": 3,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 9
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Docker",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to Dockerfile, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_name": "Dockerfile",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Docker",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to docker-compose.yml, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "15",
+    "end_line": "15",
+    "side": "RIGHT",
+    "file_name": "docker-compose.yml",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Version Control",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to .gitignore, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "164",
+    "end_line": "164",
+    "side": "RIGHT",
+    "file_name": ".gitignore",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Database",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to db_setup/init.sql, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "db_setup/init.sql",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_222/review.md b/.experiments/code_review/llama-405b/no_eval/pr_222/review.md
new file mode 100644
index 00000000..6fd1770a
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_222/review.md
@@ -0,0 +1,195 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/222
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 19
+- Critical: 8
+- Important: 4
+- Minor: 7
+- Files Affected: 12
+## 🏆 Code Quality
+[████████████████░░░░] 80% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Error handling (8 issues)</strong></summary>
+
+### 1. Error handling is missing in some critical sections of the code.
+📁 **File:** `:0`
+⚖️ **Severity:** 8/10
+🔍 **Description:** Error handling is crucial for preventing crashes and providing useful error messages.
+💡 **Solution:** Add try-except blocks to handle potential errors.
+
+### 2. Potential SQL injection vulnerability
+📁 **File:** `kaizen/retriever/custom_vector_store.py:19`
+⚖️ **Severity:** 9/10
+🔍 **Description:** Using string formatting to construct SQL queries can lead to SQL injection attacks
+💡 **Solution:** Use parameterized queries or an ORM to prevent SQL injection
+
+**Current Code:**
+```python
+query = f"""SELECT ... FROM{self.table_name}e"""
+```
+
+**Suggested Code:**
+```python
+query = """SELECT ... FROM %s e""" % self.table_name
+```
+
+### 3. Verify that the version bump is intentional and follows the project's versioning scheme
+📁 **File:** `pyproject.toml:3`
+⚖️ **Severity:** 9/10
+🔍 **Description:** Inconsistent versioning can cause confusion and break dependencies
+💡 **Solution:** Verify the version bump and update the project's versioning scheme if necessary
+
+### 4. Changes made to sensitive file
+📁 **File:** `config.json:4`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to config.json, which needs review
+💡 **Solution:** NA
+
+### 5. Changes made to sensitive file
+📁 **File:** `Dockerfile:4`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to Dockerfile, which needs review
+💡 **Solution:** NA
+
+### 6. Changes made to sensitive file
+📁 **File:** `docker-compose.yml:15`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to docker-compose.yml, which needs review
+💡 **Solution:** NA
+
+### 7. Changes made to sensitive file
+📁 **File:** `.gitignore:164`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to .gitignore, which needs review
+💡 **Solution:** NA
+
+### 8. Changes made to sensitive file
+📁 **File:** `db_setup/init.sql:1`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to db_setup/init.sql, which needs review
+💡 **Solution:** NA
+
+</details>
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Code organization (4 issues)</strong></summary>
+
+### 1. The code is well-organized, but some files have too many responsibilities.
+📁 **File:** `:0`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Separation of concerns is crucial for maintainability.
+💡 **Solution:** Consider breaking down large files into smaller ones.
+
+### 2. Missing error handling for database operations
+📁 **File:** `kaizen/retriever/custom_vector_store.py:39`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Database operations can fail due to various reasons, and error handling is necessary to prevent crashes
+💡 **Solution:** Add try-except blocks to handle database operation errors
+
+### 3. The code does not handle potential errors that may occur when connecting to the database or executing queries.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:250`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Error handling is crucial to prevent the program from crashing and to provide meaningful error messages instead.
+💡 **Solution:** Add try-except blocks to handle potential errors when connecting to the database or executing queries.
+
+### 4. Catch specific exceptions in LanguageLoader.load_language
+📁 **File:** `kaizen/retriever/tree_sitter_utils.py:15`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Broad exception catching can mask bugs and make debugging harder
+💡 **Solution:** Catch specific exceptions, such as ImportError or ModuleNotFoundError
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (7 issues)</strong></summary>
+
+<details>
+<summary><strong>Type hints (7 issues)</strong></summary>
+
+### 1. Type hints are missing in some function definitions.
+📁 **File:** `kaizen/retriever/code_chunker.py:7`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Type hints improve code readability and help catch type-related errors.
+💡 **Solution:** Add type hints for function parameters and return types.
+
+**Current Code:**
+```python
+def chunk_code(code: str, language: str) -> ParsedBody:
+```
+
+**Suggested Code:**
+```python
+def chunk_code(code: str, language: str) -> Dict[str, Dict[str, Any]]:
+```
+
+### 2. Some code is duplicated across multiple files.
+📁 **File:** `:0`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Code duplication makes maintenance harder and increases the chance of bugs.
+💡 **Solution:** Extract duplicated code into reusable functions or classes.
+
+### 3. Missing type hints for function return types
+📁 **File:** `kaizen/retriever/custom_vector_store.py:13`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Type hints improve code readability and help catch type-related errors
+💡 **Solution:** Add type hints for function return types
+
+### 4. AbstractionFeedback class has a single responsibility, but its methods are not well-organized
+📁 **File:** `kaizen/retriever/feedback_system.py:4`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Well-organized code is easier to read and maintain
+💡 **Solution:** Consider reorganizing the methods into separate classes or modules
+
+### 5. The code has duplicated logic for storing code in the database and storing function relationships.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:295`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Code duplication makes the code harder to maintain and modify.
+💡 **Solution:** Extract the duplicated logic into a separate function.
+
+### 6. The code has long and complex functions that are hard to read and understand.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:100`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Long and complex functions make the code harder to maintain and modify.
+💡 **Solution:** Break down the long and complex functions into smaller and simpler functions.
+
+### 7. Remove unused imports in tree_sitter_utils.py and test_chunker.py
+📁 **File:** `kaizen/retriever/tree_sitter_utils.py:1`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Unused imports can clutter the codebase and make it harder to maintain
+💡 **Solution:** Remove the unnecessary imports
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (azure_ai/Meta-Llama-3-405B-Instruct)
+{"prompt_tokens": 21487, "completion_tokens": 2711, "total_tokens": 24198}
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_232/comments.json b/.experiments/code_review/llama-405b/no_eval/pr_232/comments.json
new file mode 100644
index 00000000..0637a088
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_232/comments.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_232/issues.json b/.experiments/code_review/llama-405b/no_eval/pr_232/issues.json
new file mode 100644
index 00000000..1e332158
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_232/issues.json
@@ -0,0 +1,137 @@
+[
+  {
+    "topic": "Unused imports",
+    "comment": "There are several unused imports in the code, such as `redirect` in `page.tsx` and `Switch` and `Label` in `queryinput.tsx`. These imports should be removed to declutter the code.",
+    "confidence": "moderate",
+    "reason": "Unused imports can make the code harder to read and maintain.",
+    "solution": "Remove unused imports.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "page.tsx, queryinput.tsx",
+    "start_line": 0,
+    "end_line": 0,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Type annotations",
+    "comment": "Some variables and functions are missing type annotations, such as the `initialSpaces` prop in `queryinput.tsx`. Adding type annotations can improve code readability and prevent type-related errors.",
+    "confidence": "moderate",
+    "reason": "Type annotations can help catch type-related errors at compile-time.",
+    "solution": "Add type annotations for variables and functions.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "queryinput.tsx",
+    "start_line": 10,
+    "end_line": 10,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Code organization",
+    "comment": "Some files, such as `page.tsx`, contain multiple unrelated components. It would be better to separate these components into their own files to improve code organization and reusability.",
+    "confidence": "low",
+    "reason": "Separating components into their own files can improve code organization and reusability.",
+    "solution": "Separate unrelated components into their own files.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "page.tsx",
+    "start_line": 0,
+    "end_line": 0,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 3
+  },
+  {
+    "topic": "Functionality",
+    "comment": "The code seems to be implementing a memory creation feature with a dialog box. However, there are some potential issues with the functionality.",
+    "confidence": "important",
+    "reason": "The `handleSubmit` function is not properly handling errors. It should be improved to handle errors in a more robust way.",
+    "solution": "Add try-catch blocks to handle errors in the `handleSubmit` function.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "menu.tsx",
+    "start_line": 213,
+    "end_line": 233,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "Code organization",
+    "comment": "The code is not well-organized. There are many functions and variables defined inside the `DialogContentContainer` component.",
+    "confidence": "moderate",
+    "reason": "It would be better to separate the concerns of the component into smaller functions or utilities.",
+    "solution": "Extract some of the functions and variables into separate files or utilities.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "menu.tsx",
+    "start_line": 163,
+    "end_line": 346,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Type annotations",
+    "comment": "Some of the variables and functions are missing type annotations.",
+    "confidence": "low",
+    "reason": "Adding type annotations would improve the code quality and make it easier to understand.",
+    "solution": "Add type annotations for the variables and functions.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "menu.tsx",
+    "start_line": 163,
+    "end_line": 346,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 2
+  },
+  {
+    "topic": "Unused imports",
+    "comment": "The import of `useEffect` from 'react' is not used in the code.",
+    "confidence": "moderate",
+    "reason": "The import is not used anywhere in the code.",
+    "solution": "Remove the unused import.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 3,
+    "end_line": 3,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Type annotations",
+    "comment": "The type annotation for the `handleInputChange` function is missing.",
+    "confidence": "moderate",
+    "reason": "The function is not annotated with a type.",
+    "solution": "Add the type annotation for the function.",
+    "actual_code": "const handleInputChange = (e: React.ChangeEvent<HTMLInputElement>) =>{",
+    "fixed_code": "const handleInputChange: React.ChangeEventHandler<HTMLInputElement> = (e) =>{",
+    "file_name": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 41,
+    "end_line": 41,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Code organization",
+    "comment": "The code is not organized in a logical manner.",
+    "confidence": "low",
+    "reason": "The code is not separated into clear sections or functions.",
+    "solution": "Organize the code into clear sections or functions.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 1,
+    "end_line": 107,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 3
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_232/review.md b/.experiments/code_review/llama-405b/no_eval/pr_232/review.md
new file mode 100644
index 00000000..ef2f635f
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_232/review.md
@@ -0,0 +1,100 @@
+PR URL: https://github.com/supermemoryai/supermemory/pull/232
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 9
+- Critical: 0
+- Important: 1
+- Minor: 5
+- Files Affected: 5
+## 🏆 Code Quality
+[████████████████░░░░] 80% (Good)
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Functionality (1 issues)</strong></summary>
+
+### 1. The code seems to be implementing a memory creation feature with a dialog box. However, there are some potential issues with the functionality.
+📁 **File:** `menu.tsx:213`
+⚖️ **Severity:** 6/10
+🔍 **Description:** The `handleSubmit` function is not properly handling errors. It should be improved to handle errors in a more robust way.
+💡 **Solution:** Add try-catch blocks to handle errors in the `handleSubmit` function.
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (8 issues)</strong></summary>
+
+<details>
+<summary><strong>Unused imports (5 issues)</strong></summary>
+
+### 1. There are several unused imports in the code, such as `redirect` in `page.tsx` and `Switch` and `Label` in `queryinput.tsx`. These imports should be removed to declutter the code.
+📁 **File:** `page.tsx, queryinput.tsx:0`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Unused imports can make the code harder to read and maintain.
+💡 **Solution:** Remove unused imports.
+
+### 2. Some variables and functions are missing type annotations, such as the `initialSpaces` prop in `queryinput.tsx`. Adding type annotations can improve code readability and prevent type-related errors.
+📁 **File:** `queryinput.tsx:10`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Type annotations can help catch type-related errors at compile-time.
+💡 **Solution:** Add type annotations for variables and functions.
+
+### 3. The code is not well-organized. There are many functions and variables defined inside the `DialogContentContainer` component.
+📁 **File:** `menu.tsx:163`
+⚖️ **Severity:** 4/10
+🔍 **Description:** It would be better to separate the concerns of the component into smaller functions or utilities.
+💡 **Solution:** Extract some of the functions and variables into separate files or utilities.
+
+### 4. The import of `useEffect` from 'react' is not used in the code.
+📁 **File:** `packages/ui/shadcn/combobox.tsx:3`
+⚖️ **Severity:** 5/10
+🔍 **Description:** The import is not used anywhere in the code.
+💡 **Solution:** Remove the unused import.
+
+### 5. The type annotation for the `handleInputChange` function is missing.
+📁 **File:** `packages/ui/shadcn/combobox.tsx:41`
+⚖️ **Severity:** 5/10
+🔍 **Description:** The function is not annotated with a type.
+💡 **Solution:** Add the type annotation for the function.
+
+**Current Code:**
+```python
+const handleInputChange = (e: React.ChangeEvent<HTMLInputElement>) =>{
+```
+
+**Suggested Code:**
+```python
+const handleInputChange: React.ChangeEventHandler<HTMLInputElement> = (e) =>{
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (azure_ai/Meta-Llama-3-405B-Instruct)
+{"prompt_tokens": 15756, "completion_tokens": 1539, "total_tokens": 17295}
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_252/comments.json b/.experiments/code_review/llama-405b/no_eval/pr_252/comments.json
new file mode 100644
index 00000000..0637a088
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_252/comments.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_252/issues.json b/.experiments/code_review/llama-405b/no_eval/pr_252/issues.json
new file mode 100644
index 00000000..4a42628f
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_252/issues.json
@@ -0,0 +1,62 @@
+[
+  {
+    "topic": "Code Organization",
+    "comment": "The `WorkSummaryGenerator` class has multiple responsibilities, including generating work summaries, Twitter posts, and LinkedIn posts. Consider breaking this down into separate classes or functions for better organization and maintainability.",
+    "confidence": "important",
+    "reason": "Separation of Concerns (SoC) principle",
+    "solution": "Refactor the `WorkSummaryGenerator` class into separate classes or functions for each responsibility.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/reviewer/work_summarizer.py",
+    "start_line": 0,
+    "end_line": 0,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "The `generate_twitter_post` and `generate_linkedin_post` methods do not handle potential errors that may occur during the generation process. Consider adding try-except blocks to handle and log any exceptions.",
+    "confidence": "important",
+    "reason": "Error handling and logging",
+    "solution": "Add try-except blocks to handle and log any exceptions during the generation process.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/reviewer/work_summarizer.py",
+    "start_line": 58,
+    "end_line": 74,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 7
+  },
+  {
+    "topic": "Code Style",
+    "comment": "The `kaizen/llms/prompts/code_review_prompts.py` file has inconsistent indentation. Consider using a consistent number of spaces for indentation throughout the file.",
+    "confidence": "moderate",
+    "reason": "Code style and readability",
+    "solution": "Use a consistent number of spaces for indentation throughout the file.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/llms/prompts/code_review_prompts.py",
+    "start_line": 0,
+    "end_line": 0,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Code Duplication",
+    "comment": "The `generate_twitter_post` and `generate_linkedin_post` methods have similar code structures. Consider extracting a common method to avoid code duplication.",
+    "confidence": "moderate",
+    "reason": "Don't Repeat Yourself (DRY) principle",
+    "solution": "Extract a common method to avoid code duplication.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/reviewer/work_summarizer.py",
+    "start_line": 58,
+    "end_line": 74,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_252/review.md b/.experiments/code_review/llama-405b/no_eval/pr_252/review.md
new file mode 100644
index 00000000..732aa03c
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_252/review.md
@@ -0,0 +1,78 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/252
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 4
+- Critical: 0
+- Important: 2
+- Minor: 2
+- Files Affected: 2
+## 🏆 Code Quality
+[████████████████░░░░] 80% (Good)
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Code Organization (2 issues)</strong></summary>
+
+### 1. The `WorkSummaryGenerator` class has multiple responsibilities, including generating work summaries, Twitter posts, and LinkedIn posts. Consider breaking this down into separate classes or functions for better organization and maintainability.
+📁 **File:** `kaizen/reviewer/work_summarizer.py:0`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Separation of Concerns (SoC) principle
+💡 **Solution:** Refactor the `WorkSummaryGenerator` class into separate classes or functions for each responsibility.
+
+### 2. The `generate_twitter_post` and `generate_linkedin_post` methods do not handle potential errors that may occur during the generation process. Consider adding try-except blocks to handle and log any exceptions.
+📁 **File:** `kaizen/reviewer/work_summarizer.py:58`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Error handling and logging
+💡 **Solution:** Add try-except blocks to handle and log any exceptions during the generation process.
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (2 issues)</strong></summary>
+
+<details>
+<summary><strong>Code Style (2 issues)</strong></summary>
+
+### 1. The `kaizen/llms/prompts/code_review_prompts.py` file has inconsistent indentation. Consider using a consistent number of spaces for indentation throughout the file.
+📁 **File:** `kaizen/llms/prompts/code_review_prompts.py:0`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Code style and readability
+💡 **Solution:** Use a consistent number of spaces for indentation throughout the file.
+
+### 2. The `generate_twitter_post` and `generate_linkedin_post` methods have similar code structures. Consider extracting a common method to avoid code duplication.
+📁 **File:** `kaizen/reviewer/work_summarizer.py:58`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Don't Repeat Yourself (DRY) principle
+💡 **Solution:** Extract a common method to avoid code duplication.
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (azure_ai/Meta-Llama-3-405B-Instruct)
+{"prompt_tokens": 3881, "completion_tokens": 757, "total_tokens": 4638}
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_335/comments.json b/.experiments/code_review/llama-405b/no_eval/pr_335/comments.json
new file mode 100644
index 00000000..0637a088
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_335/comments.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_335/issues.json b/.experiments/code_review/llama-405b/no_eval/pr_335/issues.json
new file mode 100644
index 00000000..5b84d52d
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_335/issues.json
@@ -0,0 +1,47 @@
+[
+  {
+    "topic": "Import Statements",
+    "comment": "Unused import statements are present in the code.",
+    "confidence": "moderate",
+    "reason": "The import statements for 'output' and 'parser' are not used anywhere in the code.",
+    "solution": "Remove unused import statements to declutter the code.",
+    "actual_code": "from kaizen.helpers import output, parser",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 6,
+    "end_line": 6,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 5
+  },
+  {
+    "topic": "Function Definition",
+    "comment": "The function '__init__' has an unused parameter 'llm_provider'.",
+    "confidence": "high",
+    "reason": "The parameter 'llm_provider' is not used anywhere in the function.",
+    "solution": "Remove unused function parameters to simplify the code.",
+    "actual_code": "def __init__(self, llm_provider: LLMProvider):",
+    "fixed_code": "def __init__(self):",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 26,
+    "end_line": 26,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Variable Naming",
+    "comment": "Variable names are not following PEP 8 conventions.",
+    "confidence": "low",
+    "reason": "Variable names should be in lowercase with words separated by underscores.",
+    "solution": "Rename variables to follow PEP 8 conventions.",
+    "actual_code": "PULL_REQUEST_TITLE ={PULL_REQUEST_TITLE}",
+    "fixed_code": "pull_request_title ={pull_request_title}",
+    "file_name": "kaizen/llms/prompts/pr_desc_prompts.py",
+    "start_line": 27,
+    "end_line": 27,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 3
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_335/review.md b/.experiments/code_review/llama-405b/no_eval/pr_335/review.md
new file mode 100644
index 00000000..2a15ce48
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_335/review.md
@@ -0,0 +1,62 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/335
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 3
+- Critical: 0
+- Important: 0
+- Minor: 1
+- Files Affected: 2
+## 🏆 Code Quality
+[████████████████░░░░] 80% (Good)
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (2 issues)</strong></summary>
+
+<details>
+<summary><strong>Import Statements (1 issues)</strong></summary>
+
+### 1. Unused import statements are present in the code.
+📁 **File:** `kaizen/generator/pr_description.py:6`
+⚖️ **Severity:** 5/10
+🔍 **Description:** The import statements for 'output' and 'parser' are not used anywhere in the code.
+💡 **Solution:** Remove unused import statements to declutter the code.
+
+**Current Code:**
+```python
+from kaizen.helpers import output, parser
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (azure_ai/Meta-Llama-3-405B-Instruct)
+{"prompt_tokens": 6756, "completion_tokens": 533, "total_tokens": 7289}
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_400/comments.json b/.experiments/code_review/llama-405b/no_eval/pr_400/comments.json
new file mode 100644
index 00000000..33197e40
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_400/comments.json
@@ -0,0 +1,16 @@
+[
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "11",
+    "end_line": "11",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_400/issues.json b/.experiments/code_review/llama-405b/no_eval/pr_400/issues.json
new file mode 100644
index 00000000..318f90b8
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_400/issues.json
@@ -0,0 +1,316 @@
+[
+  {
+    "topic": "Error Handling",
+    "comment": "The `create_pr_description` function does not handle cases where `desc` or `original_desc` are not strings.",
+    "confidence": "important",
+    "reason": "The function should check the type of the input parameters and raise a meaningful error if they are not strings.",
+    "solution": "Add type checking for `desc` and `original_desc` and raise a `TypeError` if they are not strings.",
+    "actual_code": "",
+    "fixed_code": "if not isinstance(desc, str) or not isinstance(original_desc, str):\n    raise TypeError('desc and original_desc must be strings')",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
+    "start_line": 31,
+    "end_line": 31,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Performance",
+    "comment": "The `create_pr_description` function may have performance issues for large input strings.",
+    "confidence": "moderate",
+    "reason": "The function uses string concatenation, which can be inefficient for large strings.",
+    "solution": "Consider using a more efficient string concatenation method, such as using a list and joining the strings at the end.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
+    "start_line": 51,
+    "end_line": 51,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Code Organization",
+    "comment": "The test file has a mix of test cases and helper functions.",
+    "confidence": "moderate",
+    "reason": "It's better to separate test cases and helper functions into different files or modules.",
+    "solution": "Consider moving the helper functions to a separate file or module.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
+    "start_line": 1,
+    "end_line": 1,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 3
+  },
+  {
+    "topic": "Code Duplication",
+    "comment": "The `PR_COLLAPSIBLE_TEMPLATE` is duplicated in multiple tests.",
+    "confidence": "high",
+    "reason": "Code duplication can make maintenance harder.",
+    "solution": "Extract the template into a separate constant or function.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 0,
+    "end_line": 0,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Test Coverage",
+    "comment": "There are no tests for the `create_pr_review_text` function with an empty input.",
+    "confidence": "medium",
+    "reason": "Empty input can cause unexpected behavior.",
+    "solution": "Add a test for the `create_pr_review_text` function with an empty input.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 0,
+    "end_line": 0,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 5
+  },
+  {
+    "topic": "Code Readability",
+    "comment": "The `test_create_pr_review_text_mixed_reviews` test has a long and complex expected output.",
+    "confidence": "low",
+    "reason": "Long and complex expected outputs can make tests harder to understand.",
+    "solution": "Consider breaking the expected output into smaller, more manageable parts.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 96,
+    "end_line": 172,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 3
+  },
+  {
+    "topic": "Functionality",
+    "comment": "The code seems to be implementing the required functionality correctly.",
+    "confidence": "moderate",
+    "reason": "The code is using the correct libraries and functions to achieve the desired outcome.",
+    "solution": "",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 1,
+    "end_line": 247,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 5
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "The code is missing error handling for potential exceptions that may occur during file operations.",
+    "confidence": "high",
+    "reason": "The code is not handling potential exceptions that may occur during file operations, which can lead to unexpected behavior.",
+    "solution": "Add try-except blocks to handle potential exceptions during file operations.",
+    "actual_code": "",
+    "fixed_code": "try:\n    with open(file_path, 'r') as f:\n        content = f.read()\nexcept FileNotFoundError:\n    print(f'File{file_path}not found.')",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 40,
+    "end_line": 45,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Code Quality",
+    "comment": "The code has some redundant comments and docstrings that can be removed.",
+    "confidence": "low",
+    "reason": "The code has some redundant comments and docstrings that are not providing any additional information.",
+    "solution": "Remove redundant comments and docstrings.",
+    "actual_code": "# Correct implementation of get_parent_folder()",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py",
+    "start_line": 8,
+    "end_line": 8,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 2
+  },
+  {
+    "topic": "Import Statements",
+    "comment": "Unused import statement",
+    "confidence": "moderate",
+    "reason": "The import statement 'from kaizen.helpers.output import get_web_html' is not used in the code.",
+    "solution": "Remove the unused import statement.",
+    "actual_code": "from kaizen.helpers.output import get_web_html",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 4,
+    "end_line": 4,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 5
+  },
+  {
+    "topic": "Function Definition",
+    "comment": "Function 'test_get_web_html_normal_cases' is too long and complex.",
+    "confidence": "high",
+    "reason": "The function has too many lines of code and is difficult to understand.",
+    "solution": "Break down the function into smaller, more manageable functions.",
+    "actual_code": "async def test_get_web_html_normal_cases(mock_get_html, mock_nest_asyncio, html_content, expected_output):",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 76,
+    "end_line": 103,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Error handling is missing in function 'test_get_web_html_invalid_url'.",
+    "confidence": "high",
+    "reason": "The function does not handle potential errors that may occur.",
+    "solution": "Add try-except blocks to handle potential errors.",
+    "actual_code": "async def test_get_web_html_invalid_url(mock_get_html, mock_nest_asyncio):",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 85,
+    "end_line": 91,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "Code Duplication",
+    "comment": "Code duplication in functions 'test_get_web_html_normal_cases' and 'test_get_web_html_invalid_url'.",
+    "confidence": "moderate",
+    "reason": "The functions have similar code that can be extracted into a separate function.",
+    "solution": "Extract the common code into a separate function.",
+    "actual_code": "mock_get_html.return_value = html_content",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 77,
+    "end_line": 77,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Import organization",
+    "comment": "Imports are not organized alphabetically.",
+    "confidence": "moderate",
+    "reason": "Following PEP 8 guidelines for import organization improves readability.",
+    "solution": "Organize imports alphabetically.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/unit_test.py",
+    "start_line": 1,
+    "end_line": 10,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 3
+  },
+  {
+    "topic": "Docstrings",
+    "comment": "Missing docstrings for classes and methods.",
+    "confidence": "important",
+    "reason": "Docstrings provide essential documentation for users and maintainers.",
+    "solution": "Add docstrings to classes and methods.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/unit_test.py",
+    "start_line": 11,
+    "end_line": 275,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Type hints",
+    "comment": "Missing type hints for method parameters and return types.",
+    "confidence": "important",
+    "reason": "Type hints improve code readability and enable static type checking.",
+    "solution": "Add type hints for method parameters and return types.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/unit_test.py",
+    "start_line": 11,
+    "end_line": 275,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Error handling",
+    "comment": "Insufficient error handling in methods.",
+    "confidence": "important",
+    "reason": "Proper error handling ensures the program remains stable and provides useful error messages.",
+    "solution": "Implement try-except blocks to handle potential errors.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/unit_test.py",
+    "start_line": 11,
+    "end_line": 275,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Code organization",
+    "comment": "Some methods are too long and complex.",
+    "confidence": "moderate",
+    "reason": "Breaking down long methods into smaller ones improves readability and maintainability.",
+    "solution": "Refactor long methods into smaller, more focused ones.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/unit_test.py",
+    "start_line": 11,
+    "end_line": 275,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Logging Configuration",
+    "comment": "The logging configuration is not properly set up.",
+    "confidence": "important",
+    "reason": "The logging level is set to ERROR for all loggers, but the LOGLEVEL environment variable is set to INFO.",
+    "solution": "Set the logging level consistently throughout the application.",
+    "actual_code": "set_all_loggers_to_ERROR()",
+    "fixed_code": "set_all_loggers_to_INFO()",
+    "file_name": "kaizen/llms/provider.py",
+    "start_line": 13,
+    "end_line": 28,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Code Organization",
+    "comment": "The code is not properly organized.",
+    "confidence": "moderate",
+    "reason": "The set_all_loggers_to_ERROR function is defined in the middle of the file.",
+    "solution": "Move the function definition to the top of the file.",
+    "actual_code": "def set_all_loggers_to_ERROR():",
+    "fixed_code": "def set_all_loggers_to_ERROR():\n    # ...",
+    "file_name": "kaizen/llms/provider.py",
+    "start_line": 13,
+    "end_line": 28,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 4
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "11",
+    "end_line": "11",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_400/review.md b/.experiments/code_review/llama-405b/no_eval/pr_400/review.md
new file mode 100644
index 00000000..c168419b
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_400/review.md
@@ -0,0 +1,197 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/400
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 21
+- Critical: 1
+- Important: 5
+- Minor: 8
+- Files Affected: 8
+## 🏆 Code Quality
+[████████████████░░░░] 80% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Configuration (1 issues)</strong></summary>
+
+### 1. Changes made to sensitive file
+📁 **File:** `config.json:11`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to config.json, which needs review
+💡 **Solution:** NA
+
+</details>
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Error Handling (5 issues)</strong></summary>
+
+### 1. The `create_pr_description` function does not handle cases where `desc` or `original_desc` are not strings.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_description.py:31`
+⚖️ **Severity:** 6/10
+🔍 **Description:** The function should check the type of the input parameters and raise a meaningful error if they are not strings.
+💡 **Solution:** Add type checking for `desc` and `original_desc` and raise a `TypeError` if they are not strings.
+
+**Current Code:**
+```python
+
+```
+
+**Suggested Code:**
+```python
+if not isinstance(desc, str) or not isinstance(original_desc, str):
+    raise TypeError('desc and original_desc must be strings')
+```
+
+### 2. Missing docstrings for classes and methods.
+📁 **File:** `kaizen/generator/unit_test.py:11`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Docstrings provide essential documentation for users and maintainers.
+💡 **Solution:** Add docstrings to classes and methods.
+
+### 3. Missing type hints for method parameters and return types.
+📁 **File:** `kaizen/generator/unit_test.py:11`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Type hints improve code readability and enable static type checking.
+💡 **Solution:** Add type hints for method parameters and return types.
+
+### 4. Insufficient error handling in methods.
+📁 **File:** `kaizen/generator/unit_test.py:11`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Proper error handling ensures the program remains stable and provides useful error messages.
+💡 **Solution:** Implement try-except blocks to handle potential errors.
+
+### 5. The logging configuration is not properly set up.
+📁 **File:** `kaizen/llms/provider.py:13`
+⚖️ **Severity:** 6/10
+🔍 **Description:** The logging level is set to ERROR for all loggers, but the LOGLEVEL environment variable is set to INFO.
+💡 **Solution:** Set the logging level consistently throughout the application.
+
+**Current Code:**
+```python
+set_all_loggers_to_ERROR()
+```
+
+**Suggested Code:**
+```python
+set_all_loggers_to_INFO()
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (10 issues)</strong></summary>
+
+<details>
+<summary><strong>Performance (8 issues)</strong></summary>
+
+### 1. The `create_pr_description` function may have performance issues for large input strings.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_description.py:51`
+⚖️ **Severity:** 4/10
+🔍 **Description:** The function uses string concatenation, which can be inefficient for large strings.
+💡 **Solution:** Consider using a more efficient string concatenation method, such as using a list and joining the strings at the end.
+
+### 2. The test file has a mix of test cases and helper functions.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_description.py:1`
+⚖️ **Severity:** 3/10
+🔍 **Description:** It's better to separate test cases and helper functions into different files or modules.
+💡 **Solution:** Consider moving the helper functions to a separate file or module.
+
+### 3. The code seems to be implementing the required functionality correctly.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_test_files.py:1`
+⚖️ **Severity:** 5/10
+🔍 **Description:** The code is using the correct libraries and functions to achieve the desired outcome.
+💡 **Solution:** 
+
+### 4. Unused import statement
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_web_html.py:4`
+⚖️ **Severity:** 5/10
+🔍 **Description:** The import statement 'from kaizen.helpers.output import get_web_html' is not used in the code.
+💡 **Solution:** Remove the unused import statement.
+
+**Current Code:**
+```python
+from kaizen.helpers.output import get_web_html
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 5. Code duplication in functions 'test_get_web_html_normal_cases' and 'test_get_web_html_invalid_url'.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_web_html.py:77`
+⚖️ **Severity:** 6/10
+🔍 **Description:** The functions have similar code that can be extracted into a separate function.
+💡 **Solution:** Extract the common code into a separate function.
+
+**Current Code:**
+```python
+mock_get_html.return_value = html_content
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 6. Imports are not organized alphabetically.
+📁 **File:** `kaizen/generator/unit_test.py:1`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Following PEP 8 guidelines for import organization improves readability.
+💡 **Solution:** Organize imports alphabetically.
+
+### 7. Some methods are too long and complex.
+📁 **File:** `kaizen/generator/unit_test.py:11`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Breaking down long methods into smaller ones improves readability and maintainability.
+💡 **Solution:** Refactor long methods into smaller, more focused ones.
+
+### 8. The code is not properly organized.
+📁 **File:** `kaizen/llms/provider.py:13`
+⚖️ **Severity:** 4/10
+🔍 **Description:** The set_all_loggers_to_ERROR function is defined in the middle of the file.
+💡 **Solution:** Move the function definition to the top of the file.
+
+**Current Code:**
+```python
+def set_all_loggers_to_ERROR():
+```
+
+**Suggested Code:**
+```python
+def set_all_loggers_to_ERROR():
+    # ...
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (azure_ai/Meta-Llama-3-405B-Instruct)
+{"prompt_tokens": 36841, "completion_tokens": 3640, "total_tokens": 40481}
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_440/comments.json b/.experiments/code_review/llama-405b/no_eval/pr_440/comments.json
new file mode 100644
index 00000000..0637a088
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_440/comments.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_440/issues.json b/.experiments/code_review/llama-405b/no_eval/pr_440/issues.json
new file mode 100644
index 00000000..172e9fbf
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_440/issues.json
@@ -0,0 +1,47 @@
+[
+  {
+    "topic": "Unnecessary Comment",
+    "comment": "The comment 'TODO: DONT PUSH DUPLICATE' is unnecessary and should be removed.",
+    "confidence": "moderate",
+    "reason": "The comment does not provide any useful information and is not relevant to the code.",
+    "solution": "Remove the comment.",
+    "actual_code": "# TODO: DONT PUSH DUPLICATE",
+    "fixed_code": "",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 157,
+    "end_line": 157,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 2
+  },
+  {
+    "topic": "Dependency Update",
+    "comment": "The dependency 'llama-index-core' has been updated to version '0.10.65'.",
+    "confidence": "important",
+    "reason": "The updated version may include security patches or bug fixes.",
+    "solution": "Review the changelog for the updated version to ensure compatibility with the current codebase.",
+    "actual_code": "llama-index-core = \"0.10.65\"",
+    "fixed_code": "",
+    "file_name": "pyproject.toml",
+    "start_line": 27,
+    "end_line": 27,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 5
+  },
+  {
+    "topic": "Removed Dependencies",
+    "comment": "The dependencies 'llama-index-llms-openai' and 'llama-index-core' (version '^0.10.47') have been removed.",
+    "confidence": "important",
+    "reason": "The removed dependencies may be required by other parts of the codebase.",
+    "solution": "Review the codebase to ensure that the removed dependencies are not required.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "pyproject.toml",
+    "start_line": 27,
+    "end_line": 28,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 6
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_440/review.md b/.experiments/code_review/llama-405b/no_eval/pr_440/review.md
new file mode 100644
index 00000000..f509ad53
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_440/review.md
@@ -0,0 +1,92 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/440
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 3
+- Critical: 0
+- Important: 2
+- Minor: 1
+- Files Affected: 2
+## 🏆 Code Quality
+[████████████████░░░░] 80% (Good)
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Dependency Update (2 issues)</strong></summary>
+
+### 1. The dependency 'llama-index-core' has been updated to version '0.10.65'.
+📁 **File:** `pyproject.toml:27`
+⚖️ **Severity:** 5/10
+🔍 **Description:** The updated version may include security patches or bug fixes.
+💡 **Solution:** Review the changelog for the updated version to ensure compatibility with the current codebase.
+
+**Current Code:**
+```python
+llama-index-core = "0.10.65"
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 2. The dependencies 'llama-index-llms-openai' and 'llama-index-core' (version '^0.10.47') have been removed.
+📁 **File:** `pyproject.toml:27`
+⚖️ **Severity:** 6/10
+🔍 **Description:** The removed dependencies may be required by other parts of the codebase.
+💡 **Solution:** Review the codebase to ensure that the removed dependencies are not required.
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (1 issues)</strong></summary>
+
+<details>
+<summary><strong>Unnecessary Comment (1 issues)</strong></summary>
+
+### 1. The comment 'TODO: DONT PUSH DUPLICATE' is unnecessary and should be removed.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:157`
+⚖️ **Severity:** 2/10
+🔍 **Description:** The comment does not provide any useful information and is not relevant to the code.
+💡 **Solution:** Remove the comment.
+
+**Current Code:**
+```python
+# TODO: DONT PUSH DUPLICATE
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (azure_ai/Meta-Llama-3-405B-Instruct)
+{"prompt_tokens": 1362, "completion_tokens": 562, "total_tokens": 1924}
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_476/comments.json b/.experiments/code_review/llama-405b/no_eval/pr_476/comments.json
new file mode 100644
index 00000000..a9d40eac
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_476/comments.json
@@ -0,0 +1,16 @@
+[
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_476/issues.json b/.experiments/code_review/llama-405b/no_eval/pr_476/issues.json
new file mode 100644
index 00000000..a394f97b
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_476/issues.json
@@ -0,0 +1,61 @@
+[
+  {
+    "topic": "Error Handling",
+    "comment": "Broad exception handling can mask bugs and make debugging difficult.",
+    "confidence": "important",
+    "reason": "The `except Exception` block in `github_app/github_helper/pull_requests.py` (line 140) catches all exceptions, which can make it challenging to identify and fix specific issues.",
+    "solution": "Catch specific exceptions that can occur during the execution of the code, and provide meaningful error messages to aid in debugging.",
+    "actual_code": "except Exception:",
+    "fixed_code": "except requests.exceptions.RequestException as e:",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 140,
+    "end_line": 141,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Code Organization",
+    "comment": "The `sort_files` function is not necessary and can be replaced with a built-in sorting function.",
+    "confidence": "moderate",
+    "reason": "The `sort_files` function in `github_app/github_helper/pull_requests.py` (line 184) is not necessary and can be replaced with the built-in `sorted` function.",
+    "solution": "Use the built-in `sorted` function to sort the files, which is more efficient and Pythonic.",
+    "actual_code": "def sort_files(files):",
+    "fixed_code": "sorted_files = sorted(files, key=lambda x: x['filename'])",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 184,
+    "end_line": 194,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Code Quality",
+    "comment": "The `generate_tests` function is not necessary and can be replaced with a list comprehension.",
+    "confidence": "moderate",
+    "reason": "The `generate_tests` function in `github_app/github_helper/pull_requests.py` (line 199) is not necessary and can be replaced with a list comprehension.",
+    "solution": "Use a list comprehension to generate the tests, which is more efficient and Pythonic.",
+    "actual_code": "def generate_tests(pr_files):",
+    "fixed_code": "tests =[f['filename'] for f in pr_files]",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 199,
+    "end_line": 200,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_476/review.md b/.experiments/code_review/llama-405b/no_eval/pr_476/review.md
new file mode 100644
index 00000000..8f093042
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_476/review.md
@@ -0,0 +1,115 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/476
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 4
+- Critical: 1
+- Important: 1
+- Minor: 2
+- Files Affected: 2
+## 🏆 Code Quality
+[████████████████░░░░] 80% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Configuration (1 issues)</strong></summary>
+
+### 1. Changes made to sensitive file
+📁 **File:** `config.json:1`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to config.json, which needs review
+💡 **Solution:** NA
+
+</details>
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Error Handling (1 issues)</strong></summary>
+
+### 1. Broad exception handling can mask bugs and make debugging difficult.
+📁 **File:** `github_app/github_helper/pull_requests.py:140`
+⚖️ **Severity:** 6/10
+🔍 **Description:** The `except Exception` block in `github_app/github_helper/pull_requests.py` (line 140) catches all exceptions, which can make it challenging to identify and fix specific issues.
+💡 **Solution:** Catch specific exceptions that can occur during the execution of the code, and provide meaningful error messages to aid in debugging.
+
+**Current Code:**
+```python
+except Exception:
+```
+
+**Suggested Code:**
+```python
+except requests.exceptions.RequestException as e:
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (2 issues)</strong></summary>
+
+<details>
+<summary><strong>Code Organization (2 issues)</strong></summary>
+
+### 1. The `sort_files` function is not necessary and can be replaced with a built-in sorting function.
+📁 **File:** `github_app/github_helper/pull_requests.py:184`
+⚖️ **Severity:** 4/10
+🔍 **Description:** The `sort_files` function in `github_app/github_helper/pull_requests.py` (line 184) is not necessary and can be replaced with the built-in `sorted` function.
+💡 **Solution:** Use the built-in `sorted` function to sort the files, which is more efficient and Pythonic.
+
+**Current Code:**
+```python
+def sort_files(files):
+```
+
+**Suggested Code:**
+```python
+sorted_files = sorted(files, key=lambda x: x['filename'])
+```
+
+### 2. The `generate_tests` function is not necessary and can be replaced with a list comprehension.
+📁 **File:** `github_app/github_helper/pull_requests.py:199`
+⚖️ **Severity:** 4/10
+🔍 **Description:** The `generate_tests` function in `github_app/github_helper/pull_requests.py` (line 199) is not necessary and can be replaced with a list comprehension.
+💡 **Solution:** Use a list comprehension to generate the tests, which is more efficient and Pythonic.
+
+**Current Code:**
+```python
+def generate_tests(pr_files):
+```
+
+**Suggested Code:**
+```python
+tests =[f['filename'] for f in pr_files]
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (azure_ai/Meta-Llama-3-405B-Instruct)
+{"prompt_tokens": 4006, "completion_tokens": 655, "total_tokens": 4661}
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_5/comments.json b/.experiments/code_review/llama-405b/no_eval/pr_5/comments.json
new file mode 100644
index 00000000..ce61c328
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_5/comments.json
@@ -0,0 +1,47 @@
+[
+  {
+    "topic": "API Call Failure",
+    "comment": "API call failure without retry mechanism.",
+    "confidence": "critical",
+    "reason": "The API call may fail without a retry mechanism, causing the program to crash.",
+    "solution": "Implement a retry mechanism for the API call.",
+    "actual_code": "response = completion(model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages)",
+    "fixed_code": "import time\ntry:\n    response = completion(model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages)\nexcept Exception as e:\n    print(f\"API call failed:{e}\")\n    time.sleep(1)\n    # retry the API call",
+    "file_name": "main.py",
+    "start_line": 65,
+    "end_line": 65,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Silent Failure",
+    "comment": "Silent failure without logging.",
+    "confidence": "critical",
+    "reason": "The program will fail silently without logging any errors.",
+    "solution": "Implement logging for errors.",
+    "actual_code": "except json.JSONDecodeError:\n    result ={key: \"\" for key in[\"feedback\", \"review\", \"should_interview\", \"rating\", \"input_tokens\", \"output_tokens\"]}",
+    "fixed_code": "import logging\nexcept json.JSONDecodeError as e:\n    logging.error(f\"JSON decode error:{e}\")\n    result ={key: \"\" for key in[\"feedback\", \"review\", \"should_interview\", \"rating\", \"input_tokens\", \"output_tokens\"]}",
+    "file_name": "main.py",
+    "start_line": 82,
+    "end_line": 84,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Division by Zero",
+    "comment": "Division by zero potential.",
+    "confidence": "critical",
+    "reason": "The code may divide by zero, causing a runtime error.",
+    "solution": "Add a check to prevent division by zero.",
+    "actual_code": "total_tokens = total_input_tokens + total_output_tokens",
+    "fixed_code": "if total_tokens == 0:\n    total_tokens = 1\n    print(\"Warning: total tokens is zero\")",
+    "file_name": "main.py",
+    "start_line": 156,
+    "end_line": 156,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 8
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_5/issues.json b/.experiments/code_review/llama-405b/no_eval/pr_5/issues.json
new file mode 100644
index 00000000..a33e8f2e
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_5/issues.json
@@ -0,0 +1,107 @@
+[
+  {
+    "topic": "Unused Import",
+    "comment": "Unused import 'random' should be removed.",
+    "confidence": "moderate",
+    "reason": "The import 'random' is not used anywhere in the code.",
+    "solution": "Remove the line 'import random'.",
+    "actual_code": "import random",
+    "fixed_code": "",
+    "file_name": "main.py",
+    "start_line": 8,
+    "end_line": 8,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 3
+  },
+  {
+    "topic": "API Call Failure",
+    "comment": "API call failure without retry mechanism.",
+    "confidence": "critical",
+    "reason": "The API call may fail without a retry mechanism, causing the program to crash.",
+    "solution": "Implement a retry mechanism for the API call.",
+    "actual_code": "response = completion(model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages)",
+    "fixed_code": "import time\ntry:\n    response = completion(model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages)\nexcept Exception as e:\n    print(f\"API call failed:{e}\")\n    time.sleep(1)\n    # retry the API call",
+    "file_name": "main.py",
+    "start_line": 65,
+    "end_line": 65,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Silent Failure",
+    "comment": "Silent failure without logging.",
+    "confidence": "critical",
+    "reason": "The program will fail silently without logging any errors.",
+    "solution": "Implement logging for errors.",
+    "actual_code": "except json.JSONDecodeError:\n    result ={key: \"\" for key in[\"feedback\", \"review\", \"should_interview\", \"rating\", \"input_tokens\", \"output_tokens\"]}",
+    "fixed_code": "import logging\nexcept json.JSONDecodeError as e:\n    logging.error(f\"JSON decode error:{e}\")\n    result ={key: \"\" for key in[\"feedback\", \"review\", \"should_interview\", \"rating\", \"input_tokens\", \"output_tokens\"]}",
+    "file_name": "main.py",
+    "start_line": 82,
+    "end_line": 84,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Inefficient Progress Printing",
+    "comment": "Inefficient way to print progress.",
+    "confidence": "important",
+    "reason": "The progress printing is inefficient and may cause performance issues.",
+    "solution": "Use a more efficient way to print progress, such as using a progress bar library.",
+    "actual_code": "print(f\"\\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}\", end=\"\", flush=True)",
+    "fixed_code": "from tqdm import tqdm\nwith tqdm(total=total, desc=\"Processing applicants\") as pbar:\n    for index, row in df.iterrows():\n        # process applicant\n        pbar.update(1)",
+    "file_name": "main.py",
+    "start_line": 121,
+    "end_line": 121,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 5
+  },
+  {
+    "topic": "Redundant Code",
+    "comment": "Redundant code.",
+    "confidence": "moderate",
+    "reason": "The code is redundant and can be removed.",
+    "solution": "Remove the redundant code.",
+    "actual_code": "if len(df) == 0:\n    return",
+    "fixed_code": "",
+    "file_name": "main.py",
+    "start_line": 141,
+    "end_line": 142,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 3
+  },
+  {
+    "topic": "Division by Zero",
+    "comment": "Division by zero potential.",
+    "confidence": "critical",
+    "reason": "The code may divide by zero, causing a runtime error.",
+    "solution": "Add a check to prevent division by zero.",
+    "actual_code": "total_tokens = total_input_tokens + total_output_tokens",
+    "fixed_code": "if total_tokens == 0:\n    total_tokens = 1\n    print(\"Warning: total tokens is zero\")",
+    "file_name": "main.py",
+    "start_line": 156,
+    "end_line": 156,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "File Not Found",
+    "comment": "No error handling for file not found.",
+    "confidence": "important",
+    "reason": "The code does not handle the case where the file is not found.",
+    "solution": "Add error handling for file not found.",
+    "actual_code": "main(input_file)",
+    "fixed_code": "try:\n    main(input_file)\nexcept FileNotFoundError:\n    print(f\"Error: file '{input_file}' not found\")",
+    "file_name": "main.py",
+    "start_line": 174,
+    "end_line": 174,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 5
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_5/review.md b/.experiments/code_review/llama-405b/no_eval/pr_5/review.md
new file mode 100644
index 00000000..adfbba95
--- /dev/null
+++ b/.experiments/code_review/llama-405b/no_eval/pr_5/review.md
@@ -0,0 +1,193 @@
+PR URL: https://github.com/sauravpanda/applicant-screening/pull/5
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 7
+- Critical: 3
+- Important: 2
+- Minor: 2
+- Files Affected: 1
+## 🏆 Code Quality
+[██████████████░░░░░░] 70% (Fair)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>API Call Failure (3 issues)</strong></summary>
+
+### 1. API call failure without retry mechanism.
+📁 **File:** `main.py:65`
+⚖️ **Severity:** 8/10
+🔍 **Description:** The API call may fail without a retry mechanism, causing the program to crash.
+💡 **Solution:** Implement a retry mechanism for the API call.
+
+**Current Code:**
+```python
+response = completion(model=os.environ.get("model", "anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1"), messages=messages)
+```
+
+**Suggested Code:**
+```python
+import time
+try:
+    response = completion(model=os.environ.get("model", "anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1"), messages=messages)
+except Exception as e:
+    print(f"API call failed:{e}")
+    time.sleep(1)
+    # retry the API call
+```
+
+### 2. Silent failure without logging.
+📁 **File:** `main.py:82`
+⚖️ **Severity:** 8/10
+🔍 **Description:** The program will fail silently without logging any errors.
+💡 **Solution:** Implement logging for errors.
+
+**Current Code:**
+```python
+except json.JSONDecodeError:
+    result ={key: "" for key in["feedback", "review", "should_interview", "rating", "input_tokens", "output_tokens"]}
+```
+
+**Suggested Code:**
+```python
+import logging
+except json.JSONDecodeError as e:
+    logging.error(f"JSON decode error:{e}")
+    result ={key: "" for key in["feedback", "review", "should_interview", "rating", "input_tokens", "output_tokens"]}
+```
+
+### 3. Division by zero potential.
+📁 **File:** `main.py:156`
+⚖️ **Severity:** 8/10
+🔍 **Description:** The code may divide by zero, causing a runtime error.
+💡 **Solution:** Add a check to prevent division by zero.
+
+**Current Code:**
+```python
+total_tokens = total_input_tokens + total_output_tokens
+```
+
+**Suggested Code:**
+```python
+if total_tokens == 0:
+    total_tokens = 1
+    print("Warning: total tokens is zero")
+```
+
+</details>
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Inefficient Progress Printing (2 issues)</strong></summary>
+
+### 1. Inefficient way to print progress.
+📁 **File:** `main.py:121`
+⚖️ **Severity:** 5/10
+🔍 **Description:** The progress printing is inefficient and may cause performance issues.
+💡 **Solution:** Use a more efficient way to print progress, such as using a progress bar library.
+
+**Current Code:**
+```python
+print(f"\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}", end="", flush=True)
+```
+
+**Suggested Code:**
+```python
+from tqdm import tqdm
+with tqdm(total=total, desc="Processing applicants") as pbar:
+    for index, row in df.iterrows():
+        # process applicant
+        pbar.update(1)
+```
+
+### 2. No error handling for file not found.
+📁 **File:** `main.py:174`
+⚖️ **Severity:** 5/10
+🔍 **Description:** The code does not handle the case where the file is not found.
+💡 **Solution:** Add error handling for file not found.
+
+**Current Code:**
+```python
+main(input_file)
+```
+
+**Suggested Code:**
+```python
+try:
+    main(input_file)
+except FileNotFoundError:
+    print(f"Error: file '{input_file}' not found")
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (2 issues)</strong></summary>
+
+<details>
+<summary><strong>Unused Import (2 issues)</strong></summary>
+
+### 1. Unused import 'random' should be removed.
+📁 **File:** `main.py:8`
+⚖️ **Severity:** 3/10
+🔍 **Description:** The import 'random' is not used anywhere in the code.
+💡 **Solution:** Remove the line 'import random'.
+
+**Current Code:**
+```python
+import random
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 2. Redundant code.
+📁 **File:** `main.py:141`
+⚖️ **Severity:** 3/10
+🔍 **Description:** The code is redundant and can be removed.
+💡 **Solution:** Remove the redundant code.
+
+**Current Code:**
+```python
+if len(df) == 0:
+    return
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (azure_ai/Meta-Llama-3-405B-Instruct)
+{"prompt_tokens": 6128, "completion_tokens": 1277, "total_tokens": 7405}
\ No newline at end of file
diff --git a/.experiments/code_review/main.py b/.experiments/code_review/main.py
index 28375d34..c04c609f 100644
--- a/.experiments/code_review/main.py
+++ b/.experiments/code_review/main.py
@@ -35,7 +35,7 @@ def process_pr(pr_url, reeval_response=False):
     diff_text = get_diff_text(pr_diff, "")
     pr_files = get_pr_files(pr_files, "")
 
-    reviewer = CodeReviewer(llm_provider=LLMProvider())
+    reviewer = CodeReviewer(llm_provider=LLMProvider(), default_model="best")
     review_data = reviewer.review_pull_request(
         diff_text=diff_text,
         pull_request_title=pr_title,
@@ -51,16 +51,23 @@ def process_pr(pr_url, reeval_response=False):
     review_desc = create_pr_review_text(
         review_data.issues, code_quality=review_data.code_quality
     )
+    review_desc = f"PR URL: {pr_url}\n\n" + review_desc
+    review_desc += f"\n\n----- Cost Usage ({review_data.model_name})\n" + json.dumps(
+        review_data.usage
+    )
     comments, topics = create_review_comments(review_data.topics)
     logger.info(f"Model: {review_data.model_name}\nUsage: {review_data.usage}")
     logger.info(f"Completed processing PR: {pr_url}")
-    return review_desc, comments, topics
+    return review_desc, comments, review_data.issues
 
 
-def save_review(pr_number, review_desc, comments, topics, folder):
+def save_review(pr_number, review_desc, comments, issues, folder):
+    folder = os.path.join(folder, f"pr_{pr_number}")
     logger.info(f"Saving review for PR {pr_number} in {folder}")
-    review_file = os.path.join(folder, f"pr_{pr_number}_review.md")
-    comments_file = os.path.join(folder, f"pr_{pr_number}_comments.json")
+    os.makedirs(folder, exist_ok=True)
+    review_file = os.path.join(folder, "review.md")
+    comments_file = os.path.join(folder, "comments.json")
+    issues_file = os.path.join(folder, "issues.json")
 
     with open(review_file, "w") as f:
         f.write(review_desc)
@@ -68,6 +75,9 @@ def save_review(pr_number, review_desc, comments, topics, folder):
     with open(comments_file, "w") as f:
         json.dump(comments, f, indent=2)
 
+    with open(issues_file, "w") as f:
+        json.dump(issues, f, indent=2)
+
     logger.info(f"Saved review files for PR {pr_number}")
 
 
@@ -88,12 +98,12 @@ def main(pr_urls):
         logger.info(f"Starting to process PR {pr_number}")
 
         # Without re-evaluation
-        review_desc, comments, topics = process_pr(pr_url, reeval_response=False)
-        save_review(pr_number, review_desc, comments, topics, no_eval_folder)
+        review_desc, comments, issues = process_pr(pr_url, reeval_response=False)
+        save_review(pr_number, review_desc, comments, issues, no_eval_folder)
 
-        # With re-evaluation
-        review_desc, comments, topics = process_pr(pr_url, reeval_response=True)
-        save_review(pr_number, review_desc, comments, topics, with_eval_folder)
+        # # With re-evaluation
+        # review_desc, comments, topics = process_pr(pr_url, reeval_response=True)
+        # save_review(pr_number, review_desc, comments, topics, with_eval_folder)
 
         logger.info(f"Completed processing PR {pr_number}")
 
@@ -102,9 +112,15 @@ def main(pr_urls):
 
 if __name__ == "__main__":
     pr_urls = [
+        "https://github.com/sauravpanda/applicant-screening/pull/5",
         "https://github.com/Cloud-Code-AI/kaizen/pull/335",
         "https://github.com/Cloud-Code-AI/kaizen/pull/440",
         "https://github.com/Cloud-Code-AI/kaizen/pull/222",
+        "https://github.com/Cloud-Code-AI/kaizen/pull/476",
+        "https://github.com/Cloud-Code-AI/kaizen/pull/252",
+        "https://github.com/Cloud-Code-AI/kaizen/pull/400",
+        # "https://github.com/supermemoryai/supermemory/pull/164",
+        "https://github.com/supermemoryai/supermemory/pull/232",
         # Add more PR URLs here
     ]
     main(pr_urls)
diff --git a/.experiments/code_review/print_info.py b/.experiments/code_review/print_info.py
new file mode 100644
index 00000000..1e693c37
--- /dev/null
+++ b/.experiments/code_review/print_info.py
@@ -0,0 +1,40 @@
+import json
+from pathlib import Path
+
+
+def print_issues_for_pr(pr_number):
+    base_path = Path(".experiments/code_review")
+    models = [
+        "gpt-4o",
+        "gpt-4o-mini",
+        "gpt-4o-try2",
+        "haiku",
+        "llama-405b",
+        "sonnet-3.5",
+    ]
+
+    for model in models:
+        file_path = base_path / model / "no_eval" / f"pr_{pr_number}" / "issues.json"
+
+        if file_path.exists():
+            print(f"\nModel: {model}")
+            print(f"File: {file_path}")
+
+            try:
+                with open(file_path, "r") as file:
+                    data = json.load(file)
+                    formatted_json = json.dumps(data, indent=2)
+                    print("Content:")
+                    print(formatted_json)
+            except json.JSONDecodeError:
+                print("Error: Invalid JSON file")
+            except Exception as e:
+                print(f"Error reading file: {str(e)}")
+        else:
+            print(f"\nModel: {model}")
+            print(f"File not found: {file_path}")
+
+
+# Example usage
+pr_number = 476
+print_issues_for_pr(pr_number)
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_222/comments.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_222/comments.json
new file mode 100644
index 00000000..7edfffbe
--- /dev/null
+++ b/.experiments/code_review/sonnet-3.5/no_eval/pr_222/comments.json
@@ -0,0 +1,72 @@
+[
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Docker",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to Dockerfile, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_name": "Dockerfile",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Docker",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to docker-compose.yml, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "15",
+    "end_line": "15",
+    "side": "RIGHT",
+    "file_name": "docker-compose.yml",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Version Control",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to .gitignore, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "164",
+    "end_line": "164",
+    "side": "RIGHT",
+    "file_name": ".gitignore",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Database",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to db_setup/init.sql, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "db_setup/init.sql",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_222/issues.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_222/issues.json
new file mode 100644
index 00000000..13cebd2d
--- /dev/null
+++ b/.experiments/code_review/sonnet-3.5/no_eval/pr_222/issues.json
@@ -0,0 +1,312 @@
+[
+  {
+    "topic": "Docker Configuration",
+    "comment": "Consider using multi-stage builds to reduce the final image size",
+    "confidence": "moderate",
+    "reason": "Multi-stage builds can significantly reduce the size of the final Docker image by excluding build dependencies",
+    "solution": "Implement a multi-stage build in the Dockerfile",
+    "actual_code": "RUN apt-get update && apt-get install -y \\\n    git \\\n    build-essential \\\n    && rm -rf /var/lib/apt/lists/*",
+    "fixed_code": "FROM python:3.9 AS builder\n\nRUN apt-get update && apt-get install -y \\\n    git \\\n    build-essential\n\n# ... (build steps)\n\nFROM python:3.9-slim\n\nCOPY --from=builder /app /app\n\n# ... (runtime configuration)",
+    "file_name": "Dockerfile",
+    "start_line": 7,
+    "end_line": 11,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Environment Variables",
+    "comment": "Consider using a more secure method for storing sensitive information",
+    "confidence": "important",
+    "reason": "Storing sensitive information like API keys directly in environment variables can be a security risk",
+    "solution": "Use a secret management system or encrypt sensitive values",
+    "actual_code": "OPENAI_API_KEY=\nOPENAI_ORGANIZATION=",
+    "fixed_code": "# Use a secret management system to securely store and retrieve API keys\n# OPENAI_API_KEY=\n# OPENAI_ORGANIZATION=",
+    "file_name": ".env.example",
+    "start_line": 10,
+    "end_line": 11,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Database Configuration",
+    "comment": "Ensure proper indexing for performance optimization",
+    "confidence": "important",
+    "reason": "Proper indexing is crucial for database performance, especially for frequently queried columns",
+    "solution": "Review and optimize index creation based on query patterns",
+    "actual_code": "CREATE INDEX idx_file_path ON files(file_path);\n\nCREATE INDEX idx_function_name ON function_abstractions(function_name);\n\nCREATE INDEX idx_node_type ON syntax_nodes(node_type);",
+    "fixed_code": "CREATE INDEX idx_file_path ON files(file_path);\nCREATE INDEX idx_function_name ON function_abstractions(function_name);\nCREATE INDEX idx_node_type ON syntax_nodes(node_type);\n-- Consider adding composite indexes based on common query patterns\n-- CREATE INDEX idx_file_repo ON files(repo_id, file_path);\n-- CREATE INDEX idx_function_file ON function_abstractions(file_id, function_name);",
+    "file_name": "db_setup/init.sql",
+    "start_line": 72,
+    "end_line": 78,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 5
+  },
+  {
+    "topic": "Code Embedding",
+    "comment": "Hardcoded embedding dimensions may limit flexibility",
+    "confidence": "moderate",
+    "reason": "Using a fixed embedding size of 1536 may not be suitable for all models or future changes",
+    "solution": "Consider making the embedding dimensions configurable",
+    "actual_code": "response = self.provider.embedding(\n    model=\"embedding\", input=[text], dimensions=1536, encoding_format=\"float\"\n)",
+    "fixed_code": "embedding_dim = self.config.get('embedding_dimensions', 1536)\nresponse = self.provider.embedding(\n    model=\"embedding\", input=[text], dimensions=embedding_dim, encoding_format=\"float\"\n)",
+    "file_name": "kaizen/llms/provider.py",
+    "start_line": 242,
+    "end_line": 244,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Type Hinting",
+    "comment": "Consider using more specific type hints for better code clarity and maintainability.",
+    "confidence": "important",
+    "reason": "Using more specific type hints can improve code readability and catch potential type-related errors early.",
+    "solution": "Replace 'List[float]' with 'np.ndarray' for the query_embedding parameter, and use 'List[Dict[str, Any]]' for the return type.",
+    "actual_code": "def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:",
+    "fixed_code": "def custom_query(self, query_embedding: np.ndarray, repo_id: int, similarity_top_k: int) -> List[Dict[str, Any]]:",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 13,
+    "end_line": 13,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Add error handling for database operations to improve robustness.",
+    "confidence": "important",
+    "reason": "Database operations can fail due to various reasons, and proper error handling can prevent unexpected crashes and improve debugging.",
+    "solution": "Wrap the database operations in a try-except block and handle potential exceptions.",
+    "actual_code": "with self.get_client() as client:\n            with client.cursor() as cur:\n                cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))\n                results = cur.fetchall()",
+    "fixed_code": "try:\n    with self.get_client() as client:\n        with client.cursor() as cur:\n            cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))\n            results = cur.fetchall()\nexcept Exception as e:\n    # Log the error and handle it appropriately\n    print(f\"Database error:{e}\")\n    results =[]",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 39,
+    "end_line": 42,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Code Optimization",
+    "comment": "Consider using a list comprehension for creating the result list to improve performance and readability.",
+    "confidence": "moderate",
+    "reason": "List comprehensions are generally more efficient and concise than traditional for loops for creating lists.",
+    "solution": "Replace the for loop with a list comprehension.",
+    "actual_code": "return[\n{\n                \"id\": row[0],\n                \"text\": row[1],\n                \"metadata\": row[2] if isinstance(row[2], dict) else Json(row[2]),\n                \"similarity\": row[3]\n}\n            for row in results\n        ]",
+    "fixed_code": "return[{\n            \"id\": row[0],\n            \"text\": row[1],\n            \"metadata\": row[2] if isinstance(row[2], dict) else Json(row[2]),\n            \"similarity\": row[3]\n}for row in results]",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 44,
+    "end_line": 52,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Code Structure",
+    "comment": "Consider adding docstrings to methods for better documentation.",
+    "confidence": "moderate",
+    "reason": "Docstrings improve code readability and help other developers understand the purpose and usage of methods.",
+    "solution": "Add descriptive docstrings to the custom_query method and the AbstractionFeedback class methods.",
+    "actual_code": "def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:",
+    "fixed_code": "def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:\n    \"\"\"Perform a custom query on the vector store.\n\n    Args:\n        query_embedding (List[float]): The query embedding vector.\n        repo_id (int): The repository ID to filter results.\n        similarity_top_k (int): The number of top similar results to return.\n\n    Returns:\n        List[dict]: A list of dictionaries containing the query results.\n    \"\"\"",
+    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 13,
+    "end_line": 13,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Improve error handling in the parse_file method",
+    "confidence": "important",
+    "reason": "The current implementation catches all exceptions and logs them, but continues execution. This might lead to incomplete or inconsistent data.",
+    "solution": "Consider rethrowing specific exceptions or implementing a more granular error handling strategy.",
+    "actual_code": "except Exception as e:\n            logger.error(f\"Error processing file{file_path}:{str(e)}\")\n            logger.error(traceback.format_exc())",
+    "fixed_code": "except Exception as e:\n            logger.error(f\"Error processing file{file_path}:{str(e)}\")\n            logger.error(traceback.format_exc())\n            raise",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 108,
+    "end_line": 110,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Code Duplication",
+    "comment": "Repeated code for database connection string",
+    "confidence": "important",
+    "reason": "The database connection string is defined in multiple places, which violates the DRY principle and makes maintenance harder.",
+    "solution": "Extract the database connection string creation into a separate method or constant.",
+    "actual_code": "f\"postgresql://{os.environ['POSTGRES_USER']}:{os.environ['POSTGRES_PASSWORD']}@{os.environ['POSTGRES_HOST']}:{os.environ['POSTGRES_PORT']}/{os.environ['POSTGRES_DB']}\"",
+    "fixed_code": "self.db_connection_string = f\"postgresql://{os.environ['POSTGRES_USER']}:{os.environ['POSTGRES_PASSWORD']}@{os.environ['POSTGRES_HOST']}:{os.environ['POSTGRES_PORT']}/{os.environ['POSTGRES_DB']}\"\n        self.engine = create_engine(\n            self.db_connection_string,\n            pool_size=10,\n            max_overflow=20,\n        )",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 36,
+    "end_line": 37,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Code Optimization",
+    "comment": "Potential performance issue in store_function_relationships method",
+    "confidence": "moderate",
+    "reason": "The method executes a database query for each edge in the graph, which could be inefficient for large graphs.",
+    "solution": "Consider batching the inserts or using a more efficient bulk insert method if supported by the database.",
+    "actual_code": "for caller, callee in self.graph.edges():\n                query = text(\n                    \"\"\"\n                    INSERT INTO node_relationships (parent_node_id, child_node_id, relationship_type)\n                    VALUES (\n                        (SELECT node_id FROM syntax_nodes WHERE node_content LIKE :caller),\n                        (SELECT node_id FROM syntax_nodes WHERE node_content LIKE :callee),\n                        'calls'\n                    )\n                    ON CONFLICT DO NOTHING\n                \"\"\")\n                connection.execute(\n                    query,{\"caller\": f\"%{caller}%\", \"callee\": f\"%{callee}%\"}\n                )",
+    "fixed_code": "relationships =[(caller, callee) for caller, callee in self.graph.edges()]\n            query = text(\"\"\"\n                INSERT INTO node_relationships (parent_node_id, child_node_id, relationship_type)\n                VALUES (\n                    (SELECT node_id FROM syntax_nodes WHERE node_content LIKE :caller),\n                    (SELECT node_id FROM syntax_nodes WHERE node_content LIKE :callee),\n                    'calls'\n                )\n                ON CONFLICT DO NOTHING\n            \"\"\")\n            connection.execute(query,[{\"caller\": f\"%{caller}%\", \"callee\": f\"%{callee}%\"}for caller, callee in relationships])",
+    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 298,
+    "end_line": 312,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 5
+  },
+  {
+    "topic": "Python Version Upgrade",
+    "comment": "The minimum Python version has been increased from 3.8.1 to 3.9.0.",
+    "confidence": "important",
+    "reason": "This change may break compatibility with environments using Python 3.8.x.",
+    "solution": "Ensure all development and production environments are updated to Python 3.9.0 or higher. Update CI/CD pipelines and deployment scripts accordingly.",
+    "actual_code": "python = \"^3.9.0\"",
+    "fixed_code": "",
+    "file_name": "pyproject.toml",
+    "start_line": 13,
+    "end_line": 13,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 7
+  },
+  {
+    "topic": "New Dependencies",
+    "comment": "Several new dependencies have been added, including llama-index and tree-sitter related packages.",
+    "confidence": "important",
+    "reason": "New dependencies may introduce compatibility issues or increase the project's complexity.",
+    "solution": "Review each new dependency for necessity and potential impact on the project. Ensure they are compatible with existing dependencies and the project's requirements.",
+    "actual_code": "llama-index-core = \"^0.10.47\"\nllama-index-llms-openai = \"^0.1.22\"\nllama-index-readers-file = \"^0.1.25\"\nllama-index-vector-stores-postgres = \"^0.1.11\"\nsqlalchemy = \"^2.0.31\"\nesprima = \"^4.0.1\"\nescodegen = \"^1.0.11\"\ntree-sitter = \"^0.22.3\"\nllama-index = \"^0.10.65\"\ntree-sitter-python = \"^0.21.0\"\ntree-sitter-javascript = \"^0.21.4\"\ntree-sitter-typescript = \"^0.21.2\"\ntree-sitter-rust = \"^0.21.2\"\nllama-index-llms-litellm = \"^0.1.4\"\nllama-index-embeddings-litellm = \"^0.1.1\"",
+    "fixed_code": "",
+    "file_name": "pyproject.toml",
+    "start_line": 27,
+    "end_line": 43,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "Error Handling in LanguageLoader",
+    "comment": "The error handling in the LanguageLoader class could be improved for better debugging.",
+    "confidence": "moderate",
+    "reason": "The current error handling catches all exceptions and logs them, which might hide specific issues.",
+    "solution": "Consider catching specific exceptions (e.g., ImportError) separately and provide more detailed error messages.",
+    "actual_code": "except Exception as e:\n    logger.error(f\"Failed to load language{language}:{str(e)}\")\n    raise",
+    "fixed_code": "except ImportError as e:\n    logger.error(f\"Failed to import language module for{language}:{str(e)}\")\n    raise\nexcept Exception as e:\n    logger.error(f\"Unexpected error loading language{language}:{str(e)}\")\n    raise",
+    "file_name": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 28,
+    "end_line": 30,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 5
+  },
+  {
+    "topic": "Unused Import in Test File",
+    "comment": "The 'json' module is imported but not used in the test file.",
+    "confidence": "moderate",
+    "reason": "Unused imports can clutter the code and potentially confuse other developers.",
+    "solution": "Remove the unused import to improve code cleanliness.",
+    "actual_code": "import json",
+    "fixed_code": "",
+    "file_name": "tests/retriever/test_chunker.py",
+    "start_line": 2,
+    "end_line": 2,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 3
+  },
+  {
+    "topic": "Commented Out Code in Test File",
+    "comment": "There are several blocks of commented-out code in the test file.",
+    "confidence": "moderate",
+    "reason": "Commented-out code can make the file harder to read and maintain.",
+    "solution": "Remove commented-out code if it's no longer needed, or add a clear comment explaining why it's kept if it might be useful in the future.",
+    "actual_code": "# print(\"\\nFunctions:\")\n# for name, func in chunks[\"functions\"].items():\n#     print(f\"\\n{name}:\\n{func}\")\n\n# print(\"\\nClasses:\")\n# for name, class_info in chunks[\"classes\"].items():\n#     print(f\"\\n{name}:\")\n#     print(f\"Definition:\\n{class_info['definition']}\")\n#     print(\"Methods:\")\n#     for method_name, method in class_info[\"methods\"].items():\n#         print(f\"\\n{method_name}:\\n{method}\")\n\n# print(\"\\nOther Blocks:\")\n# for i, block in enumerate(chunks[\"other_blocks\"], 1):\n#     print(f\"\\nBlock{i}:\\n{block}\")",
+    "fixed_code": "",
+    "file_name": "tests/retriever/test_chunker.py",
+    "start_line": 81,
+    "end_line": 95,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 4
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Docker",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to Dockerfile, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_name": "Dockerfile",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Docker",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to docker-compose.yml, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "15",
+    "end_line": "15",
+    "side": "RIGHT",
+    "file_name": "docker-compose.yml",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Version Control",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to .gitignore, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "164",
+    "end_line": "164",
+    "side": "RIGHT",
+    "file_name": ".gitignore",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Database",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to db_setup/init.sql, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "db_setup/init.sql",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_222/review.md b/.experiments/code_review/sonnet-3.5/no_eval/pr_222/review.md
new file mode 100644
index 00000000..abca4169
--- /dev/null
+++ b/.experiments/code_review/sonnet-3.5/no_eval/pr_222/review.md
@@ -0,0 +1,481 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/222
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 21
+- Critical: 5
+- Important: 8
+- Minor: 8
+- Files Affected: 12
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Configuration (5 issues)</strong></summary>
+
+### 1. Changes made to sensitive file
+📁 **File:** `config.json:4`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to config.json, which needs review
+💡 **Solution:** NA
+
+### 2. Changes made to sensitive file
+📁 **File:** `Dockerfile:4`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to Dockerfile, which needs review
+💡 **Solution:** NA
+
+### 3. Changes made to sensitive file
+📁 **File:** `docker-compose.yml:15`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to docker-compose.yml, which needs review
+💡 **Solution:** NA
+
+### 4. Changes made to sensitive file
+📁 **File:** `.gitignore:164`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to .gitignore, which needs review
+💡 **Solution:** NA
+
+### 5. Changes made to sensitive file
+📁 **File:** `db_setup/init.sql:1`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to db_setup/init.sql, which needs review
+💡 **Solution:** NA
+
+</details>
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Environment Variables (8 issues)</strong></summary>
+
+### 1. Consider using a more secure method for storing sensitive information
+📁 **File:** `.env.example:10`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Storing sensitive information like API keys directly in environment variables can be a security risk
+💡 **Solution:** Use a secret management system or encrypt sensitive values
+
+**Current Code:**
+```python
+OPENAI_API_KEY=
+OPENAI_ORGANIZATION=
+```
+
+**Suggested Code:**
+```python
+# Use a secret management system to securely store and retrieve API keys
+# OPENAI_API_KEY=
+# OPENAI_ORGANIZATION=
+```
+
+### 2. Ensure proper indexing for performance optimization
+📁 **File:** `db_setup/init.sql:72`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Proper indexing is crucial for database performance, especially for frequently queried columns
+💡 **Solution:** Review and optimize index creation based on query patterns
+
+**Current Code:**
+```python
+CREATE INDEX idx_file_path ON files(file_path);
+
+CREATE INDEX idx_function_name ON function_abstractions(function_name);
+
+CREATE INDEX idx_node_type ON syntax_nodes(node_type);
+```
+
+**Suggested Code:**
+```python
+CREATE INDEX idx_file_path ON files(file_path);
+CREATE INDEX idx_function_name ON function_abstractions(function_name);
+CREATE INDEX idx_node_type ON syntax_nodes(node_type);
+-- Consider adding composite indexes based on common query patterns
+-- CREATE INDEX idx_file_repo ON files(repo_id, file_path);
+-- CREATE INDEX idx_function_file ON function_abstractions(file_id, function_name);
+```
+
+### 3. Consider using more specific type hints for better code clarity and maintainability.
+📁 **File:** `kaizen/retriever/custom_vector_store.py:13`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Using more specific type hints can improve code readability and catch potential type-related errors early.
+💡 **Solution:** Replace 'List[float]' with 'np.ndarray' for the query_embedding parameter, and use 'List[Dict[str, Any]]' for the return type.
+
+**Current Code:**
+```python
+def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:
+```
+
+**Suggested Code:**
+```python
+def custom_query(self, query_embedding: np.ndarray, repo_id: int, similarity_top_k: int) -> List[Dict[str, Any]]:
+```
+
+### 4. Add error handling for database operations to improve robustness.
+📁 **File:** `kaizen/retriever/custom_vector_store.py:39`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Database operations can fail due to various reasons, and proper error handling can prevent unexpected crashes and improve debugging.
+💡 **Solution:** Wrap the database operations in a try-except block and handle potential exceptions.
+
+**Current Code:**
+```python
+with self.get_client() as client:
+            with client.cursor() as cur:
+                cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))
+                results = cur.fetchall()
+```
+
+**Suggested Code:**
+```python
+try:
+    with self.get_client() as client:
+        with client.cursor() as cur:
+            cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))
+            results = cur.fetchall()
+except Exception as e:
+    # Log the error and handle it appropriately
+    print(f"Database error:{e}")
+    results =[]
+```
+
+### 5. Improve error handling in the parse_file method
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:108`
+⚖️ **Severity:** 7/10
+🔍 **Description:** The current implementation catches all exceptions and logs them, but continues execution. This might lead to incomplete or inconsistent data.
+💡 **Solution:** Consider rethrowing specific exceptions or implementing a more granular error handling strategy.
+
+**Current Code:**
+```python
+except Exception as e:
+            logger.error(f"Error processing file{file_path}:{str(e)}")
+            logger.error(traceback.format_exc())
+```
+
+**Suggested Code:**
+```python
+except Exception as e:
+            logger.error(f"Error processing file{file_path}:{str(e)}")
+            logger.error(traceback.format_exc())
+            raise
+```
+
+### 6. Repeated code for database connection string
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:36`
+⚖️ **Severity:** 6/10
+🔍 **Description:** The database connection string is defined in multiple places, which violates the DRY principle and makes maintenance harder.
+💡 **Solution:** Extract the database connection string creation into a separate method or constant.
+
+**Current Code:**
+```python
+f"postgresql://{os.environ['POSTGRES_USER']}:{os.environ['POSTGRES_PASSWORD']}@{os.environ['POSTGRES_HOST']}:{os.environ['POSTGRES_PORT']}/{os.environ['POSTGRES_DB']}"
+```
+
+**Suggested Code:**
+```python
+self.db_connection_string = f"postgresql://{os.environ['POSTGRES_USER']}:{os.environ['POSTGRES_PASSWORD']}@{os.environ['POSTGRES_HOST']}:{os.environ['POSTGRES_PORT']}/{os.environ['POSTGRES_DB']}"
+        self.engine = create_engine(
+            self.db_connection_string,
+            pool_size=10,
+            max_overflow=20,
+        )
+```
+
+### 7. The minimum Python version has been increased from 3.8.1 to 3.9.0.
+📁 **File:** `pyproject.toml:13`
+⚖️ **Severity:** 7/10
+🔍 **Description:** This change may break compatibility with environments using Python 3.8.x.
+💡 **Solution:** Ensure all development and production environments are updated to Python 3.9.0 or higher. Update CI/CD pipelines and deployment scripts accordingly.
+
+**Current Code:**
+```python
+python = "^3.9.0"
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 8. Several new dependencies have been added, including llama-index and tree-sitter related packages.
+📁 **File:** `pyproject.toml:27`
+⚖️ **Severity:** 6/10
+🔍 **Description:** New dependencies may introduce compatibility issues or increase the project's complexity.
+💡 **Solution:** Review each new dependency for necessity and potential impact on the project. Ensure they are compatible with existing dependencies and the project's requirements.
+
+**Current Code:**
+```python
+llama-index-core = "^0.10.47"
+llama-index-llms-openai = "^0.1.22"
+llama-index-readers-file = "^0.1.25"
+llama-index-vector-stores-postgres = "^0.1.11"
+sqlalchemy = "^2.0.31"
+esprima = "^4.0.1"
+escodegen = "^1.0.11"
+tree-sitter = "^0.22.3"
+llama-index = "^0.10.65"
+tree-sitter-python = "^0.21.0"
+tree-sitter-javascript = "^0.21.4"
+tree-sitter-typescript = "^0.21.2"
+tree-sitter-rust = "^0.21.2"
+llama-index-llms-litellm = "^0.1.4"
+llama-index-embeddings-litellm = "^0.1.1"
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (8 issues)</strong></summary>
+
+<details>
+<summary><strong>Docker Configuration (8 issues)</strong></summary>
+
+### 1. Consider using multi-stage builds to reduce the final image size
+📁 **File:** `Dockerfile:7`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Multi-stage builds can significantly reduce the size of the final Docker image by excluding build dependencies
+💡 **Solution:** Implement a multi-stage build in the Dockerfile
+
+**Current Code:**
+```python
+RUN apt-get update && apt-get install -y \
+    git \
+    build-essential \
+    && rm -rf /var/lib/apt/lists/*
+```
+
+**Suggested Code:**
+```python
+FROM python:3.9 AS builder
+
+RUN apt-get update && apt-get install -y \
+    git \
+    build-essential
+
+# ... (build steps)
+
+FROM python:3.9-slim
+
+COPY --from=builder /app /app
+
+# ... (runtime configuration)
+```
+
+### 2. Hardcoded embedding dimensions may limit flexibility
+📁 **File:** `kaizen/llms/provider.py:242`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Using a fixed embedding size of 1536 may not be suitable for all models or future changes
+💡 **Solution:** Consider making the embedding dimensions configurable
+
+**Current Code:**
+```python
+response = self.provider.embedding(
+    model="embedding", input=[text], dimensions=1536, encoding_format="float"
+)
+```
+
+**Suggested Code:**
+```python
+embedding_dim = self.config.get('embedding_dimensions', 1536)
+response = self.provider.embedding(
+    model="embedding", input=[text], dimensions=embedding_dim, encoding_format="float"
+)
+```
+
+### 3. Consider using a list comprehension for creating the result list to improve performance and readability.
+📁 **File:** `kaizen/retriever/custom_vector_store.py:44`
+⚖️ **Severity:** 3/10
+🔍 **Description:** List comprehensions are generally more efficient and concise than traditional for loops for creating lists.
+💡 **Solution:** Replace the for loop with a list comprehension.
+
+**Current Code:**
+```python
+return[
+{
+                "id": row[0],
+                "text": row[1],
+                "metadata": row[2] if isinstance(row[2], dict) else Json(row[2]),
+                "similarity": row[3]
+}
+            for row in results
+        ]
+```
+
+**Suggested Code:**
+```python
+return[{
+            "id": row[0],
+            "text": row[1],
+            "metadata": row[2] if isinstance(row[2], dict) else Json(row[2]),
+            "similarity": row[3]
+}for row in results]
+```
+
+### 4. Consider adding docstrings to methods for better documentation.
+📁 **File:** `kaizen/retriever/custom_vector_store.py:13`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Docstrings improve code readability and help other developers understand the purpose and usage of methods.
+💡 **Solution:** Add descriptive docstrings to the custom_query method and the AbstractionFeedback class methods.
+
+**Current Code:**
+```python
+def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:
+```
+
+**Suggested Code:**
+```python
+def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:
+    """Perform a custom query on the vector store.
+
+    Args:
+        query_embedding (List[float]): The query embedding vector.
+        repo_id (int): The repository ID to filter results.
+        similarity_top_k (int): The number of top similar results to return.
+
+    Returns:
+        List[dict]: A list of dictionaries containing the query results.
+    """
+```
+
+### 5. Potential performance issue in store_function_relationships method
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:298`
+⚖️ **Severity:** 5/10
+🔍 **Description:** The method executes a database query for each edge in the graph, which could be inefficient for large graphs.
+💡 **Solution:** Consider batching the inserts or using a more efficient bulk insert method if supported by the database.
+
+**Current Code:**
+```python
+for caller, callee in self.graph.edges():
+                query = text(
+                    """
+                    INSERT INTO node_relationships (parent_node_id, child_node_id, relationship_type)
+                    VALUES (
+                        (SELECT node_id FROM syntax_nodes WHERE node_content LIKE :caller),
+                        (SELECT node_id FROM syntax_nodes WHERE node_content LIKE :callee),
+                        'calls'
+                    )
+                    ON CONFLICT DO NOTHING
+                """)
+                connection.execute(
+                    query,{"caller": f"%{caller}%", "callee": f"%{callee}%"}
+                )
+```
+
+**Suggested Code:**
+```python
+relationships =[(caller, callee) for caller, callee in self.graph.edges()]
+            query = text("""
+                INSERT INTO node_relationships (parent_node_id, child_node_id, relationship_type)
+                VALUES (
+                    (SELECT node_id FROM syntax_nodes WHERE node_content LIKE :caller),
+                    (SELECT node_id FROM syntax_nodes WHERE node_content LIKE :callee),
+                    'calls'
+                )
+                ON CONFLICT DO NOTHING
+            """)
+            connection.execute(query,[{"caller": f"%{caller}%", "callee": f"%{callee}%"}for caller, callee in relationships])
+```
+
+### 6. The error handling in the LanguageLoader class could be improved for better debugging.
+📁 **File:** `kaizen/retriever/tree_sitter_utils.py:28`
+⚖️ **Severity:** 5/10
+🔍 **Description:** The current error handling catches all exceptions and logs them, which might hide specific issues.
+💡 **Solution:** Consider catching specific exceptions (e.g., ImportError) separately and provide more detailed error messages.
+
+**Current Code:**
+```python
+except Exception as e:
+    logger.error(f"Failed to load language{language}:{str(e)}")
+    raise
+```
+
+**Suggested Code:**
+```python
+except ImportError as e:
+    logger.error(f"Failed to import language module for{language}:{str(e)}")
+    raise
+except Exception as e:
+    logger.error(f"Unexpected error loading language{language}:{str(e)}")
+    raise
+```
+
+### 7. The 'json' module is imported but not used in the test file.
+📁 **File:** `tests/retriever/test_chunker.py:2`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Unused imports can clutter the code and potentially confuse other developers.
+💡 **Solution:** Remove the unused import to improve code cleanliness.
+
+**Current Code:**
+```python
+import json
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 8. There are several blocks of commented-out code in the test file.
+📁 **File:** `tests/retriever/test_chunker.py:81`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Commented-out code can make the file harder to read and maintain.
+💡 **Solution:** Remove commented-out code if it's no longer needed, or add a clear comment explaining why it's kept if it might be useful in the future.
+
+**Current Code:**
+```python
+# print("\nFunctions:")
+# for name, func in chunks["functions"].items():
+#     print(f"\n{name}:\n{func}")
+
+# print("\nClasses:")
+# for name, class_info in chunks["classes"].items():
+#     print(f"\n{name}:")
+#     print(f"Definition:\n{class_info['definition']}")
+#     print("Methods:")
+#     for method_name, method in class_info["methods"].items():
+#         print(f"\n{method_name}:\n{method}")
+
+# print("\nOther Blocks:")
+# for i, block in enumerate(chunks["other_blocks"], 1):
+#     print(f"\nBlock{i}:\n{block}")
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (anthropic.claude-3-5-sonnet-20240620-v1:0)
+{"prompt_tokens": 24844, "completion_tokens": 5105, "total_tokens": 29949}
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_232/comments.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_232/comments.json
new file mode 100644
index 00000000..0637a088
--- /dev/null
+++ b/.experiments/code_review/sonnet-3.5/no_eval/pr_232/comments.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_232/issues.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_232/issues.json
new file mode 100644
index 00000000..5a4abbcf
--- /dev/null
+++ b/.experiments/code_review/sonnet-3.5/no_eval/pr_232/issues.json
@@ -0,0 +1,227 @@
+[
+  {
+    "topic": "Type Definition",
+    "comment": "Improved type definition for MemoriesPage props",
+    "confidence": "important",
+    "reason": "Using a separate type definition improves code readability and maintainability",
+    "solution": "The change is already implemented correctly",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 40,
+    "end_line": 45,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Error Message",
+    "comment": "Improved error message for space deletion",
+    "confidence": "moderate",
+    "reason": "More specific error message provides better user feedback",
+    "solution": "The change is already implemented correctly",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 73,
+    "end_line": 73,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 2
+  },
+  {
+    "topic": "Component Naming",
+    "comment": "Renamed components for better clarity",
+    "confidence": "important",
+    "reason": "More descriptive component names improve code readability",
+    "solution": "The changes are already implemented correctly",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 231,
+    "end_line": 231,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "UI Improvements",
+    "comment": "Enhanced UI elements with better styling and layout",
+    "confidence": "moderate",
+    "reason": "Improved visual consistency and user experience",
+    "solution": "The changes are already implemented correctly",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 140,
+    "end_line": 152,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Import Optimization",
+    "comment": "Removed unused import",
+    "confidence": "important",
+    "reason": "Removing unused imports improves code cleanliness and potentially reduces bundle size",
+    "solution": "The change is already implemented correctly",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "apps/web/app/(dash)/home/page.tsx",
+    "start_line": 6,
+    "end_line": 6,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Import Cleanup",
+    "comment": "Removed unused imports",
+    "confidence": "important",
+    "reason": "Removing unused imports improves code cleanliness and potentially reduces bundle size",
+    "solution": "The change is already implemented correctly",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "apps/web/app/(dash)/home/queryinput.tsx",
+    "start_line": 3,
+    "end_line": 3,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Code Structure",
+    "comment": "Improved code organization by extracting dialog content into a separate component",
+    "confidence": "important",
+    "reason": "Separating concerns improves readability and maintainability",
+    "solution": "The change is already implemented correctly",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 163,
+    "end_line": 346,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 7
+  },
+  {
+    "topic": "State Management",
+    "comment": "Moved state management for spaces and selectedSpaces into the DialogContentContainer component",
+    "confidence": "important",
+    "reason": "Localizing state management to the component that uses it improves encapsulation",
+    "solution": "The change is already implemented correctly",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 168,
+    "end_line": 170,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 6
+  },
+  {
+    "topic": "Prop Drilling",
+    "comment": "Consider using context or state management library to avoid prop drilling",
+    "confidence": "moderate",
+    "reason": "The setDialogClose function is passed down as a prop, which could lead to prop drilling in larger components",
+    "solution": "Implement React Context or use a state management library like Redux for managing global state",
+    "actual_code": "function DialogContentContainer({\n\tsetDialogClose,\n}:{\n\tsetDialogClose: () => void;\n}){",
+    "fixed_code": "const DialogContext = React.createContext();\n\nfunction DialogContentContainer(){\n  const{setDialogClose}= useContext(DialogContext);",
+    "file_name": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 163,
+    "end_line": 167,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Improve error handling in the handleSubmit function",
+    "confidence": "important",
+    "reason": "The current implementation throws an error but then continues execution",
+    "solution": "Remove the return statement after throwing the error",
+    "actual_code": "throw new Error(`Memory creation failed: ${cont.error}`);\nreturn cont;",
+    "fixed_code": "throw new Error(`Memory creation failed: ${cont.error}`);",
+    "file_name": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 230,
+    "end_line": 231,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Unused Import",
+    "comment": "The useEffect import is removed but not replaced with any other import.",
+    "confidence": "important",
+    "reason": "Removing unused imports improves code cleanliness and potentially reduces bundle size.",
+    "solution": "Ensure all necessary hooks are imported and remove any unused imports.",
+    "actual_code": "import{useState}from \"react\";",
+    "fixed_code": "import{useState, useEffect}from \"react\";",
+    "file_name": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 3,
+    "end_line": 3,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 3
+  },
+  {
+    "topic": "Type Safety",
+    "comment": "The component definition has been changed from a typed functional component to a regular function without explicit typing.",
+    "confidence": "important",
+    "reason": "Removing explicit typing can lead to potential type-related bugs and reduces code readability.",
+    "solution": "Maintain explicit typing for the component to ensure type safety and improve code clarity.",
+    "actual_code": "const ComboboxWithCreate = ({",
+    "fixed_code": "const ComboboxWithCreate: React.FC<ComboboxWithCreateProps> = ({",
+    "file_name": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 32,
+    "end_line": 32,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "The new handleKeyDown function doesn't handle potential undefined values when accessing selectedSpaces.",
+    "confidence": "moderate",
+    "reason": "Not checking for undefined values can lead to runtime errors if selectedSpaces is not properly initialized.",
+    "solution": "Add a null check before accessing selectedSpaces.length.",
+    "actual_code": "if (\n\t\t\te.key === \"Backspace\" &&\n\t\t\tinputValue === \"\" &&\n\t\t\tselectedSpaces.length > 0\n\t\t){\n\t\t\tsetSelectedSpaces((prev) => prev.slice(0, -1));\n\t\t}",
+    "fixed_code": "if (\n\t\t\te.key === \"Backspace\" &&\n\t\t\tinputValue === \"\" &&\n\t\t\tselectedSpaces?.length > 0\n\t\t){\n\t\t\tsetSelectedSpaces((prev) => prev.slice(0, -1));\n\t\t}",
+    "file_name": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 46,
+    "end_line": 52,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 5
+  },
+  {
+    "topic": "Performance Optimization",
+    "comment": "The filteredOptions array is being recalculated on every render, which could be inefficient for large arrays.",
+    "confidence": "moderate",
+    "reason": "Recalculating filtered options on every render can lead to unnecessary computations and potential performance issues.",
+    "solution": "Consider using useMemo to memoize the filteredOptions calculation.",
+    "actual_code": "const filteredOptions = options.filter(\n\t\t(option) => !selectedSpaces.includes(parseInt(option.value)),\n\t);",
+    "fixed_code": "const filteredOptions = useMemo(() => options.filter(\n\t\t(option) => !selectedSpaces.includes(parseInt(option.value)),\n\t),[options, selectedSpaces]);",
+    "file_name": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 55,
+    "end_line": 57,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Accessibility",
+    "comment": "The button for removing selected spaces lacks an aria-label for better accessibility.",
+    "confidence": "moderate",
+    "reason": "Missing aria-labels can make it difficult for screen reader users to understand the purpose of interactive elements.",
+    "solution": "Add an appropriate aria-label to the button for removing selected spaces.",
+    "actual_code": "<button\n\t\t\t\t\t\t\tkey={spaceId}\n\t\t\t\t\t\t\ttype=\"button\"\n\t\t\t\t\t\t\tonClick={() =>\n\t\t\t\t\t\t\t\tsetSelectedSpaces((prev) => prev.filter((id) => id !== spaceId))\n\t\t\t\t\t\t\t}\n\t\t\t\t\t\t\tclassName=\"relative group rounded-md py-1 px-2 bg-[#3C464D] max-w-32\"\n\t\t\t\t\t\t>",
+    "fixed_code": "<button\n\t\t\t\t\t\t\tkey={spaceId}\n\t\t\t\t\t\t\ttype=\"button\"\n\t\t\t\t\t\t\tonClick={() =>\n\t\t\t\t\t\t\t\tsetSelectedSpaces((prev) => prev.filter((id) => id !== spaceId))\n\t\t\t\t\t\t\t}\n\t\t\t\t\t\t\tclassName=\"relative group rounded-md py-1 px-2 bg-[#3C464D] max-w-32\"\n\t\t\t\t\t\t\taria-label={`Remove ${options.find((opt) => opt.value === spaceId.toString())?.label}`}\n\t\t\t\t\t\t>",
+    "file_name": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 65,
+    "end_line": 72,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_232/review.md b/.experiments/code_review/sonnet-3.5/no_eval/pr_232/review.md
new file mode 100644
index 00000000..0c5cbfbc
--- /dev/null
+++ b/.experiments/code_review/sonnet-3.5/no_eval/pr_232/review.md
@@ -0,0 +1,253 @@
+PR URL: https://github.com/supermemoryai/supermemory/pull/232
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 15
+- Critical: 0
+- Important: 9
+- Minor: 6
+- Files Affected: 5
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Type Definition (9 issues)</strong></summary>
+
+### 1. Improved type definition for MemoriesPage props
+📁 **File:** `apps/web/app/(dash)/(memories)/content.tsx:40`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Using a separate type definition improves code readability and maintainability
+💡 **Solution:** The change is already implemented correctly
+
+### 2. Renamed components for better clarity
+📁 **File:** `apps/web/app/(dash)/(memories)/content.tsx:231`
+⚖️ **Severity:** 3/10
+🔍 **Description:** More descriptive component names improve code readability
+💡 **Solution:** The changes are already implemented correctly
+
+### 3. Removed unused import
+📁 **File:** `apps/web/app/(dash)/home/page.tsx:6`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Removing unused imports improves code cleanliness and potentially reduces bundle size
+💡 **Solution:** The change is already implemented correctly
+
+### 4. Removed unused imports
+📁 **File:** `apps/web/app/(dash)/home/queryinput.tsx:3`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Removing unused imports improves code cleanliness and potentially reduces bundle size
+💡 **Solution:** The change is already implemented correctly
+
+### 5. Improved code organization by extracting dialog content into a separate component
+📁 **File:** `apps/web/app/(dash)/menu.tsx:163`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Separating concerns improves readability and maintainability
+💡 **Solution:** The change is already implemented correctly
+
+### 6. Moved state management for spaces and selectedSpaces into the DialogContentContainer component
+📁 **File:** `apps/web/app/(dash)/menu.tsx:168`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Localizing state management to the component that uses it improves encapsulation
+💡 **Solution:** The change is already implemented correctly
+
+### 7. Improve error handling in the handleSubmit function
+📁 **File:** `apps/web/app/(dash)/menu.tsx:230`
+⚖️ **Severity:** 7/10
+🔍 **Description:** The current implementation throws an error but then continues execution
+💡 **Solution:** Remove the return statement after throwing the error
+
+**Current Code:**
+```python
+throw new Error(`Memory creation failed: ${cont.error}`);
+return cont;
+```
+
+**Suggested Code:**
+```python
+throw new Error(`Memory creation failed: ${cont.error}`);
+```
+
+### 8. The useEffect import is removed but not replaced with any other import.
+📁 **File:** `packages/ui/shadcn/combobox.tsx:3`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Removing unused imports improves code cleanliness and potentially reduces bundle size.
+💡 **Solution:** Ensure all necessary hooks are imported and remove any unused imports.
+
+**Current Code:**
+```python
+import{useState}from "react";
+```
+
+**Suggested Code:**
+```python
+import{useState, useEffect}from "react";
+```
+
+### 9. The component definition has been changed from a typed functional component to a regular function without explicit typing.
+📁 **File:** `packages/ui/shadcn/combobox.tsx:32`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Removing explicit typing can lead to potential type-related bugs and reduces code readability.
+💡 **Solution:** Maintain explicit typing for the component to ensure type safety and improve code clarity.
+
+**Current Code:**
+```python
+const ComboboxWithCreate = ({
+```
+
+**Suggested Code:**
+```python
+const ComboboxWithCreate: React.FC<ComboboxWithCreateProps> = ({
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (6 issues)</strong></summary>
+
+<details>
+<summary><strong>Error Message (6 issues)</strong></summary>
+
+### 1. Improved error message for space deletion
+📁 **File:** `apps/web/app/(dash)/(memories)/content.tsx:73`
+⚖️ **Severity:** 2/10
+🔍 **Description:** More specific error message provides better user feedback
+💡 **Solution:** The change is already implemented correctly
+
+### 2. Enhanced UI elements with better styling and layout
+📁 **File:** `apps/web/app/(dash)/(memories)/content.tsx:140`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Improved visual consistency and user experience
+💡 **Solution:** The changes are already implemented correctly
+
+### 3. Consider using context or state management library to avoid prop drilling
+📁 **File:** `apps/web/app/(dash)/menu.tsx:163`
+⚖️ **Severity:** 4/10
+🔍 **Description:** The setDialogClose function is passed down as a prop, which could lead to prop drilling in larger components
+💡 **Solution:** Implement React Context or use a state management library like Redux for managing global state
+
+**Current Code:**
+```python
+function DialogContentContainer({
+	setDialogClose,
+}:{
+	setDialogClose: () => void;
+}){
+```
+
+**Suggested Code:**
+```python
+const DialogContext = React.createContext();
+
+function DialogContentContainer(){
+  const{setDialogClose}= useContext(DialogContext);
+```
+
+### 4. The new handleKeyDown function doesn't handle potential undefined values when accessing selectedSpaces.
+📁 **File:** `packages/ui/shadcn/combobox.tsx:46`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Not checking for undefined values can lead to runtime errors if selectedSpaces is not properly initialized.
+💡 **Solution:** Add a null check before accessing selectedSpaces.length.
+
+**Current Code:**
+```python
+if (
+			e.key === "Backspace" &&
+			inputValue === "" &&
+			selectedSpaces.length > 0
+		){
+			setSelectedSpaces((prev) => prev.slice(0, -1));
+		}
+```
+
+**Suggested Code:**
+```python
+if (
+			e.key === "Backspace" &&
+			inputValue === "" &&
+			selectedSpaces?.length > 0
+		){
+			setSelectedSpaces((prev) => prev.slice(0, -1));
+		}
+```
+
+### 5. The filteredOptions array is being recalculated on every render, which could be inefficient for large arrays.
+📁 **File:** `packages/ui/shadcn/combobox.tsx:55`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Recalculating filtered options on every render can lead to unnecessary computations and potential performance issues.
+💡 **Solution:** Consider using useMemo to memoize the filteredOptions calculation.
+
+**Current Code:**
+```python
+const filteredOptions = options.filter(
+		(option) => !selectedSpaces.includes(parseInt(option.value)),
+	);
+```
+
+**Suggested Code:**
+```python
+const filteredOptions = useMemo(() => options.filter(
+		(option) => !selectedSpaces.includes(parseInt(option.value)),
+	),[options, selectedSpaces]);
+```
+
+### 6. The button for removing selected spaces lacks an aria-label for better accessibility.
+📁 **File:** `packages/ui/shadcn/combobox.tsx:65`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Missing aria-labels can make it difficult for screen reader users to understand the purpose of interactive elements.
+💡 **Solution:** Add an appropriate aria-label to the button for removing selected spaces.
+
+**Current Code:**
+```python
+<button
+							key={spaceId}
+							type="button"
+							onClick={() =>
+								setSelectedSpaces((prev) => prev.filter((id) => id !== spaceId))
+							}
+							className="relative group rounded-md py-1 px-2 bg-[#3C464D] max-w-32"
+						>
+```
+
+**Suggested Code:**
+```python
+<button
+							key={spaceId}
+							type="button"
+							onClick={() =>
+								setSelectedSpaces((prev) => prev.filter((id) => id !== spaceId))
+							}
+							className="relative group rounded-md py-1 px-2 bg-[#3C464D] max-w-32"
+							aria-label={`Remove ${options.find((opt) => opt.value === spaceId.toString())?.label}`}
+						>
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (anthropic.claude-3-5-sonnet-20240620-v1:0)
+{"prompt_tokens": 17838, "completion_tokens": 3363, "total_tokens": 21201}
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_252/comments.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_252/comments.json
new file mode 100644
index 00000000..0637a088
--- /dev/null
+++ b/.experiments/code_review/sonnet-3.5/no_eval/pr_252/comments.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_252/issues.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_252/issues.json
new file mode 100644
index 00000000..0b1ff3f5
--- /dev/null
+++ b/.experiments/code_review/sonnet-3.5/no_eval/pr_252/issues.json
@@ -0,0 +1,62 @@
+[
+  {
+    "topic": "Code Structure",
+    "comment": "The `generate_linkedin_post` method call has been split across multiple lines, which improves readability for long function calls.",
+    "confidence": "moderate",
+    "reason": "Multi-line function calls can improve code readability, especially for functions with long parameter lists.",
+    "solution": "The current implementation is good. No changes needed.",
+    "actual_code": "linkedin_post = work_summary_generator.generate_linkedin_post(\n    summary, user=\"oss_example\"\n)",
+    "fixed_code": "",
+    "file_name": "examples/work_summarizer/main.py",
+    "start_line": 60,
+    "end_line": 62,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 2
+  },
+  {
+    "topic": "Code Consistency",
+    "comment": "The `generate_twitter_post` method call is not formatted consistently with the `generate_linkedin_post` call.",
+    "confidence": "moderate",
+    "reason": "Consistent formatting improves code readability and maintainability.",
+    "solution": "Consider formatting the `generate_twitter_post` call similarly to the `generate_linkedin_post` call for consistency.",
+    "actual_code": "twitter_post = work_summary_generator.generate_twitter_post(summary, user=\"oss_example\")",
+    "fixed_code": "twitter_post = work_summary_generator.generate_twitter_post(\n    summary, user=\"oss_example\"\n)",
+    "file_name": "examples/work_summarizer/main.py",
+    "start_line": 59,
+    "end_line": 59,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 3
+  },
+  {
+    "topic": "Code Duplication",
+    "comment": "The print statement for LinkedIn post is duplicated.",
+    "confidence": "important",
+    "reason": "Code duplication can lead to maintenance issues and inconsistencies.",
+    "solution": "Remove the duplicated print statement for the LinkedIn post.",
+    "actual_code": "print(f\" LinkedIn Post: \\n{linkedin_post}\\n\")",
+    "fixed_code": "",
+    "file_name": "examples/work_summarizer/main.py",
+    "start_line": 68,
+    "end_line": 68,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Code Documentation",
+    "comment": "The new `severity_level` field in the code review prompt is not explained in detail.",
+    "confidence": "moderate",
+    "reason": "Clear documentation helps users understand how to use the severity level correctly.",
+    "solution": "Add a brief explanation of what each severity level represents (e.g., what constitutes a level 1 vs. level 10 issue).",
+    "actual_code": "For \"severity_level\" score in range of 1 to 10, 1 being not severe and 10 being critical.",
+    "fixed_code": "For \"severity_level\" score in range of 1 to 10:\n1-3: Minor issues (style, small optimizations)\n4-6: Moderate issues (potential bugs, performance concerns)\n7-8: Major issues (definite bugs, security vulnerabilities)\n9-10: Critical issues (severe security risks, system-breaking bugs)",
+    "file_name": "kaizen/llms/prompts/code_review_prompts.py",
+    "start_line": 100,
+    "end_line": 100,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_252/review.md b/.experiments/code_review/sonnet-3.5/no_eval/pr_252/review.md
new file mode 100644
index 00000000..d9e48407
--- /dev/null
+++ b/.experiments/code_review/sonnet-3.5/no_eval/pr_252/review.md
@@ -0,0 +1,126 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/252
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 4
+- Critical: 0
+- Important: 1
+- Minor: 3
+- Files Affected: 2
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Code Duplication (1 issues)</strong></summary>
+
+### 1. The print statement for LinkedIn post is duplicated.
+📁 **File:** `examples/work_summarizer/main.py:68`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Code duplication can lead to maintenance issues and inconsistencies.
+💡 **Solution:** Remove the duplicated print statement for the LinkedIn post.
+
+**Current Code:**
+```python
+print(f" LinkedIn Post: \n{linkedin_post}\n")
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (3 issues)</strong></summary>
+
+<details>
+<summary><strong>Code Structure (3 issues)</strong></summary>
+
+### 1. The `generate_linkedin_post` method call has been split across multiple lines, which improves readability for long function calls.
+📁 **File:** `examples/work_summarizer/main.py:60`
+⚖️ **Severity:** 2/10
+🔍 **Description:** Multi-line function calls can improve code readability, especially for functions with long parameter lists.
+💡 **Solution:** The current implementation is good. No changes needed.
+
+**Current Code:**
+```python
+linkedin_post = work_summary_generator.generate_linkedin_post(
+    summary, user="oss_example"
+)
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 2. The `generate_twitter_post` method call is not formatted consistently with the `generate_linkedin_post` call.
+📁 **File:** `examples/work_summarizer/main.py:59`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Consistent formatting improves code readability and maintainability.
+💡 **Solution:** Consider formatting the `generate_twitter_post` call similarly to the `generate_linkedin_post` call for consistency.
+
+**Current Code:**
+```python
+twitter_post = work_summary_generator.generate_twitter_post(summary, user="oss_example")
+```
+
+**Suggested Code:**
+```python
+twitter_post = work_summary_generator.generate_twitter_post(
+    summary, user="oss_example"
+)
+```
+
+### 3. The new `severity_level` field in the code review prompt is not explained in detail.
+📁 **File:** `kaizen/llms/prompts/code_review_prompts.py:100`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Clear documentation helps users understand how to use the severity level correctly.
+💡 **Solution:** Add a brief explanation of what each severity level represents (e.g., what constitutes a level 1 vs. level 10 issue).
+
+**Current Code:**
+```python
+For "severity_level" score in range of 1 to 10, 1 being not severe and 10 being critical.
+```
+
+**Suggested Code:**
+```python
+For "severity_level" score in range of 1 to 10:
+1-3: Minor issues (style, small optimizations)
+4-6: Moderate issues (potential bugs, performance concerns)
+7-8: Major issues (definite bugs, security vulnerabilities)
+9-10: Critical issues (severe security risks, system-breaking bugs)
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (anthropic.claude-3-5-sonnet-20240620-v1:0)
+{"prompt_tokens": 4436, "completion_tokens": 952, "total_tokens": 5388}
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_335/comments.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_335/comments.json
new file mode 100644
index 00000000..0637a088
--- /dev/null
+++ b/.experiments/code_review/sonnet-3.5/no_eval/pr_335/comments.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_335/issues.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_335/issues.json
new file mode 100644
index 00000000..5b6fae70
--- /dev/null
+++ b/.experiments/code_review/sonnet-3.5/no_eval/pr_335/issues.json
@@ -0,0 +1,62 @@
+[
+  {
+    "topic": "Code Refactoring",
+    "comment": "Removal of reevaluation logic and associated prompts",
+    "confidence": "important",
+    "reason": "The code has been simplified by removing the reevaluation logic, which could potentially improve performance and reduce complexity.",
+    "solution": "Ensure that the removal of reevaluation doesn't impact the quality of PR descriptions. Consider adding unit tests to verify the new behavior.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 86,
+    "end_line": 88,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "API Change",
+    "comment": "Change from chat_completion_with_json to chat_completion",
+    "confidence": "important",
+    "reason": "The API call has been changed, which might affect the format of the response and how it's processed.",
+    "solution": "Ensure that the new chat_completion method returns the expected format. Update any dependent code that might be affected by this change.",
+    "actual_code": "resp, usage = self.provider.chat_completion(prompt, user=user)",
+    "fixed_code": "resp, usage = self.provider.chat_completion(prompt, user=user)\ndesc = parser.extract_code_from_markdown(resp)",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 79,
+    "end_line": 80,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "Code Simplification",
+    "comment": "Removal of reeval_response parameter from multiple methods",
+    "confidence": "moderate",
+    "reason": "The reeval_response parameter has been removed from several method signatures, simplifying the code.",
+    "solution": "Verify that the removal of this parameter doesn't affect any calling code. Update any documentation or comments related to these methods.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/pr_description.py",
+    "start_line": 52,
+    "end_line": 52,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "File Reorganization",
+    "comment": "Move of PR description prompts to a new file",
+    "confidence": "important",
+    "reason": "PR description prompts have been moved from code_review_prompts.py to pr_desc_prompts.py, improving code organization.",
+    "solution": "Update any import statements in other files that might be affected by this change. Ensure that all necessary prompts have been moved correctly.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/llms/prompts/pr_desc_prompts.py",
+    "start_line": 1,
+    "end_line": 92,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_335/review.md b/.experiments/code_review/sonnet-3.5/no_eval/pr_335/review.md
new file mode 100644
index 00000000..8c444640
--- /dev/null
+++ b/.experiments/code_review/sonnet-3.5/no_eval/pr_335/review.md
@@ -0,0 +1,89 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/335
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 4
+- Critical: 0
+- Important: 3
+- Minor: 1
+- Files Affected: 2
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Code Refactoring (3 issues)</strong></summary>
+
+### 1. Removal of reevaluation logic and associated prompts
+📁 **File:** `kaizen/generator/pr_description.py:86`
+⚖️ **Severity:** 5/10
+🔍 **Description:** The code has been simplified by removing the reevaluation logic, which could potentially improve performance and reduce complexity.
+💡 **Solution:** Ensure that the removal of reevaluation doesn't impact the quality of PR descriptions. Consider adding unit tests to verify the new behavior.
+
+### 2. Change from chat_completion_with_json to chat_completion
+📁 **File:** `kaizen/generator/pr_description.py:79`
+⚖️ **Severity:** 6/10
+🔍 **Description:** The API call has been changed, which might affect the format of the response and how it's processed.
+💡 **Solution:** Ensure that the new chat_completion method returns the expected format. Update any dependent code that might be affected by this change.
+
+**Current Code:**
+```python
+resp, usage = self.provider.chat_completion(prompt, user=user)
+```
+
+**Suggested Code:**
+```python
+resp, usage = self.provider.chat_completion(prompt, user=user)
+desc = parser.extract_code_from_markdown(resp)
+```
+
+### 3. Move of PR description prompts to a new file
+📁 **File:** `kaizen/llms/prompts/pr_desc_prompts.py:1`
+⚖️ **Severity:** 4/10
+🔍 **Description:** PR description prompts have been moved from code_review_prompts.py to pr_desc_prompts.py, improving code organization.
+💡 **Solution:** Update any import statements in other files that might be affected by this change. Ensure that all necessary prompts have been moved correctly.
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (1 issues)</strong></summary>
+
+<details>
+<summary><strong>Code Simplification (1 issues)</strong></summary>
+
+### 1. Removal of reeval_response parameter from multiple methods
+📁 **File:** `kaizen/generator/pr_description.py:52`
+⚖️ **Severity:** 3/10
+🔍 **Description:** The reeval_response parameter has been removed from several method signatures, simplifying the code.
+💡 **Solution:** Verify that the removal of this parameter doesn't affect any calling code. Update any documentation or comments related to these methods.
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (anthropic.claude-3-5-sonnet-20240620-v1:0)
+{"prompt_tokens": 7614, "completion_tokens": 826, "total_tokens": 8440}
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_400/comments.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_400/comments.json
new file mode 100644
index 00000000..f470a30b
--- /dev/null
+++ b/.experiments/code_review/sonnet-3.5/no_eval/pr_400/comments.json
@@ -0,0 +1,31 @@
+[
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "11",
+    "end_line": "11",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  },
+  {
+    "topic": "Asynchronous Testing",
+    "comment": "The tests have been updated to use async/await syntax, which is correct for testing asynchronous functions.",
+    "confidence": "critical",
+    "reason": "Proper async testing is crucial for accurate results when testing asynchronous code.",
+    "solution": "The implementation is correct. Ensure all test functions are defined with 'async def'.",
+    "actual_code": "async def test_get_web_html_normal_cases(mock_get_html, mock_nest_asyncio, html_content, expected_output):",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 76,
+    "end_line": 76,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 8
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_400/issues.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_400/issues.json
new file mode 100644
index 00000000..ccc761ea
--- /dev/null
+++ b/.experiments/code_review/sonnet-3.5/no_eval/pr_400/issues.json
@@ -0,0 +1,406 @@
+[
+  {
+    "topic": "Test Coverage",
+    "comment": "Improved test coverage with more comprehensive test cases",
+    "confidence": "important",
+    "reason": "The new test cases cover a wider range of scenarios, including edge cases and error handling",
+    "solution": "No changes needed. The improvements are good.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
+    "start_line": 7,
+    "end_line": 64,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 2
+  },
+  {
+    "topic": "Performance Testing",
+    "comment": "Added performance testing for large inputs, but removed arbitrary time limit",
+    "confidence": "moderate",
+    "reason": "Performance testing is important, but the removal of the 1-second boundary condition is a good change as it avoids potential flaky tests",
+    "solution": "Consider adding a more flexible performance assertion based on input size",
+    "actual_code": "print(f\"Execution time:{execution_time}seconds\")",
+    "fixed_code": "assert execution_time < len(desc + original_desc) * 0.0001, f\"Execution time ({execution_time}seconds) exceeded expected limit\"",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
+    "start_line": 61,
+    "end_line": 61,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Added error handling tests for invalid input types",
+    "confidence": "important",
+    "reason": "Proper error handling is crucial for robust code",
+    "solution": "No changes needed. The added error handling tests are beneficial.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
+    "start_line": 31,
+    "end_line": 43,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Test Structure",
+    "comment": "Improved test structure using pytest.mark.parametrize",
+    "confidence": "important",
+    "reason": "Parameterized tests reduce code duplication and make it easier to add new test cases",
+    "solution": "No changes needed. The use of pytest.mark.parametrize is a good practice.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
+    "start_line": 7,
+    "end_line": 28,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 2
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Updated .flake8 configuration to exclude unit test directory",
+    "confidence": "moderate",
+    "reason": "Excluding unit tests from flake8 checks can be beneficial, but it's important to maintain code quality in tests as well",
+    "solution": "Consider running flake8 on test files with a separate, less strict configuration",
+    "actual_code": "exclude = docs/*, venv/*, .kaizen/unit_test/*",
+    "fixed_code": "exclude = docs/*, venv/*",
+    "file_name": ".flake8",
+    "start_line": 2,
+    "end_line": 2,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 3
+  },
+  {
+    "topic": "Environment Configuration",
+    "comment": "Added LITELLM_LOG environment variable",
+    "confidence": "moderate",
+    "reason": "Adding logging configuration can improve debugging and monitoring",
+    "solution": "Ensure this change is documented in the project's README or documentation",
+    "actual_code": "LITELLM_LOG=\"ERROR\"",
+    "fixed_code": "",
+    "file_name": ".env.example",
+    "start_line": 6,
+    "end_line": 6,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 2
+  },
+  {
+    "topic": "Test Coverage",
+    "comment": "New test cases have been added to improve test coverage, which is a positive change.",
+    "confidence": "important",
+    "reason": "Comprehensive test coverage is crucial for maintaining code quality and catching potential bugs.",
+    "solution": "Continue to add test cases for edge cases and ensure all new functionality is covered by tests.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 18,
+    "end_line": 33,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 7
+  },
+  {
+    "topic": "Code Structure",
+    "comment": "The PR_COLLAPSIBLE_TEMPLATE has been updated to use a multi-line string, which improves readability.",
+    "confidence": "moderate",
+    "reason": "Multi-line strings are more readable and maintainable for long string templates.",
+    "solution": "Consider using f-strings for better performance and readability if Python 3.6+ is supported.",
+    "actual_code": "PR_COLLAPSIBLE_TEMPLATE = \"\"\"\n<details>\n<summary>Review Comment</summary>\n<p>{comment}</p>\n<p><strong>Reason:</strong>{reason}</p>\n<p><strong>Solution:</strong>{solution}</p>\n<p><strong>Confidence:</strong>{confidence}</p>\n<p><strong>Start Line:</strong>{start_line}</p>\n<p><strong>End Line:</strong>{end_line}</p>\n<p><strong>File Name:</strong>{file_name}</p>\n<p><strong>Severity:</strong>{severity}</p>\n</details>\n\"\"\"",
+    "fixed_code": "PR_COLLAPSIBLE_TEMPLATE = f\"\"\"\n<details>\n<summary>Review Comment</summary>\n<p>{{comment}}</p>\n<p><strong>Reason:</strong>{{reason}}</p>\n<p><strong>Solution:</strong>{{solution}}</p>\n<p><strong>Confidence:</strong>{{confidence}}</p>\n<p><strong>Start Line:</strong>{{start_line}}</p>\n<p><strong>End Line:</strong>{{end_line}}</p>\n<p><strong>File Name:</strong>{{file_name}}</p>\n<p><strong>Severity:</strong>{{severity}}</p>\n</details>\n\"\"\"",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 4,
+    "end_line": 16,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Test Case Design",
+    "comment": "New test cases for handling missing fields and empty lists have been added, which improves robustness.",
+    "confidence": "important",
+    "reason": "Testing edge cases and error handling is crucial for ensuring the reliability of the code.",
+    "solution": "Continue to add test cases for other potential edge cases and error scenarios.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 142,
+    "end_line": 276,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 7
+  },
+  {
+    "topic": "Test Coverage",
+    "comment": "The new tests provide better coverage and include more edge cases.",
+    "confidence": "important",
+    "reason": "Comprehensive test coverage is crucial for maintaining code quality and preventing regressions.",
+    "solution": "No changes needed. The new tests are well-structured and cover various scenarios.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 25,
+    "end_line": 246,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 2
+  },
+  {
+    "topic": "Code Organization",
+    "comment": "The test functions are well-organized and use pytest fixtures effectively.",
+    "confidence": "important",
+    "reason": "Good organization improves readability and maintainability of test code.",
+    "solution": "No changes needed. The use of fixtures and utility functions is appropriate.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 7,
+    "end_line": 23,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 2
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "The tests now include checks for file writing permission issues, which is a good practice.",
+    "confidence": "important",
+    "reason": "Testing error handling scenarios ensures the code behaves correctly under various conditions.",
+    "solution": "No changes needed. The error handling tests are well-implemented.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 175,
+    "end_line": 199,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 2
+  },
+  {
+    "topic": "Test Case Naming",
+    "comment": "Test function names are descriptive and follow a consistent naming convention.",
+    "confidence": "moderate",
+    "reason": "Clear and consistent naming improves code readability and helps understand test purposes.",
+    "solution": "No changes needed. The test function names are appropriate.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 25,
+    "end_line": 246,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Test Implementation",
+    "comment": "The test_get_parent_folder_normal function doesn't mock os.getcwd, which may lead to inconsistent results.",
+    "confidence": "important",
+    "reason": "Not mocking external dependencies can make tests unreliable and environment-dependent.",
+    "solution": "Mock os.getcwd to ensure consistent test results across different environments.",
+    "actual_code": "def test_get_parent_folder_normal():\n    expected = os.path.dirname(os.getcwd())\n    result = get_parent_folder()\n    assert result == expected, f\"Expected{expected}, but got{result}\"",
+    "fixed_code": "def test_get_parent_folder_normal():\n    with mock.patch('os.getcwd', return_value='/home/user/project'):\n        expected = '/home/user'\n        result = get_parent_folder()\n        assert result == expected, f\"Expected{expected}, but got{result}\"",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py",
+    "start_line": 13,
+    "end_line": 16,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Test Structure",
+    "comment": "The test structure has been significantly improved with the use of parametrized tests.",
+    "confidence": "important",
+    "reason": "Parametrized tests allow for more comprehensive testing with less code duplication.",
+    "solution": "The changes are already an improvement. Consider adding more test cases to cover edge cases.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 20,
+    "end_line": 75,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Asynchronous Testing",
+    "comment": "The tests have been updated to use async/await syntax, which is correct for testing asynchronous functions.",
+    "confidence": "critical",
+    "reason": "Proper async testing is crucial for accurate results when testing asynchronous code.",
+    "solution": "The implementation is correct. Ensure all test functions are defined with 'async def'.",
+    "actual_code": "async def test_get_web_html_normal_cases(mock_get_html, mock_nest_asyncio, html_content, expected_output):",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 76,
+    "end_line": 76,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 8
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "A new test case for invalid URL has been added, which is a good practice for error handling.",
+    "confidence": "important",
+    "reason": "Testing error scenarios is crucial for robust code.",
+    "solution": "The implementation is correct. Consider adding more error scenarios if applicable.",
+    "actual_code": "async def test_get_web_html_invalid_url(mock_get_html, mock_nest_asyncio):",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 85,
+    "end_line": 91,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 7
+  },
+  {
+    "topic": "Performance Testing",
+    "comment": "A new test case for large content has been added, which is good for performance testing.",
+    "confidence": "moderate",
+    "reason": "Testing with large inputs helps identify potential performance issues.",
+    "solution": "The implementation is good. Consider adding assertions for execution time if performance is critical.",
+    "actual_code": "async def test_get_web_html_large_content(mock_get_html, mock_nest_asyncio):",
+    "fixed_code": "",
+    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 93,
+    "end_line": 102,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 5
+  },
+  {
+    "topic": "Configuration Update",
+    "comment": "New 'base_model' fields have been added to the configuration file.",
+    "confidence": "important",
+    "reason": "Keeping configuration up-to-date is crucial for proper system functionality.",
+    "solution": "Ensure that the code using this configuration is updated to handle the new 'base_model' field.",
+    "actual_code": "\"base_model\": \"azure/gpt-4o-mini\"",
+    "fixed_code": "",
+    "file_name": "config.json",
+    "start_line": 15,
+    "end_line": 15,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 6
+  },
+  {
+    "topic": "Test Generation",
+    "comment": "The test generation call has been updated with new parameters.",
+    "confidence": "moderate",
+    "reason": "New parameters may affect the test generation process and output.",
+    "solution": "Ensure that the 'enable_critique' and 'verbose' options are properly handled in the generator code.",
+    "actual_code": "generator.generate_tests(\n    file_path=\"kaizen/helpers/output.py\", enable_critique=True, verbose=True\n)",
+    "fixed_code": "",
+    "file_name": "examples/unittest/main.py",
+    "start_line": 35,
+    "end_line": 37,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Code Structure",
+    "comment": "The class has been significantly refactored with improved modularity and organization.",
+    "confidence": "important",
+    "reason": "The changes introduce better separation of concerns and more focused methods.",
+    "solution": "No immediate action required, but continue to monitor for potential further improvements.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/unit_test.py",
+    "start_line": 24,
+    "end_line": 40,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "The _read_file_content method lacks error handling for file operations.",
+    "confidence": "important",
+    "reason": "File operations can fail due to various reasons (e.g., permissions, file not found).",
+    "solution": "Add try-except block to handle potential IOError or FileNotFoundError.",
+    "actual_code": "def _read_file_content(self, file_path):\n    with open(file_path, \"r\") as file:\n        return file.read()",
+    "fixed_code": "def _read_file_content(self, file_path):\n    try:\n        with open(file_path, \"r\") as file:\n            return file.read()\n    except (IOError, FileNotFoundError) as e:\n        self.logger.error(f\"Error reading file{file_path}:{e}\")\n        raise",
+    "file_name": "kaizen/generator/unit_test.py",
+    "start_line": 108,
+    "end_line": 110,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Code Duplication",
+    "comment": "The update_usage method duplicates functionality already present in the LLMProvider class.",
+    "confidence": "moderate",
+    "reason": "Duplicating functionality can lead to maintenance issues and inconsistencies.",
+    "solution": "Consider removing the update_usage method and directly using the provider's method.",
+    "actual_code": "def update_usage(self, usage):\n    self.total_usage = self.provider.update_usage(self.total_usage, usage)\n    print(f\"@ Token usage: current_step:{usage}, total:{self.total_usage}\")",
+    "fixed_code": "",
+    "file_name": "kaizen/generator/unit_test.py",
+    "start_line": 273,
+    "end_line": 275,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 5
+  },
+  {
+    "topic": "Code Improvement",
+    "comment": "The _review_tests_by_critique method could benefit from a more descriptive variable name for 'counter'.",
+    "confidence": "low",
+    "reason": "Using more descriptive variable names enhances code readability.",
+    "solution": "Rename 'counter' to 'critique_attempt' or similar for better clarity.",
+    "actual_code": "counter = 1",
+    "fixed_code": "critique_attempt = 1",
+    "file_name": "kaizen/generator/unit_test.py",
+    "start_line": 180,
+    "end_line": 180,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 2
+  },
+  {
+    "topic": "Logging Configuration",
+    "comment": "The new logging configuration might override the existing logging setup",
+    "confidence": "important",
+    "reason": "The added function `set_all_loggers_to_ERROR()` sets all loggers to ERROR level, which may interfere with the existing logging configuration set by `logging.basicConfig()`",
+    "solution": "Consider removing the `set_all_loggers_to_ERROR()` function or adjusting it to respect the `LOGLEVEL` environment variable",
+    "actual_code": "set_all_loggers_to_ERROR()",
+    "fixed_code": "# Consider removing this line or adjusting the function to respect LOGLEVEL\n# set_all_loggers_to_ERROR()",
+    "file_name": "kaizen/llms/provider.py",
+    "start_line": 23,
+    "end_line": 23,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Code Organization",
+    "comment": "The logging configuration is mixed with import statements",
+    "confidence": "moderate",
+    "reason": "Logging setup is typically done after imports for better code organization",
+    "solution": "Move the logging configuration to a separate section after all imports",
+    "actual_code": "def set_all_loggers_to_ERROR():\n    print(\"All Loggers and their levels:\")\n    for name, logger in logging.Logger.manager.loggerDict.items():\n        if isinstance(logger, logging.Logger):\n            print(f\"Logger:{name}, Level:{logging.getLevelName(logger.level)}\")\n            logging.getLogger(name).setLevel(logging.ERROR)\n        else:\n            print(f\"PlaceHolder:{name}\")\n\n\nset_all_loggers_to_ERROR()\n\n# Set litellm log level to ERROR\nlogging.getLogger(\"LiteLLM\").setLevel(logging.ERROR)\nlogging.getLogger(\"LiteLLM Router\").setLevel(logging.ERROR)\nlogging.getLogger(\"LiteLLM Proxy\").setLevel(logging.ERROR)",
+    "fixed_code": "import logging\n\n# Rest of the imports...\n\n# Logging configuration\ndef set_all_loggers_to_ERROR():\n    print(\"All Loggers and their levels:\")\n    for name, logger in logging.Logger.manager.loggerDict.items():\n        if isinstance(logger, logging.Logger):\n            print(f\"Logger:{name}, Level:{logging.getLevelName(logger.level)}\")\n            logging.getLogger(name).setLevel(logging.ERROR)\n        else:\n            print(f\"PlaceHolder:{name}\")\n\nset_all_loggers_to_ERROR()\n\n# Set litellm log level to ERROR\nlogging.getLogger(\"LiteLLM\").setLevel(logging.ERROR)\nlogging.getLogger(\"LiteLLM Router\").setLevel(logging.ERROR)\nlogging.getLogger(\"LiteLLM Proxy\").setLevel(logging.ERROR)",
+    "file_name": "kaizen/llms/provider.py",
+    "start_line": 13,
+    "end_line": 28,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "11",
+    "end_line": "11",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_400/review.md b/.experiments/code_review/sonnet-3.5/no_eval/pr_400/review.md
new file mode 100644
index 00000000..73268d79
--- /dev/null
+++ b/.experiments/code_review/sonnet-3.5/no_eval/pr_400/review.md
@@ -0,0 +1,435 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/400
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 27
+- Critical: 2
+- Important: 15
+- Minor: 9
+- Files Affected: 11
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Asynchronous Testing (2 issues)</strong></summary>
+
+### 1. The tests have been updated to use async/await syntax, which is correct for testing asynchronous functions.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_web_html.py:76`
+⚖️ **Severity:** 8/10
+🔍 **Description:** Proper async testing is crucial for accurate results when testing asynchronous code.
+💡 **Solution:** The implementation is correct. Ensure all test functions are defined with 'async def'.
+
+**Current Code:**
+```python
+async def test_get_web_html_normal_cases(mock_get_html, mock_nest_asyncio, html_content, expected_output):
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 2. Changes made to sensitive file
+📁 **File:** `config.json:11`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to config.json, which needs review
+💡 **Solution:** NA
+
+</details>
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Test Coverage (15 issues)</strong></summary>
+
+### 1. Improved test coverage with more comprehensive test cases
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_description.py:7`
+⚖️ **Severity:** 2/10
+🔍 **Description:** The new test cases cover a wider range of scenarios, including edge cases and error handling
+💡 **Solution:** No changes needed. The improvements are good.
+
+### 2. Added error handling tests for invalid input types
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_description.py:31`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Proper error handling is crucial for robust code
+💡 **Solution:** No changes needed. The added error handling tests are beneficial.
+
+### 3. Improved test structure using pytest.mark.parametrize
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_description.py:7`
+⚖️ **Severity:** 2/10
+🔍 **Description:** Parameterized tests reduce code duplication and make it easier to add new test cases
+💡 **Solution:** No changes needed. The use of pytest.mark.parametrize is a good practice.
+
+### 4. New test cases have been added to improve test coverage, which is a positive change.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py:18`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Comprehensive test coverage is crucial for maintaining code quality and catching potential bugs.
+💡 **Solution:** Continue to add test cases for edge cases and ensure all new functionality is covered by tests.
+
+### 5. New test cases for handling missing fields and empty lists have been added, which improves robustness.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py:142`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Testing edge cases and error handling is crucial for ensuring the reliability of the code.
+💡 **Solution:** Continue to add test cases for other potential edge cases and error scenarios.
+
+### 6. The new tests provide better coverage and include more edge cases.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_test_files.py:25`
+⚖️ **Severity:** 2/10
+🔍 **Description:** Comprehensive test coverage is crucial for maintaining code quality and preventing regressions.
+💡 **Solution:** No changes needed. The new tests are well-structured and cover various scenarios.
+
+### 7. The test functions are well-organized and use pytest fixtures effectively.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_test_files.py:7`
+⚖️ **Severity:** 2/10
+🔍 **Description:** Good organization improves readability and maintainability of test code.
+💡 **Solution:** No changes needed. The use of fixtures and utility functions is appropriate.
+
+### 8. The tests now include checks for file writing permission issues, which is a good practice.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_test_files.py:175`
+⚖️ **Severity:** 2/10
+🔍 **Description:** Testing error handling scenarios ensures the code behaves correctly under various conditions.
+💡 **Solution:** No changes needed. The error handling tests are well-implemented.
+
+### 9. The test_get_parent_folder_normal function doesn't mock os.getcwd, which may lead to inconsistent results.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py:13`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Not mocking external dependencies can make tests unreliable and environment-dependent.
+💡 **Solution:** Mock os.getcwd to ensure consistent test results across different environments.
+
+**Current Code:**
+```python
+def test_get_parent_folder_normal():
+    expected = os.path.dirname(os.getcwd())
+    result = get_parent_folder()
+    assert result == expected, f"Expected{expected}, but got{result}"
+```
+
+**Suggested Code:**
+```python
+def test_get_parent_folder_normal():
+    with mock.patch('os.getcwd', return_value='/home/user/project'):
+        expected = '/home/user'
+        result = get_parent_folder()
+        assert result == expected, f"Expected{expected}, but got{result}"
+```
+
+### 10. The test structure has been significantly improved with the use of parametrized tests.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_web_html.py:20`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Parametrized tests allow for more comprehensive testing with less code duplication.
+💡 **Solution:** The changes are already an improvement. Consider adding more test cases to cover edge cases.
+
+### 11. A new test case for invalid URL has been added, which is a good practice for error handling.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_web_html.py:85`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Testing error scenarios is crucial for robust code.
+💡 **Solution:** The implementation is correct. Consider adding more error scenarios if applicable.
+
+**Current Code:**
+```python
+async def test_get_web_html_invalid_url(mock_get_html, mock_nest_asyncio):
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 12. New 'base_model' fields have been added to the configuration file.
+📁 **File:** `config.json:15`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Keeping configuration up-to-date is crucial for proper system functionality.
+💡 **Solution:** Ensure that the code using this configuration is updated to handle the new 'base_model' field.
+
+**Current Code:**
+```python
+"base_model": "azure/gpt-4o-mini"
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 13. The class has been significantly refactored with improved modularity and organization.
+📁 **File:** `kaizen/generator/unit_test.py:24`
+⚖️ **Severity:** 3/10
+🔍 **Description:** The changes introduce better separation of concerns and more focused methods.
+💡 **Solution:** No immediate action required, but continue to monitor for potential further improvements.
+
+### 14. The _read_file_content method lacks error handling for file operations.
+📁 **File:** `kaizen/generator/unit_test.py:108`
+⚖️ **Severity:** 7/10
+🔍 **Description:** File operations can fail due to various reasons (e.g., permissions, file not found).
+💡 **Solution:** Add try-except block to handle potential IOError or FileNotFoundError.
+
+**Current Code:**
+```python
+def _read_file_content(self, file_path):
+    with open(file_path, "r") as file:
+        return file.read()
+```
+
+**Suggested Code:**
+```python
+def _read_file_content(self, file_path):
+    try:
+        with open(file_path, "r") as file:
+            return file.read()
+    except (IOError, FileNotFoundError) as e:
+        self.logger.error(f"Error reading file{file_path}:{e}")
+        raise
+```
+
+### 15. The new logging configuration might override the existing logging setup
+📁 **File:** `kaizen/llms/provider.py:23`
+⚖️ **Severity:** 7/10
+🔍 **Description:** The added function `set_all_loggers_to_ERROR()` sets all loggers to ERROR level, which may interfere with the existing logging configuration set by `logging.basicConfig()`
+💡 **Solution:** Consider removing the `set_all_loggers_to_ERROR()` function or adjusting it to respect the `LOGLEVEL` environment variable
+
+**Current Code:**
+```python
+set_all_loggers_to_ERROR()
+```
+
+**Suggested Code:**
+```python
+# Consider removing this line or adjusting the function to respect LOGLEVEL
+# set_all_loggers_to_ERROR()
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (10 issues)</strong></summary>
+
+<details>
+<summary><strong>Performance Testing (9 issues)</strong></summary>
+
+### 1. Added performance testing for large inputs, but removed arbitrary time limit
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_description.py:61`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Performance testing is important, but the removal of the 1-second boundary condition is a good change as it avoids potential flaky tests
+💡 **Solution:** Consider adding a more flexible performance assertion based on input size
+
+**Current Code:**
+```python
+print(f"Execution time:{execution_time}seconds")
+```
+
+**Suggested Code:**
+```python
+assert execution_time < len(desc + original_desc) * 0.0001, f"Execution time ({execution_time}seconds) exceeded expected limit"
+```
+
+### 2. Updated .flake8 configuration to exclude unit test directory
+📁 **File:** `.flake8:2`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Excluding unit tests from flake8 checks can be beneficial, but it's important to maintain code quality in tests as well
+💡 **Solution:** Consider running flake8 on test files with a separate, less strict configuration
+
+**Current Code:**
+```python
+exclude = docs/*, venv/*, .kaizen/unit_test/*
+```
+
+**Suggested Code:**
+```python
+exclude = docs/*, venv/*
+```
+
+### 3. Added LITELLM_LOG environment variable
+📁 **File:** `.env.example:6`
+⚖️ **Severity:** 2/10
+🔍 **Description:** Adding logging configuration can improve debugging and monitoring
+💡 **Solution:** Ensure this change is documented in the project's README or documentation
+
+**Current Code:**
+```python
+LITELLM_LOG="ERROR"
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 4. The PR_COLLAPSIBLE_TEMPLATE has been updated to use a multi-line string, which improves readability.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py:4`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Multi-line strings are more readable and maintainable for long string templates.
+💡 **Solution:** Consider using f-strings for better performance and readability if Python 3.6+ is supported.
+
+**Current Code:**
+```python
+PR_COLLAPSIBLE_TEMPLATE = """
+<details>
+<summary>Review Comment</summary>
+<p>{comment}</p>
+<p><strong>Reason:</strong>{reason}</p>
+<p><strong>Solution:</strong>{solution}</p>
+<p><strong>Confidence:</strong>{confidence}</p>
+<p><strong>Start Line:</strong>{start_line}</p>
+<p><strong>End Line:</strong>{end_line}</p>
+<p><strong>File Name:</strong>{file_name}</p>
+<p><strong>Severity:</strong>{severity}</p>
+</details>
+"""
+```
+
+**Suggested Code:**
+```python
+PR_COLLAPSIBLE_TEMPLATE = f"""
+<details>
+<summary>Review Comment</summary>
+<p>{{comment}}</p>
+<p><strong>Reason:</strong>{{reason}}</p>
+<p><strong>Solution:</strong>{{solution}}</p>
+<p><strong>Confidence:</strong>{{confidence}}</p>
+<p><strong>Start Line:</strong>{{start_line}}</p>
+<p><strong>End Line:</strong>{{end_line}}</p>
+<p><strong>File Name:</strong>{{file_name}}</p>
+<p><strong>Severity:</strong>{{severity}}</p>
+</details>
+"""
+```
+
+### 5. Test function names are descriptive and follow a consistent naming convention.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_test_files.py:25`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Clear and consistent naming improves code readability and helps understand test purposes.
+💡 **Solution:** No changes needed. The test function names are appropriate.
+
+### 6. A new test case for large content has been added, which is good for performance testing.
+📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_web_html.py:93`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Testing with large inputs helps identify potential performance issues.
+💡 **Solution:** The implementation is good. Consider adding assertions for execution time if performance is critical.
+
+**Current Code:**
+```python
+async def test_get_web_html_large_content(mock_get_html, mock_nest_asyncio):
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 7. The test generation call has been updated with new parameters.
+📁 **File:** `examples/unittest/main.py:35`
+⚖️ **Severity:** 4/10
+🔍 **Description:** New parameters may affect the test generation process and output.
+💡 **Solution:** Ensure that the 'enable_critique' and 'verbose' options are properly handled in the generator code.
+
+**Current Code:**
+```python
+generator.generate_tests(
+    file_path="kaizen/helpers/output.py", enable_critique=True, verbose=True
+)
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 8. The update_usage method duplicates functionality already present in the LLMProvider class.
+📁 **File:** `kaizen/generator/unit_test.py:273`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Duplicating functionality can lead to maintenance issues and inconsistencies.
+💡 **Solution:** Consider removing the update_usage method and directly using the provider's method.
+
+**Current Code:**
+```python
+def update_usage(self, usage):
+    self.total_usage = self.provider.update_usage(self.total_usage, usage)
+    print(f"@ Token usage: current_step:{usage}, total:{self.total_usage}")
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 9. The logging configuration is mixed with import statements
+📁 **File:** `kaizen/llms/provider.py:13`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Logging setup is typically done after imports for better code organization
+💡 **Solution:** Move the logging configuration to a separate section after all imports
+
+**Current Code:**
+```python
+def set_all_loggers_to_ERROR():
+    print("All Loggers and their levels:")
+    for name, logger in logging.Logger.manager.loggerDict.items():
+        if isinstance(logger, logging.Logger):
+            print(f"Logger:{name}, Level:{logging.getLevelName(logger.level)}")
+            logging.getLogger(name).setLevel(logging.ERROR)
+        else:
+            print(f"PlaceHolder:{name}")
+
+
+set_all_loggers_to_ERROR()
+
+# Set litellm log level to ERROR
+logging.getLogger("LiteLLM").setLevel(logging.ERROR)
+logging.getLogger("LiteLLM Router").setLevel(logging.ERROR)
+logging.getLogger("LiteLLM Proxy").setLevel(logging.ERROR)
+```
+
+**Suggested Code:**
+```python
+import logging
+
+# Rest of the imports...
+
+# Logging configuration
+def set_all_loggers_to_ERROR():
+    print("All Loggers and their levels:")
+    for name, logger in logging.Logger.manager.loggerDict.items():
+        if isinstance(logger, logging.Logger):
+            print(f"Logger:{name}, Level:{logging.getLevelName(logger.level)}")
+            logging.getLogger(name).setLevel(logging.ERROR)
+        else:
+            print(f"PlaceHolder:{name}")
+
+set_all_loggers_to_ERROR()
+
+# Set litellm log level to ERROR
+logging.getLogger("LiteLLM").setLevel(logging.ERROR)
+logging.getLogger("LiteLLM Router").setLevel(logging.ERROR)
+logging.getLogger("LiteLLM Proxy").setLevel(logging.ERROR)
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (anthropic.claude-3-5-sonnet-20240620-v1:0)
+{"prompt_tokens": 42996, "completion_tokens": 6157, "total_tokens": 49153}
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_476/comments.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_476/comments.json
new file mode 100644
index 00000000..a9d40eac
--- /dev/null
+++ b/.experiments/code_review/sonnet-3.5/no_eval/pr_476/comments.json
@@ -0,0 +1,16 @@
+[
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_476/issues.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_476/issues.json
new file mode 100644
index 00000000..c12e22ed
--- /dev/null
+++ b/.experiments/code_review/sonnet-3.5/no_eval/pr_476/issues.json
@@ -0,0 +1,106 @@
+[
+  {
+    "topic": "Error Handling",
+    "comment": "Exception handling is too broad and prints a generic error message.",
+    "confidence": "important",
+    "reason": "Catching all exceptions and printing a generic error message can hide important errors and make debugging difficult.",
+    "solution": "Catch specific exceptions and provide more informative error messages.",
+    "actual_code": "except Exception:\n    print(\"Error\")",
+    "fixed_code": "except KeyError as e:\n    print(f\"Invalid confidence level:{e}\")\nexcept Exception as e:\n    print(f\"Unexpected error:{e}\")",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 140,
+    "end_line": 141,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Code Efficiency",
+    "comment": "The sort_files function implements a manual insertion sort, which is inefficient for large lists.",
+    "confidence": "important",
+    "reason": "Insertion sort has O(n^2) time complexity, which can be slow for large numbers of files.",
+    "solution": "Use Python's built-in sorted() function with a key function for better performance.",
+    "actual_code": "def sort_files(files):\n    sorted_files =[]\n    for file in files:\n        min_index = len(sorted_files)\n        file_name = file[\"filename\"]\n        for i, sorted_file in enumerate(sorted_files):\n            if file_name < sorted_file[\"filename\"]:\n                min_index = i\n                break\n        sorted_files.insert(min_index, file)\n    return sorted_files",
+    "fixed_code": "def sort_files(files):\n    return sorted(files, key=lambda x: x[\"filename\"])",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 184,
+    "end_line": 194,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Code Simplification",
+    "comment": "The generate_tests function can be simplified using a list comprehension.",
+    "confidence": "moderate",
+    "reason": "The current implementation is unnecessarily verbose for a simple operation.",
+    "solution": "Use a list comprehension to create the list of filenames.",
+    "actual_code": "def generate_tests(pr_files):\n    return[f[\"filename\"] for f in pr_files]",
+    "fixed_code": "def generate_tests(pr_files):\n    return[f[\"filename\"] for f in pr_files]",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 199,
+    "end_line": 200,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 2
+  },
+  {
+    "topic": "Code Consistency",
+    "comment": "Inconsistent use of print statements for debugging.",
+    "confidence": "low",
+    "reason": "Some print statements are commented out while others are added, which may lead to inconsistent debugging output.",
+    "solution": "Decide on a consistent approach for debug logging, preferably using a proper logging system.",
+    "actual_code": "print(\"diff: \", diff_text)\nprint(\"pr_files\", pr_files)",
+    "fixed_code": "import logging\n\nlogging.debug(f\"diff:{diff_text}\")\nlogging.debug(f\"pr_files:{pr_files}\")",
+    "file_name": "examples/code_review/main.py",
+    "start_line": 21,
+    "end_line": 22,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 3
+  },
+  {
+    "topic": "Code Improvement",
+    "comment": "The create_pr_review_text function now includes a code_quality parameter, which is a good improvement.",
+    "confidence": "moderate",
+    "reason": "Including code quality in the review text provides more comprehensive feedback.",
+    "solution": "No change needed, this is a positive improvement.",
+    "actual_code": "review_desc = create_pr_review_text(topics, code_quality=review_data.code_quality)",
+    "fixed_code": "",
+    "file_name": "examples/code_review/main.py",
+    "start_line": 36,
+    "end_line": 36,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 1
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Removal of 'enable_observability_logging' from config.json",
+    "confidence": "moderate",
+    "reason": "Removing configuration options without proper documentation or migration path can lead to issues for existing users.",
+    "solution": "If the feature is no longer supported, provide a migration guide or deprecation notice.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "config.json",
+    "start_line": 4,
+    "end_line": 4,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_476/review.md b/.experiments/code_review/sonnet-3.5/no_eval/pr_476/review.md
new file mode 100644
index 00000000..0c63220f
--- /dev/null
+++ b/.experiments/code_review/sonnet-3.5/no_eval/pr_476/review.md
@@ -0,0 +1,154 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/476
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 7
+- Critical: 1
+- Important: 2
+- Minor: 3
+- Files Affected: 3
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Configuration (1 issues)</strong></summary>
+
+### 1. Changes made to sensitive file
+📁 **File:** `config.json:1`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes were made to config.json, which needs review
+💡 **Solution:** NA
+
+</details>
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Error Handling (2 issues)</strong></summary>
+
+### 1. Exception handling is too broad and prints a generic error message.
+📁 **File:** `github_app/github_helper/pull_requests.py:140`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Catching all exceptions and printing a generic error message can hide important errors and make debugging difficult.
+💡 **Solution:** Catch specific exceptions and provide more informative error messages.
+
+**Current Code:**
+```python
+except Exception:
+    print("Error")
+```
+
+**Suggested Code:**
+```python
+except KeyError as e:
+    print(f"Invalid confidence level:{e}")
+except Exception as e:
+    print(f"Unexpected error:{e}")
+```
+
+### 2. The sort_files function implements a manual insertion sort, which is inefficient for large lists.
+📁 **File:** `github_app/github_helper/pull_requests.py:184`
+⚖️ **Severity:** 6/10
+🔍 **Description:** Insertion sort has O(n^2) time complexity, which can be slow for large numbers of files.
+💡 **Solution:** Use Python's built-in sorted() function with a key function for better performance.
+
+**Current Code:**
+```python
+def sort_files(files):
+    sorted_files =[]
+    for file in files:
+        min_index = len(sorted_files)
+        file_name = file["filename"]
+        for i, sorted_file in enumerate(sorted_files):
+            if file_name < sorted_file["filename"]:
+                min_index = i
+                break
+        sorted_files.insert(min_index, file)
+    return sorted_files
+```
+
+**Suggested Code:**
+```python
+def sort_files(files):
+    return sorted(files, key=lambda x: x["filename"])
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (4 issues)</strong></summary>
+
+<details>
+<summary><strong>Code Simplification (3 issues)</strong></summary>
+
+### 1. The generate_tests function can be simplified using a list comprehension.
+📁 **File:** `github_app/github_helper/pull_requests.py:199`
+⚖️ **Severity:** 2/10
+🔍 **Description:** The current implementation is unnecessarily verbose for a simple operation.
+💡 **Solution:** Use a list comprehension to create the list of filenames.
+
+**Current Code:**
+```python
+def generate_tests(pr_files):
+    return[f["filename"] for f in pr_files]
+```
+
+**Suggested Code:**
+```python
+def generate_tests(pr_files):
+    return[f["filename"] for f in pr_files]
+```
+
+### 2. The create_pr_review_text function now includes a code_quality parameter, which is a good improvement.
+📁 **File:** `examples/code_review/main.py:36`
+⚖️ **Severity:** 1/10
+🔍 **Description:** Including code quality in the review text provides more comprehensive feedback.
+💡 **Solution:** No change needed, this is a positive improvement.
+
+**Current Code:**
+```python
+review_desc = create_pr_review_text(topics, code_quality=review_data.code_quality)
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 3. Removal of 'enable_observability_logging' from config.json
+📁 **File:** `config.json:4`
+⚖️ **Severity:** 4/10
+🔍 **Description:** Removing configuration options without proper documentation or migration path can lead to issues for existing users.
+💡 **Solution:** If the feature is no longer supported, provide a migration guide or deprecation notice.
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (anthropic.claude-3-5-sonnet-20240620-v1:0)
+{"prompt_tokens": 4511, "completion_tokens": 1381, "total_tokens": 5892}
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_5/comments.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_5/comments.json
new file mode 100644
index 00000000..eddca222
--- /dev/null
+++ b/.experiments/code_review/sonnet-3.5/no_eval/pr_5/comments.json
@@ -0,0 +1,47 @@
+[
+  {
+    "topic": "Error Handling",
+    "comment": "The API call in process_applicant() lacks a retry mechanism.",
+    "confidence": "critical",
+    "reason": "Without retries, temporary network issues could cause the application to fail.",
+    "solution": "Implement a retry mechanism with exponential backoff for the API call.",
+    "actual_code": "response = completion(\n    model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n)",
+    "fixed_code": "from tenacity import retry, stop_after_attempt, wait_exponential\n\n@retry(stop=stop_after_attempt(3), wait=wait_exponential(multiplier=1, min=4, max=10))\ndef make_completion_call(model, messages):\n    return completion(model=model, messages=messages)\n\nresponse = make_completion_call(\n    model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"),\n    messages=messages\n)",
+    "file_name": "main.py",
+    "start_line": 66,
+    "end_line": 68,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Potential division by zero when calculating token percentages.",
+    "confidence": "critical",
+    "reason": "If total_tokens is zero, it could lead to a runtime error.",
+    "solution": "Add a check to avoid division by zero when calculating percentages.",
+    "actual_code": "print(\"\\nProcessing Summary:\")\nprint(f\"Total applicants processed:{len(df)}\")\nprint(f\"Total tokens used:{total_tokens:,}\")\nprint(f\"  - Input tokens:{total_input_tokens:,}\")\nprint(f\"  - Output tokens:{total_output_tokens:,}\")",
+    "fixed_code": "print(\"\\nProcessing Summary:\")\nprint(f\"Total applicants processed:{len(df)}\")\nprint(f\"Total tokens used:{total_tokens:,}\")\nif total_tokens > 0:\n    print(f\"  - Input tokens:{total_input_tokens:,}({total_input_tokens/total_tokens:.2%})\")\n    print(f\"  - Output tokens:{total_output_tokens:,}({total_output_tokens/total_tokens:.2%})\")\nelse:\n    print(\"  - No tokens used.\")",
+    "file_name": "main.py",
+    "start_line": 159,
+    "end_line": 163,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Error Logging",
+    "comment": "JSONDecodeError is caught but not logged, leading to silent failures.",
+    "confidence": "critical",
+    "reason": "Silent failures make debugging difficult and may hide important issues.",
+    "solution": "Add logging for the JSONDecodeError to track parsing failures.",
+    "actual_code": "except json.JSONDecodeError:\n    # Critical: Silent failure without logging\n    result ={\n        key: \"\"\n        for key in[\n            \"feedback\",\n            \"review\",\n            \"should_interview\",\n            \"rating\",\n            \"input_tokens\",\n            \"output_tokens\",\n        ]\n}",
+    "fixed_code": "import logging\n\nlogging.basicConfig(level=logging.ERROR)\nlogger = logging.getLogger(__name__)\n\ntry:\n    parsed_content = extract_json(content)\n    # ... existing code ...\nexcept json.JSONDecodeError as e:\n    logger.error(f\"Failed to parse JSON content:{e}\")\n    result ={\n        key: \"\"\n        for key in[\n            \"feedback\",\n            \"review\",\n            \"should_interview\",\n            \"rating\",\n            \"input_tokens\",\n            \"output_tokens\",\n        ]\n}",
+    "file_name": "main.py",
+    "start_line": 82,
+    "end_line": 94,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_5/issues.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_5/issues.json
new file mode 100644
index 00000000..1c775bfe
--- /dev/null
+++ b/.experiments/code_review/sonnet-3.5/no_eval/pr_5/issues.json
@@ -0,0 +1,107 @@
+[
+  {
+    "topic": "Unused Import",
+    "comment": "The 'random' module is imported but never used in the code.",
+    "confidence": "important",
+    "reason": "Unused imports clutter the code and may lead to confusion.",
+    "solution": "Remove the unused import to improve code clarity.",
+    "actual_code": "import random  # Unused import",
+    "fixed_code": "# Remove this line",
+    "file_name": "main.py",
+    "start_line": 8,
+    "end_line": 8,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 3
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "The API call in process_applicant() lacks a retry mechanism.",
+    "confidence": "critical",
+    "reason": "Without retries, temporary network issues could cause the application to fail.",
+    "solution": "Implement a retry mechanism with exponential backoff for the API call.",
+    "actual_code": "response = completion(\n    model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n)",
+    "fixed_code": "from tenacity import retry, stop_after_attempt, wait_exponential\n\n@retry(stop=stop_after_attempt(3), wait=wait_exponential(multiplier=1, min=4, max=10))\ndef make_completion_call(model, messages):\n    return completion(model=model, messages=messages)\n\nresponse = make_completion_call(\n    model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"),\n    messages=messages\n)",
+    "file_name": "main.py",
+    "start_line": 66,
+    "end_line": 68,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 8
+  },
+  {
+    "topic": "Error Logging",
+    "comment": "JSONDecodeError is caught but not logged, leading to silent failures.",
+    "confidence": "critical",
+    "reason": "Silent failures make debugging difficult and may hide important issues.",
+    "solution": "Add logging for the JSONDecodeError to track parsing failures.",
+    "actual_code": "except json.JSONDecodeError:\n    # Critical: Silent failure without logging\n    result ={\n        key: \"\"\n        for key in[\n            \"feedback\",\n            \"review\",\n            \"should_interview\",\n            \"rating\",\n            \"input_tokens\",\n            \"output_tokens\",\n        ]\n}",
+    "fixed_code": "import logging\n\nlogging.basicConfig(level=logging.ERROR)\nlogger = logging.getLogger(__name__)\n\ntry:\n    parsed_content = extract_json(content)\n    # ... existing code ...\nexcept json.JSONDecodeError as e:\n    logger.error(f\"Failed to parse JSON content:{e}\")\n    result ={\n        key: \"\"\n        for key in[\n            \"feedback\",\n            \"review\",\n            \"should_interview\",\n            \"rating\",\n            \"input_tokens\",\n            \"output_tokens\",\n        ]\n}",
+    "file_name": "main.py",
+    "start_line": 82,
+    "end_line": 94,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Performance",
+    "comment": "Inefficient progress printing in non-tqdm mode.",
+    "confidence": "important",
+    "reason": "Frequent console updates can slow down processing, especially for large datasets.",
+    "solution": "Update progress less frequently, e.g., every 1% or 5% of completion.",
+    "actual_code": "print(f\"\\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}\", end=\"\", flush=True)",
+    "fixed_code": "if index % max(1, len(df) // 100) == 0:  # Update every 1%\n    print(f\"\\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}\", end=\"\", flush=True)",
+    "file_name": "main.py",
+    "start_line": 122,
+    "end_line": 122,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 5
+  },
+  {
+    "topic": "Redundant Code",
+    "comment": "Unnecessary check for empty DataFrame in main function.",
+    "confidence": "moderate",
+    "reason": "The check is redundant as an empty DataFrame would not affect the subsequent operations.",
+    "solution": "Remove the unnecessary check to simplify the code.",
+    "actual_code": "if len(df) == 0:\n    return",
+    "fixed_code": "# Remove these lines",
+    "file_name": "main.py",
+    "start_line": 142,
+    "end_line": 143,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 3
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "Potential division by zero when calculating token percentages.",
+    "confidence": "critical",
+    "reason": "If total_tokens is zero, it could lead to a runtime error.",
+    "solution": "Add a check to avoid division by zero when calculating percentages.",
+    "actual_code": "print(\"\\nProcessing Summary:\")\nprint(f\"Total applicants processed:{len(df)}\")\nprint(f\"Total tokens used:{total_tokens:,}\")\nprint(f\"  - Input tokens:{total_input_tokens:,}\")\nprint(f\"  - Output tokens:{total_output_tokens:,}\")",
+    "fixed_code": "print(\"\\nProcessing Summary:\")\nprint(f\"Total applicants processed:{len(df)}\")\nprint(f\"Total tokens used:{total_tokens:,}\")\nif total_tokens > 0:\n    print(f\"  - Input tokens:{total_input_tokens:,}({total_input_tokens/total_tokens:.2%})\")\n    print(f\"  - Output tokens:{total_output_tokens:,}({total_output_tokens/total_tokens:.2%})\")\nelse:\n    print(\"  - No tokens used.\")",
+    "file_name": "main.py",
+    "start_line": 159,
+    "end_line": 163,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Error Handling",
+    "comment": "No error handling for file not found in main function.",
+    "confidence": "important",
+    "reason": "The program may crash if the specified CSV file doesn't exist.",
+    "solution": "Add try-except block to handle FileNotFoundError.",
+    "actual_code": "main(input_file)",
+    "fixed_code": "try:\n    main(input_file)\nexcept FileNotFoundError:\n    print(f\"Error: The file '{input_file}' was not found.\")\nexcept Exception as e:\n    print(f\"An error occurred:{e}\")",
+    "file_name": "main.py",
+    "start_line": 175,
+    "end_line": 175,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_5/review.md b/.experiments/code_review/sonnet-3.5/no_eval/pr_5/review.md
new file mode 100644
index 00000000..a4a27b42
--- /dev/null
+++ b/.experiments/code_review/sonnet-3.5/no_eval/pr_5/review.md
@@ -0,0 +1,234 @@
+PR URL: https://github.com/sauravpanda/applicant-screening/pull/5
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 7
+- Critical: 3
+- Important: 3
+- Minor: 1
+- Files Affected: 1
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Error Handling (3 issues)</strong></summary>
+
+### 1. The API call in process_applicant() lacks a retry mechanism.
+📁 **File:** `main.py:66`
+⚖️ **Severity:** 8/10
+🔍 **Description:** Without retries, temporary network issues could cause the application to fail.
+💡 **Solution:** Implement a retry mechanism with exponential backoff for the API call.
+
+**Current Code:**
+```python
+response = completion(
+    model=os.environ.get("model", "anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1"), messages=messages
+)
+```
+
+**Suggested Code:**
+```python
+from tenacity import retry, stop_after_attempt, wait_exponential
+
+@retry(stop=stop_after_attempt(3), wait=wait_exponential(multiplier=1, min=4, max=10))
+def make_completion_call(model, messages):
+    return completion(model=model, messages=messages)
+
+response = make_completion_call(
+    model=os.environ.get("model", "anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1"),
+    messages=messages
+)
+```
+
+### 2. JSONDecodeError is caught but not logged, leading to silent failures.
+📁 **File:** `main.py:82`
+⚖️ **Severity:** 7/10
+🔍 **Description:** Silent failures make debugging difficult and may hide important issues.
+💡 **Solution:** Add logging for the JSONDecodeError to track parsing failures.
+
+**Current Code:**
+```python
+except json.JSONDecodeError:
+    # Critical: Silent failure without logging
+    result ={
+        key: ""
+        for key in[
+            "feedback",
+            "review",
+            "should_interview",
+            "rating",
+            "input_tokens",
+            "output_tokens",
+        ]
+}
+```
+
+**Suggested Code:**
+```python
+import logging
+
+logging.basicConfig(level=logging.ERROR)
+logger = logging.getLogger(__name__)
+
+try:
+    parsed_content = extract_json(content)
+    # ... existing code ...
+except json.JSONDecodeError as e:
+    logger.error(f"Failed to parse JSON content:{e}")
+    result ={
+        key: ""
+        for key in[
+            "feedback",
+            "review",
+            "should_interview",
+            "rating",
+            "input_tokens",
+            "output_tokens",
+        ]
+}
+```
+
+### 3. Potential division by zero when calculating token percentages.
+📁 **File:** `main.py:159`
+⚖️ **Severity:** 7/10
+🔍 **Description:** If total_tokens is zero, it could lead to a runtime error.
+💡 **Solution:** Add a check to avoid division by zero when calculating percentages.
+
+**Current Code:**
+```python
+print("\nProcessing Summary:")
+print(f"Total applicants processed:{len(df)}")
+print(f"Total tokens used:{total_tokens:,}")
+print(f"  - Input tokens:{total_input_tokens:,}")
+print(f"  - Output tokens:{total_output_tokens:,}")
+```
+
+**Suggested Code:**
+```python
+print("\nProcessing Summary:")
+print(f"Total applicants processed:{len(df)}")
+print(f"Total tokens used:{total_tokens:,}")
+if total_tokens > 0:
+    print(f"  - Input tokens:{total_input_tokens:,}({total_input_tokens/total_tokens:.2%})")
+    print(f"  - Output tokens:{total_output_tokens:,}({total_output_tokens/total_tokens:.2%})")
+else:
+    print("  - No tokens used.")
+```
+
+</details>
+
+## 🟠 Refinement Suggestions:
+These are not critical issues, but addressing them could further improve the code:
+
+<details>
+<summary><strong>Unused Import (3 issues)</strong></summary>
+
+### 1. The 'random' module is imported but never used in the code.
+📁 **File:** `main.py:8`
+⚖️ **Severity:** 3/10
+🔍 **Description:** Unused imports clutter the code and may lead to confusion.
+💡 **Solution:** Remove the unused import to improve code clarity.
+
+**Current Code:**
+```python
+import random  # Unused import
+```
+
+**Suggested Code:**
+```python
+# Remove this line
+```
+
+### 2. Inefficient progress printing in non-tqdm mode.
+📁 **File:** `main.py:122`
+⚖️ **Severity:** 5/10
+🔍 **Description:** Frequent console updates can slow down processing, especially for large datasets.
+💡 **Solution:** Update progress less frequently, e.g., every 1% or 5% of completion.
+
+**Current Code:**
+```python
+print(f"\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}", end="", flush=True)
+```
+
+**Suggested Code:**
+```python
+if index % max(1, len(df) // 100) == 0:  # Update every 1%
+    print(f"\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}", end="", flush=True)
+```
+
+### 3. No error handling for file not found in main function.
+📁 **File:** `main.py:175`
+⚖️ **Severity:** 6/10
+🔍 **Description:** The program may crash if the specified CSV file doesn't exist.
+💡 **Solution:** Add try-except block to handle FileNotFoundError.
+
+**Current Code:**
+```python
+main(input_file)
+```
+
+**Suggested Code:**
+```python
+try:
+    main(input_file)
+except FileNotFoundError:
+    print(f"Error: The file '{input_file}' was not found.")
+except Exception as e:
+    print(f"An error occurred:{e}")
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (1 issues)</strong></summary>
+
+<details>
+<summary><strong>Redundant Code (1 issues)</strong></summary>
+
+### 1. Unnecessary check for empty DataFrame in main function.
+📁 **File:** `main.py:142`
+⚖️ **Severity:** 3/10
+🔍 **Description:** The check is redundant as an empty DataFrame would not affect the subsequent operations.
+💡 **Solution:** Remove the unnecessary check to simplify the code.
+
+**Current Code:**
+```python
+if len(df) == 0:
+    return
+```
+
+**Suggested Code:**
+```python
+# Remove these lines
+```
+
+</details>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (anthropic.claude-3-5-sonnet-20240620-v1:0)
+{"prompt_tokens": 7009, "completion_tokens": 1992, "total_tokens": 9001}
\ No newline at end of file
diff --git a/github_app/github_helper/pull_requests.py b/github_app/github_helper/pull_requests.py
index a4385cad..3c6d7ea9 100644
--- a/github_app/github_helper/pull_requests.py
+++ b/github_app/github_helper/pull_requests.py
@@ -67,7 +67,10 @@ def create_review_comments(topics, confidence_level=4):
     comments = []
     for _, reviews in topics.items():
         for review in reviews:
-            if confidence_mapping[review["confidence"]] > confidence_level:
+            if (
+                confidence_mapping.get(review.get("impact", "low"), 1)
+                > confidence_level
+            ):
                 comments.append(review)
     return comments, topics
 
@@ -145,7 +148,7 @@ def clean_keys(topics, min_confidence=None):
         for review in reviews:
             if not review.get("reasoning"):
                 review["reasoning"] = review["comment"]
-            if confidence_mapping[review["confidence"]] >= min_value:
+            if confidence_mapping[review["impact"]] >= min_value:
                 rev.append(review)
         new_topics[topic] = rev
     return new_topics
@@ -159,7 +162,7 @@ def post_pull_request_comments(url, review, installation_id):
         "event": "REQUEST_CHANGES",
         "comments": [
             {
-                "path": review["file_name"],
+                "path": review["file_path"],
                 "start_line": review["start_line"],
                 "line": review["end_line"],
                 "body": review["comment"],
diff --git a/kaizen/formatters/code_review_formatter.py b/kaizen/formatters/code_review_formatter.py
index e6a37aaa..67ae259b 100644
--- a/kaizen/formatters/code_review_formatter.py
+++ b/kaizen/formatters/code_review_formatter.py
@@ -4,10 +4,9 @@
 def create_pr_review_text(
     reviews: List[Dict], code_quality: float, tests: List = None
 ) -> str:
-
     markdown_output = "# 🔍 Code Review Summary\n\n"
 
-    if sum(1 for review in reviews if review["confidence"] == "critical") == 0:
+    if sum(1 for review in reviews if review["impact"] == "critical") == 0:
         markdown_output += "✅ **All Clear:** This commit looks good! 👍\n\n"
     else:
         markdown_output += (
@@ -32,7 +31,7 @@ def create_pr_review_text(
         "trivial": [],
     }
     for review in reviews:
-        categories[review["confidence"]].append(review)
+        categories.get(review.get("impact", "low"), []).append(review)
 
     # Add issues sections
     for confidence, emoji in [
@@ -91,12 +90,10 @@ def create_pr_review_text(
 
 def create_stats_section(reviews: List[Dict]) -> str:
     total_issues = len(reviews)
-    critical_issues = sum(1 for review in reviews if review["confidence"] == "critical")
-    important_issues = sum(
-        1 for review in reviews if review["confidence"] == "important"
-    )
-    minor_issues = sum(1 for review in reviews if review["confidence"] in ["moderate"])
-    files_affected = len(set(review["file_name"] for review in reviews))
+    critical_issues = sum(1 for review in reviews if review["impact"] == "critical")
+    important_issues = sum(1 for review in reviews if review["impact"] == "important")
+    minor_issues = sum(1 for review in reviews if review["impact"] in ["moderate"])
+    files_affected = len(set(review["file_path"] for review in reviews))
 
     output = "## 📊 Stats\n"
     output += f"- Total Issues: {total_issues}\n"
@@ -120,11 +117,11 @@ def create_issue_section(issue: Dict, index: int) -> str:
     output = f"### {index}. {issue['comment']}\n"
     output += f"📁 **File:** `{issue['file_name']}:{issue['start_line']}`\n"
     output += f"⚖️ **Severity:** {issue['severity_level']}/10\n"
-    output += f"🔍 **Description:** {issue['reason']}\n"
-    output += f"💡 **Solution:** {issue['solution']}\n\n"
-    if issue.get("actual_code", None) or issue.get("fixed_code", ""):
+    output += f"🔍 **Description:** {issue.get('reason', '')}\n"
+    output += f"💡 **Solution:** {issue.get('solution', '')}\n\n"
+    if issue.get("current_code", None) or issue.get("fixed_code", ""):
         output += "**Current Code:**\n"
-        output += f"```python\n{issue.get('actual_code', '')}\n```\n\n"
+        output += f"```python\n{issue.get('current_code', '')}\n```\n\n"
         output += "**Suggested Code:**\n"
         output += f"```python\n{issue.get('fixed_code', '')}\n```\n\n"
     return output
diff --git a/kaizen/llms/prompts/code_review_prompts.py b/kaizen/llms/prompts/code_review_prompts.py
index f262ea06..ff153a87 100644
--- a/kaizen/llms/prompts/code_review_prompts.py
+++ b/kaizen/llms/prompts/code_review_prompts.py
@@ -8,19 +8,19 @@
   "code_quality_percentage": <0_TO_100>,
   "review": [
     {{
-      "topic": "<SECTION_TOPIC>",
-      "comment": "<CONCISE_ISSUE_DESCRIPTION>",
-      "confidence": "critical|important|moderate|low|trivial",
-      "reason": "<ISSUE_REASONING>",
-      "solution": "<HIGH_LEVEL_SOLUTION>",
-      "actual_code": "<PEICE_OF_CODE_WHICH_HAS_ISSUES>",
-      "fixed_code": "<CORRECTED_CODE>",
-      "file_name": "<FULL_FILE_PATH>",
-      "start_line": <START_LINE_NUMBER>,
-      "end_line": <END_LINE_NUMBER>,
-      "side": "LEFT|RIGHT",
+      "topic": "<ISSUE_CATEGORY>",
+      "description": "<CONCISE_ISSUE_DESCRIPTION>",
+      "impact": "critical|high|medium|low|trivial",
+      "rationale": "<DETAILED_EXPLANATION>",
+      "recommendation": "<SPECIFIC_IMPROVEMENT_SUGGESTION>",
+      "current_code": "<PROBLEMATIC_CODE_SNIPPET>",
+      "suggested_code": "<IMPROVED_CODE_SNIPPET>",
+      "file_path": "<FULL_FILE_PATH>",
+      "start_line": <STARTING_LINE_NUMBER>,
+      "end_line": <ENDING_LINE_NUMBER>,
+      "change_type": "addition|deletion|modification",
       "sentiment": "positive|negative|neutral",
-      "severity_level": <1_TO_10>
+      "severity": <1_TO_10>
     }}
   ]
 }}
@@ -57,13 +57,19 @@
 4. A new line 'result += y' was added, incorporating the new parameter.
 5. The return statement remained unchanged.
 
-
 ## Review Focus:
 1. Removals (-1:[-]): Identify if removal causes problems in remaining code. Remember any line having -1:[-] is removed line from the new code.
 2. Additions (+1:[+]): Provide detailed feedback and suggest improvements. Remember any line having +1:[+] is added line.
 3. Consider impact of changes on overall code structure and functionality.
 4. Note unchanged lines (0:[.]) for context.
 5. For 'fixed_code' -> always suggest changes for Additions. 
+6. Evaluate code for adherence to best practices and potential optimizations.
+7. Consider performance implications of the changes.
+8. Check for proper error handling and exception management.
+9. Assess the quality and coverage of unit tests, if present.
+10. Evaluate the clarity and completeness of comments and documentation.
+11. Look for potential security vulnerabilities or privacy concerns.
+12. Consider the scalability and maintainability of the changes.
 
 ## Field Guidelines:
 - "fixed_code": Corrected code for additions only, between start_line and end_line. make sure start_line you suggest does not has `0:[.]`.
@@ -71,6 +77,19 @@
 - "start_line" and "end_line": Actual line numbers in the additions.
 - "severity_level": 1 (least severe) to 10 (most critical).
 
+## Suggestion Guidelines:
+- Prioritize issues based on their potential impact on code quality, functionality, and maintainability.
+- Provide concrete examples or code snippets when suggesting improvements.
+- Consider both short-term fixes and long-term architectural improvements.
+- Suggest automated tools or techniques that could help prevent similar issues in the future.
+- When appropriate, reference relevant design patterns or best practices in your suggestions.
+
+## Holistic Review:
+- Consider how the changes fit into the broader context of the project.
+- Assess whether the changes introduce or resolve technical debt.
+- Evaluate if the changes are consistent with the overall architecture and coding style of the project.
+- Consider potential side effects or unintended consequences of the changes.
+
 ## PATCH DATA:
 ```{CODE_DIFF}```
 """
diff --git a/kaizen/llms/provider.py b/kaizen/llms/provider.py
index 4274d736..4757700b 100644
--- a/kaizen/llms/provider.py
+++ b/kaizen/llms/provider.py
@@ -13,7 +13,7 @@
 
 
 def set_all_loggers_to_ERROR():
-    print("All Loggers and their levels:")
+    # print("All Loggers and their levels:")
     for name, logger in logging.Logger.manager.loggerDict.items():
         if isinstance(logger, logging.Logger):
             # print(f"Logger: {name}, Level: {logging.getLevelName(logger.level)}")
@@ -265,9 +265,12 @@ def update_usage(
     def get_usage_cost(self, total_usage: Dict[str, int], model: str = None) -> float:
         if not model:
             model = self.model
-        return litellm.cost_per_token(
-            model, total_usage["prompt_tokens"], total_usage["completion_tokens"]
-        )
+        try:
+            return litellm.cost_per_token(
+                model, total_usage["prompt_tokens"], total_usage["completion_tokens"]
+            )
+        except Exception:
+            return 0, 0
 
     def get_text_embedding(self, text):
         # for model in self.config["language_model"]["models"]:
diff --git a/kaizen/reviewer/code_review.py b/kaizen/reviewer/code_review.py
index 176bbbd7..37902bac 100644
--- a/kaizen/reviewer/code_review.py
+++ b/kaizen/reviewer/code_review.py
@@ -286,7 +286,7 @@ def _process_file_chunk(
         if reeval_response:
             resp = self._reevaluate_response(prompt, resp, user)
 
-        return resp["review"], resp.get("code_quality_percentage", None)
+        return resp.get("review", []), resp.get("code_quality_percentage", None)
 
     def _reevaluate_response(self, prompt: str, resp: str, user: Optional[str]) -> str:
         new_prompt = PR_REVIEW_EVALUATION_PROMPT.format(
diff --git a/poetry.lock b/poetry.lock
index 9b2c9127..b11ee42b 100644
--- a/poetry.lock
+++ b/poetry.lock
@@ -252,6 +252,47 @@ d = ["aiohttp (>=3.7.4)", "aiohttp (>=3.7.4,!=3.9.0)"]
 jupyter = ["ipython (>=7.8.0)", "tokenize-rt (>=3.2.0)"]
 uvloop = ["uvloop (>=0.15.2)"]
 
+[[package]]
+name = "boto3"
+version = "1.35.5"
+description = "The AWS SDK for Python"
+optional = false
+python-versions = ">=3.8"
+files = [
+    {file = "boto3-1.35.5-py3-none-any.whl", hash = "sha256:2cef3aa476181395c260f4b6e6c5565e5a3022a874fb6b579d8e6b169f94e0b3"},
+    {file = "boto3-1.35.5.tar.gz", hash = "sha256:5724ddeda8e18c7614c20a09c20159ed87ff7439755cf5e250a1a3feaf9afb7e"},
+]
+
+[package.dependencies]
+botocore = ">=1.35.5,<1.36.0"
+jmespath = ">=0.7.1,<2.0.0"
+s3transfer = ">=0.10.0,<0.11.0"
+
+[package.extras]
+crt = ["botocore[crt] (>=1.21.0,<2.0a0)"]
+
+[[package]]
+name = "botocore"
+version = "1.35.5"
+description = "Low-level, data-driven core of boto 3."
+optional = false
+python-versions = ">=3.8"
+files = [
+    {file = "botocore-1.35.5-py3-none-any.whl", hash = "sha256:8116b72c7ae845c195146e437e2afd9d17538a37b3f3548dcf67c12c86ba0742"},
+    {file = "botocore-1.35.5.tar.gz", hash = "sha256:3a0086c7124cb3b0d9f98563d00ffd14a942c3f9e731d8d1ccf0d3a1ac7ed884"},
+]
+
+[package.dependencies]
+jmespath = ">=0.7.1,<2.0.0"
+python-dateutil = ">=2.1,<3.0.0"
+urllib3 = [
+    {version = ">=1.25.4,<1.27", markers = "python_version < \"3.10\""},
+    {version = ">=1.25.4,<2.2.0 || >2.2.0,<3", markers = "python_version >= \"3.10\""},
+]
+
+[package.extras]
+crt = ["awscrt (==0.21.2)"]
+
 [[package]]
 name = "bs4"
 version = "0.0.2"
@@ -1291,6 +1332,17 @@ files = [
     {file = "jiter-0.5.0.tar.gz", hash = "sha256:1d916ba875bcab5c5f7d927df998c4cb694d27dceddf3392e58beaf10563368a"},
 ]
 
+[[package]]
+name = "jmespath"
+version = "1.0.1"
+description = "JSON Matching Expressions"
+optional = false
+python-versions = ">=3.7"
+files = [
+    {file = "jmespath-1.0.1-py3-none-any.whl", hash = "sha256:02e2e4cc71b5bcab88332eebf907519190dd9e6e82107fa7f83b1003a6252980"},
+    {file = "jmespath-1.0.1.tar.gz", hash = "sha256:90261b206d6defd58fdd5e85f478bf633a2901798906be2ad389150c5c60edbe"},
+]
+
 [[package]]
 name = "joblib"
 version = "1.4.2"
@@ -1337,6 +1389,108 @@ files = [
 [package.dependencies]
 referencing = ">=0.31.0"
 
+[[package]]
+name = "levenshtein"
+version = "0.25.1"
+description = "Python extension for computing string edit distances and similarities."
+optional = false
+python-versions = ">=3.8"
+files = [
+    {file = "Levenshtein-0.25.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:eb4d1ec9f2dcbde1757c4b7fb65b8682bc2de45b9552e201988f287548b7abdf"},
+    {file = "Levenshtein-0.25.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:b4d9fa3affef48a7e727cdbd0d9502cd060da86f34d8b3627edd769d347570e2"},
+    {file = "Levenshtein-0.25.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:c1b6cd186e58196ff8b402565317e9346b408d0c04fa0ed12ce4868c0fcb6d03"},
+    {file = "Levenshtein-0.25.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:82637ef5428384dd1812849dd7328992819bf0c4a20bff0a3b3ee806821af7ed"},
+    {file = "Levenshtein-0.25.1-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:e73656da6cc3e32a6e4bcd48562fcb64599ef124997f2c91f5320d7f1532c069"},
+    {file = "Levenshtein-0.25.1-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:5abff796f92cdfba69b9cbf6527afae918d0e95cbfac000bd84017f74e0bd427"},
+    {file = "Levenshtein-0.25.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:38827d82f2ca9cb755da6f03e686866f2f411280db005f4304272378412b4cba"},
+    {file = "Levenshtein-0.25.1-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2b989df1e3231261a87d68dfa001a2070771e178b09650f9cf99a20e3d3abc28"},
+    {file = "Levenshtein-0.25.1-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:2011d3b3897d438a2f88ef7aed7747f28739cae8538ec7c18c33dd989930c7a0"},
+    {file = "Levenshtein-0.25.1-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:6c375b33ec7acc1c6855e8ee8c7c8ac6262576ffed484ff5c556695527f49686"},
+    {file = "Levenshtein-0.25.1-cp310-cp310-musllinux_1_1_ppc64le.whl", hash = "sha256:ce0cb9dd012ef1bf4d5b9d40603e7709b6581aec5acd32fcea9b371b294ca7aa"},
+    {file = "Levenshtein-0.25.1-cp310-cp310-musllinux_1_1_s390x.whl", hash = "sha256:9da9ecb81bae67d784defed7274f894011259b038ec31f2339c4958157970115"},
+    {file = "Levenshtein-0.25.1-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:3bd7be5dbe5f4a1b691f381e39512927b39d1e195bd0ad61f9bf217a25bf36c9"},
+    {file = "Levenshtein-0.25.1-cp310-cp310-win32.whl", hash = "sha256:f6abb9ced98261de67eb495b95e1d2325fa42b0344ed5763f7c0f36ee2e2bdba"},
+    {file = "Levenshtein-0.25.1-cp310-cp310-win_amd64.whl", hash = "sha256:97581af3e0a6d359af85c6cf06e51f77f4d635f7109ff7f8ed7fd634d8d8c923"},
+    {file = "Levenshtein-0.25.1-cp310-cp310-win_arm64.whl", hash = "sha256:9ba008f490788c6d8d5a10735fcf83559965be97e4ef0812db388a84b1cc736a"},
+    {file = "Levenshtein-0.25.1-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:f57d9cf06dac55c2d2f01f0d06e32acc074ab9a902921dc8fddccfb385053ad5"},
+    {file = "Levenshtein-0.25.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:22b60c6d791f4ca67a3686b557ddb2a48de203dae5214f220f9dddaab17f44bb"},
+    {file = "Levenshtein-0.25.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:d0444ee62eccf1e6cedc7c5bc01a9face6ff70cc8afa3f3ca9340e4e16f601a4"},
+    {file = "Levenshtein-0.25.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7e8758be8221a274c83924bae8dd8f42041792565a3c3bdd3c10e3f9b4a5f94e"},
+    {file = "Levenshtein-0.25.1-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:147221cfb1d03ed81d22fdd2a4c7fc2112062941b689e027a30d2b75bbced4a3"},
+    {file = "Levenshtein-0.25.1-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a454d5bc4f4a289f5471418788517cc122fcc00d5a8aba78c54d7984840655a2"},
+    {file = "Levenshtein-0.25.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:5c25f3778bbac78286bef2df0ca80f50517b42b951af0a5ddaec514412f79fac"},
+    {file = "Levenshtein-0.25.1-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:181486cf465aff934694cc9a19f3898a1d28025a9a5f80fc1608217e7cd1c799"},
+    {file = "Levenshtein-0.25.1-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:b8db9f672a5d150706648b37b044dba61f36ab7216c6a121cebbb2899d7dfaa3"},
+    {file = "Levenshtein-0.25.1-cp311-cp311-musllinux_1_1_i686.whl", hash = "sha256:f2a69fe5ddea586d439f9a50d0c51952982f6c0db0e3573b167aa17e6d1dfc48"},
+    {file = "Levenshtein-0.25.1-cp311-cp311-musllinux_1_1_ppc64le.whl", hash = "sha256:3b684675a3bd35efa6997856e73f36c8a41ef62519e0267dcbeefd15e26cae71"},
+    {file = "Levenshtein-0.25.1-cp311-cp311-musllinux_1_1_s390x.whl", hash = "sha256:cc707ef7edb71f6bf8339198b929ead87c022c78040e41668a4db68360129cef"},
+    {file = "Levenshtein-0.25.1-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:41512c436b8c691326e2d07786d906cba0e92b5e3f455bf338befb302a0ca76d"},
+    {file = "Levenshtein-0.25.1-cp311-cp311-win32.whl", hash = "sha256:2a3830175c01ade832ba0736091283f14a6506a06ffe8c846f66d9fbca91562f"},
+    {file = "Levenshtein-0.25.1-cp311-cp311-win_amd64.whl", hash = "sha256:9e0af4e6e023e0c8f79af1d1ca5f289094eb91201f08ad90f426d71e4ae84052"},
+    {file = "Levenshtein-0.25.1-cp311-cp311-win_arm64.whl", hash = "sha256:38e5d9a1d737d7b49fa17d6a4c03a0359288154bf46dc93b29403a9dd0cd1a7d"},
+    {file = "Levenshtein-0.25.1-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:4a40fa16ecd0bf9e557db67131aabeea957f82fe3e8df342aa413994c710c34e"},
+    {file = "Levenshtein-0.25.1-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:4f7d2045d5927cffa65a0ac671c263edbfb17d880fdce2d358cd0bda9bcf2b6d"},
+    {file = "Levenshtein-0.25.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:40f96590539f9815be70e330b4d2efcce0219db31db5a22fffe99565192f5662"},
+    {file = "Levenshtein-0.25.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2d78512dd25b572046ff86d8903bec283c373063349f8243430866b6a9946425"},
+    {file = "Levenshtein-0.25.1-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:c161f24a1b216e8555c874c7dd70c1a0d98f783f252a16c9face920a8b8a6f3e"},
+    {file = "Levenshtein-0.25.1-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:06ebbfd010a00490795f478d18d7fa2ffc79c9c03fc03b678081f31764d16bab"},
+    {file = "Levenshtein-0.25.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:eaa9ec0a4489ebfb25a9ec2cba064ed68d0d2485b8bc8b7203f84a7874755e0f"},
+    {file = "Levenshtein-0.25.1-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:26408938a6db7b252824a701545d50dc9cdd7a3e4c7ee70834cca17953b76ad8"},
+    {file = "Levenshtein-0.25.1-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:330ec2faff957281f4e6a1a8c88286d1453e1d73ee273ea0f937e0c9281c2156"},
+    {file = "Levenshtein-0.25.1-cp312-cp312-musllinux_1_1_i686.whl", hash = "sha256:9115d1b08626dfdea6f3955cb49ba5a578f7223205f80ead0038d6fc0442ce13"},
+    {file = "Levenshtein-0.25.1-cp312-cp312-musllinux_1_1_ppc64le.whl", hash = "sha256:bbd602edab758e93a5c67bf0d8322f374a47765f1cdb6babaf593a64dc9633ad"},
+    {file = "Levenshtein-0.25.1-cp312-cp312-musllinux_1_1_s390x.whl", hash = "sha256:b930b4df32cd3aabbed0e9f0c4fdd1ea4090a5c022ba9f1ae4ab70ccf1cf897a"},
+    {file = "Levenshtein-0.25.1-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:dd66fb51f88a3f73a802e1ff19a14978ddc9fbcb7ce3a667ca34f95ef54e0e44"},
+    {file = "Levenshtein-0.25.1-cp312-cp312-win32.whl", hash = "sha256:386de94bd1937a16ae3c8f8b7dd2eff1b733994ecf56ce4d05dfdd0e776d0261"},
+    {file = "Levenshtein-0.25.1-cp312-cp312-win_amd64.whl", hash = "sha256:9ee1902153d47886c9787598a4a5c324ce7fde44d44daa34fcf3652ac0de21bc"},
+    {file = "Levenshtein-0.25.1-cp312-cp312-win_arm64.whl", hash = "sha256:b56a7e7676093c3aee50402226f4079b15bd21b5b8f1820f9d6d63fe99dc4927"},
+    {file = "Levenshtein-0.25.1-cp38-cp38-macosx_10_9_universal2.whl", hash = "sha256:6b5dfdf6a0e2f35fd155d4c26b03398499c24aba7bc5db40245789c46ad35c04"},
+    {file = "Levenshtein-0.25.1-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:355ff797f704459ddd8b95354d699d0d0642348636c92d5e67b49be4b0e6112b"},
+    {file = "Levenshtein-0.25.1-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:933b827a3b721210fff522f3dca9572f9f374a0e88fa3a6c7ee3164406ae7794"},
+    {file = "Levenshtein-0.25.1-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:be1da669a240f272d904ab452ad0a1603452e190f4e03e886e6b3a9904152b89"},
+    {file = "Levenshtein-0.25.1-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:265cbd78962503a26f2bea096258a3b70b279bb1a74a525c671d3ee43a190f9c"},
+    {file = "Levenshtein-0.25.1-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:63cc4d53a35e673b12b721a58b197b4a65734688fb72aa1987ce63ed612dca96"},
+    {file = "Levenshtein-0.25.1-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:75fee0c471b8799c70dad9d0d5b70f1f820249257f9617601c71b6c1b37bee92"},
+    {file = "Levenshtein-0.25.1-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:045d6b0db124fbd37379b2b91f6d0786c2d9220e7a848e2dd31b99509a321240"},
+    {file = "Levenshtein-0.25.1-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:db7a2e9c51ac9cc2fd5679484f1eac6e0ab2085cb181240445f7fbf10df73230"},
+    {file = "Levenshtein-0.25.1-cp38-cp38-musllinux_1_1_i686.whl", hash = "sha256:c379c588aa0d93d4607db7eb225fd683263d49669b1bbe49e28c978aa6a4305d"},
+    {file = "Levenshtein-0.25.1-cp38-cp38-musllinux_1_1_ppc64le.whl", hash = "sha256:966dd00424df7f69b78da02a29b530fbb6c1728e9002a2925ed7edf26b231924"},
+    {file = "Levenshtein-0.25.1-cp38-cp38-musllinux_1_1_s390x.whl", hash = "sha256:09daa6b068709cc1e68b670a706d928ed8f0b179a26161dd04b3911d9f757525"},
+    {file = "Levenshtein-0.25.1-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:d6bed0792635081accf70a7e11cfece986f744fddf46ce26808cd8bfc067e430"},
+    {file = "Levenshtein-0.25.1-cp38-cp38-win32.whl", hash = "sha256:28e7b7faf5a745a690d1b1706ab82a76bbe9fa6b729d826f0cfdd24fd7c19740"},
+    {file = "Levenshtein-0.25.1-cp38-cp38-win_amd64.whl", hash = "sha256:8ca0cc9b9e07316b5904f158d5cfa340d55b4a3566ac98eaac9f087c6efb9a1a"},
+    {file = "Levenshtein-0.25.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:45682cdb3ac4a5465c01b2dce483bdaa1d5dcd1a1359fab37d26165b027d3de2"},
+    {file = "Levenshtein-0.25.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:f8dc3e63c4cd746ec162a4cd744c6dde857e84aaf8c397daa46359c3d54e6219"},
+    {file = "Levenshtein-0.25.1-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:01ad1eb09933a499a49923e74e05b1428ca4ef37fed32965fef23f1334a11563"},
+    {file = "Levenshtein-0.25.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:cbb4e8c4b8b7bbe0e1aa64710b806b6c3f31d93cb14969ae2c0eff0f3a592db8"},
+    {file = "Levenshtein-0.25.1-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:b48d1fe224b365975002e3e2ea947cbb91d2936a16297859b71c4abe8a39932c"},
+    {file = "Levenshtein-0.25.1-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a164df16d876aab0a400f72aeac870ea97947ea44777c89330e9a16c7dc5cc0e"},
+    {file = "Levenshtein-0.25.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:995d3bcedcf64be6ceca423f6cfe29184a36d7c4cbac199fdc9a0a5ec7196cf5"},
+    {file = "Levenshtein-0.25.1-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:bdaf62d637bef6711d6f3457e2684faab53b2db2ed53c05bc0dc856464c74742"},
+    {file = "Levenshtein-0.25.1-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:af9de3b5f8f5f3530cfd97daab9ab480d1b121ef34d8c0aa5bab0c645eae219e"},
+    {file = "Levenshtein-0.25.1-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:78fba73c352383b356a30c4674e39f086ffef7122fa625e7550b98be2392d387"},
+    {file = "Levenshtein-0.25.1-cp39-cp39-musllinux_1_1_ppc64le.whl", hash = "sha256:9e0df0dcea3943321398f72e330c089b5d5447318310db6f17f5421642f3ade6"},
+    {file = "Levenshtein-0.25.1-cp39-cp39-musllinux_1_1_s390x.whl", hash = "sha256:387f768bb201b9bc45f0f49557e2fb9a3774d9d087457bab972162dcd4fd352b"},
+    {file = "Levenshtein-0.25.1-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:5dcf931b64311039b43495715e9b795fbd97ab44ba3dd6bf24360b15e4e87649"},
+    {file = "Levenshtein-0.25.1-cp39-cp39-win32.whl", hash = "sha256:2449f8668c0bd62a2b305a5e797348984c06ac20903b38b3bab74e55671ddd51"},
+    {file = "Levenshtein-0.25.1-cp39-cp39-win_amd64.whl", hash = "sha256:28803fd6ec7b58065621f5ec0d24e44e2a7dc4842b64dcab690cb0a7ea545210"},
+    {file = "Levenshtein-0.25.1-cp39-cp39-win_arm64.whl", hash = "sha256:0b074d452dff8ee86b5bdb6031aa32bb2ed3c8469a56718af5e010b9bb5124dc"},
+    {file = "Levenshtein-0.25.1-pp38-pypy38_pp73-macosx_10_9_x86_64.whl", hash = "sha256:e9e060ef3925a68aeb12276f0e524fb1264592803d562ec0306c7c3f5c68eae0"},
+    {file = "Levenshtein-0.25.1-pp38-pypy38_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5f84b84049318d44722db307c448f9dcb8d27c73525a378e901189a94889ba61"},
+    {file = "Levenshtein-0.25.1-pp38-pypy38_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:07e23fdf330cb185a0c7913ca5bd73a189dfd1742eae3a82e31ed8688b191800"},
+    {file = "Levenshtein-0.25.1-pp38-pypy38_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d06958e4a81ea0f0b2b7768a2ad05bcd50a9ad04c4d521dd37d5730ff12decdc"},
+    {file = "Levenshtein-0.25.1-pp38-pypy38_pp73-win_amd64.whl", hash = "sha256:2ea7c34ec22b2fce21299b0caa6dde6bdebafcc2970e265853c9cfea8d1186da"},
+    {file = "Levenshtein-0.25.1-pp39-pypy39_pp73-macosx_10_9_x86_64.whl", hash = "sha256:fddc0ccbdd94f57aa32e2eb3ac8310d08df2e175943dc20b3e1fc7a115850af4"},
+    {file = "Levenshtein-0.25.1-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7d52249cb3448bfe661d3d7db3a6673e835c7f37b30b0aeac499a1601bae873d"},
+    {file = "Levenshtein-0.25.1-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e8dd4c201b15f8c1e612f9074335392c8208ac147acbce09aff04e3974bf9b16"},
+    {file = "Levenshtein-0.25.1-pp39-pypy39_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:23a4d95ce9d44161c7aa87ab76ad6056bc1093c461c60c097054a46dc957991f"},
+    {file = "Levenshtein-0.25.1-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:65eea8a9c33037b23069dca4b3bc310e3c28ca53f60ec0c958d15c0952ba39fa"},
+    {file = "Levenshtein-0.25.1.tar.gz", hash = "sha256:2df14471c778c75ffbd59cb64bbecfd4b0ef320ef9f80e4804764be7d5678980"},
+]
+
+[package.dependencies]
+rapidfuzz = ">=3.8.0,<4.0.0"
+
 [[package]]
 name = "litellm"
 version = "1.42.12"
@@ -2701,6 +2855,20 @@ files = [
 [package.extras]
 cli = ["click (>=5.0)"]
 
+[[package]]
+name = "python-levenshtein"
+version = "0.25.1"
+description = "Python extension for computing string edit distances and similarities."
+optional = false
+python-versions = ">=3.8"
+files = [
+    {file = "python-Levenshtein-0.25.1.tar.gz", hash = "sha256:b21e7efe83c8e8dc8260f2143b2393c6c77cb2956f0c53de6c4731c4d8006acc"},
+    {file = "python_Levenshtein-0.25.1-py3-none-any.whl", hash = "sha256:654446d1ea4acbcc573d44c43775854834a7547e4cb2f79f638f738134d72037"},
+]
+
+[package.dependencies]
+Levenshtein = "0.25.1"
+
 [[package]]
 name = "python-slugify"
 version = "8.0.4"
@@ -2839,6 +3007,126 @@ urllib3 = ">=1.26.14,<3"
 fastembed = ["fastembed (==0.3.4)"]
 fastembed-gpu = ["fastembed-gpu (==0.3.4)"]
 
+[[package]]
+name = "rapidfuzz"
+version = "3.9.6"
+description = "rapid fuzzy string matching"
+optional = false
+python-versions = ">=3.8"
+files = [
+    {file = "rapidfuzz-3.9.6-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:a7ed0d0b9c85720f0ae33ac5efc8dc3f60c1489dad5c29d735fbdf2f66f0431f"},
+    {file = "rapidfuzz-3.9.6-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:f3deff6ab7017ed21b9aec5874a07ad13e6b2a688af055837f88b743c7bfd947"},
+    {file = "rapidfuzz-3.9.6-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5c3f9fc060160507b2704f7d1491bd58453d69689b580cbc85289335b14fe8ca"},
+    {file = "rapidfuzz-3.9.6-cp310-cp310-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c4e86c2b3827fa6169ad6e7d4b790ce02a20acefb8b78d92fa4249589bbc7a2c"},
+    {file = "rapidfuzz-3.9.6-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:f982e1aafb4bd8207a5e073b1efef9e68a984e91330e1bbf364f9ed157ed83f0"},
+    {file = "rapidfuzz-3.9.6-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:9196a51d0ec5eaaaf5bca54a85b7b1e666fc944c332f68e6427503af9fb8c49e"},
+    {file = "rapidfuzz-3.9.6-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fb5a514064e02585b1cc09da2fe406a6dc1a7e5f3e92dd4f27c53e5f1465ec81"},
+    {file = "rapidfuzz-3.9.6-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:e3a4244f65dbc3580b1275480118c3763f9dc29fc3dd96610560cb5e140a4d4a"},
+    {file = "rapidfuzz-3.9.6-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:f6ebb910a702e41641e1e1dada3843bc11ba9107a33c98daef6945a885a40a07"},
+    {file = "rapidfuzz-3.9.6-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:624fbe96115fb39addafa288d583b5493bc76dab1d34d0ebba9987d6871afdf9"},
+    {file = "rapidfuzz-3.9.6-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:1c59f1c1507b7a557cf3c410c76e91f097460da7d97e51c985343798e9df7a3c"},
+    {file = "rapidfuzz-3.9.6-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:f6f0256cb27b6a0fb2e1918477d1b56473cd04acfa245376a342e7c15806a396"},
+    {file = "rapidfuzz-3.9.6-cp310-cp310-win32.whl", hash = "sha256:24d473d00d23a30a85802b502b417a7f5126019c3beec91a6739fe7b95388b24"},
+    {file = "rapidfuzz-3.9.6-cp310-cp310-win_amd64.whl", hash = "sha256:248f6d2612e661e2b5f9a22bbd5862a1600e720da7bb6ad8a55bb1548cdfa423"},
+    {file = "rapidfuzz-3.9.6-cp310-cp310-win_arm64.whl", hash = "sha256:e03fdf0e74f346ed7e798135df5f2a0fb8d6b96582b00ebef202dcf2171e1d1d"},
+    {file = "rapidfuzz-3.9.6-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:52e4675f642fbc85632f691b67115a243cd4d2a47bdcc4a3d9a79e784518ff97"},
+    {file = "rapidfuzz-3.9.6-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:1f93a2f13038700bd245b927c46a2017db3dcd4d4ff94687d74b5123689b873b"},
+    {file = "rapidfuzz-3.9.6-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:42b70500bca460264b8141d8040caee22e9cf0418c5388104ff0c73fb69ee28f"},
+    {file = "rapidfuzz-3.9.6-cp311-cp311-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:a1e037fb89f714a220f68f902fc6300ab7a33349f3ce8ffae668c3b3a40b0b06"},
+    {file = "rapidfuzz-3.9.6-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:6792f66d59b86ccfad5e247f2912e255c85c575789acdbad8e7f561412ffed8a"},
+    {file = "rapidfuzz-3.9.6-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:68d9cffe710b67f1969cf996983608cee4490521d96ea91d16bd7ea5dc80ea98"},
+    {file = "rapidfuzz-3.9.6-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:63daaeeea76da17fa0bbe7fb05cba8ed8064bb1a0edf8360636557f8b6511961"},
+    {file = "rapidfuzz-3.9.6-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:d214e063bffa13e3b771520b74f674b22d309b5720d4df9918ff3e0c0f037720"},
+    {file = "rapidfuzz-3.9.6-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:ed443a2062460f44c0346cb9d269b586496b808c2419bbd6057f54061c9b9c75"},
+    {file = "rapidfuzz-3.9.6-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:5b0c9b227ee0076fb2d58301c505bb837a290ae99ee628beacdb719f0626d749"},
+    {file = "rapidfuzz-3.9.6-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:82c9722b7dfaa71e8b61f8c89fed0482567fb69178e139fe4151fc71ed7df782"},
+    {file = "rapidfuzz-3.9.6-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:c18897c95c0a288347e29537b63608a8f63a5c3cb6da258ac46fcf89155e723e"},
+    {file = "rapidfuzz-3.9.6-cp311-cp311-win32.whl", hash = "sha256:3e910cf08944da381159587709daaad9e59d8ff7bca1f788d15928f3c3d49c2a"},
+    {file = "rapidfuzz-3.9.6-cp311-cp311-win_amd64.whl", hash = "sha256:59c4a61fab676d37329fc3a671618a461bfeef53a4d0b8b12e3bc24a14e166f8"},
+    {file = "rapidfuzz-3.9.6-cp311-cp311-win_arm64.whl", hash = "sha256:8b4afea244102332973377fddbe54ce844d0916e1c67a5123432291717f32ffa"},
+    {file = "rapidfuzz-3.9.6-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:70591b28b218fff351b88cdd7f2359a01a71f9f7f5a2e465ce3715ed4b3c422b"},
+    {file = "rapidfuzz-3.9.6-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:ee2d8355c7343c631a03e57540ea06e8717c19ecf5ff64ea07e0498f7f161457"},
+    {file = "rapidfuzz-3.9.6-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:708fb675de0f47b9635d1cc6fbbf80d52cb710d0a1abbfae5c84c46e3abbddc3"},
+    {file = "rapidfuzz-3.9.6-cp312-cp312-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1d66c247c2d3bb7a9b60567c395a15a929d0ebcc5f4ceedb55bfa202c38c6e0c"},
+    {file = "rapidfuzz-3.9.6-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:15146301b32e6e3d2b7e8146db1a26747919d8b13690c7f83a4cb5dc111b3a08"},
+    {file = "rapidfuzz-3.9.6-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a7a03da59b6c7c97e657dd5cd4bcaab5fe4a2affd8193958d6f4d938bee36679"},
+    {file = "rapidfuzz-3.9.6-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0d2c2fe19e392dbc22695b6c3b2510527e2b774647e79936bbde49db7742d6f1"},
+    {file = "rapidfuzz-3.9.6-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:91aaee4c94cb45930684f583ffc4e7c01a52b46610971cede33586cf8a04a12e"},
+    {file = "rapidfuzz-3.9.6-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:3f5702828c10768f9281180a7ff8597da1e5002803e1304e9519dd0f06d79a85"},
+    {file = "rapidfuzz-3.9.6-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:ccd1763b608fb4629a0b08f00b3c099d6395e67c14e619f6341b2c8429c2f310"},
+    {file = "rapidfuzz-3.9.6-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:cc7a0d4b2cb166bc46d02c8c9f7551cde8e2f3c9789df3827309433ee9771163"},
+    {file = "rapidfuzz-3.9.6-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:7496f53d40560a58964207b52586783633f371683834a8f719d6d965d223a2eb"},
+    {file = "rapidfuzz-3.9.6-cp312-cp312-win32.whl", hash = "sha256:5eb1a9272ca71bc72be5415c2fa8448a6302ea4578e181bb7da9db855b367df0"},
+    {file = "rapidfuzz-3.9.6-cp312-cp312-win_amd64.whl", hash = "sha256:0d21fc3c0ca507a1180152a6dbd129ebaef48facde3f943db5c1055b6e6be56a"},
+    {file = "rapidfuzz-3.9.6-cp312-cp312-win_arm64.whl", hash = "sha256:43bb27a57c29dc5fa754496ba6a1a508480d21ae99ac0d19597646c16407e9f3"},
+    {file = "rapidfuzz-3.9.6-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:83a5ac6547a9d6eedaa212975cb8f2ce2aa07e6e30833b40e54a52b9f9999aa4"},
+    {file = "rapidfuzz-3.9.6-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:10f06139142ecde67078ebc9a745965446132b998f9feebffd71acdf218acfcc"},
+    {file = "rapidfuzz-3.9.6-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:74720c3f24597f76c7c3e2c4abdff55f1664f4766ff5b28aeaa689f8ffba5fab"},
+    {file = "rapidfuzz-3.9.6-cp313-cp313-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ce2bce52b5c150878e558a0418c2b637fb3dbb6eb38e4eb27d24aa839920483e"},
+    {file = "rapidfuzz-3.9.6-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:1611199f178793ca9a060c99b284e11f6d7d124998191f1cace9a0245334d219"},
+    {file = "rapidfuzz-3.9.6-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:0308b2ad161daf502908a6e21a57c78ded0258eba9a8f5e2545e2dafca312507"},
+    {file = "rapidfuzz-3.9.6-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3eda91832201b86e3b70835f91522587725bec329ec68f2f7faf5124091e5ca7"},
+    {file = "rapidfuzz-3.9.6-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:ece873c093aedd87fc07c2a7e333d52e458dc177016afa1edaf157e82b6914d8"},
+    {file = "rapidfuzz-3.9.6-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:d97d3c9d209d5c30172baea5966f2129e8a198fec4a1aeb2f92abb6e82a2edb1"},
+    {file = "rapidfuzz-3.9.6-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:6c4550d0db4931f5ebe9f0678916d1b06f06f5a99ba0b8a48b9457fd8959a7d4"},
+    {file = "rapidfuzz-3.9.6-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:b6b8dd4af6324fc325d9483bec75ecf9be33e590928c9202d408e4eafff6a0a6"},
+    {file = "rapidfuzz-3.9.6-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:16122ae448bc89e2bea9d81ce6cb0f751e4e07da39bd1e70b95cae2493857853"},
+    {file = "rapidfuzz-3.9.6-cp313-cp313-win32.whl", hash = "sha256:71cc168c305a4445109cd0d4925406f6e66bcb48fde99a1835387c58af4ecfe9"},
+    {file = "rapidfuzz-3.9.6-cp313-cp313-win_amd64.whl", hash = "sha256:59ee78f2ecd53fef8454909cda7400fe2cfcd820f62b8a5d4dfe930102268054"},
+    {file = "rapidfuzz-3.9.6-cp313-cp313-win_arm64.whl", hash = "sha256:58b4ce83f223605c358ae37e7a2d19a41b96aa65b1fede99cc664c9053af89ac"},
+    {file = "rapidfuzz-3.9.6-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:9f469dbc9c4aeaac7dd005992af74b7dff94aa56a3ea063ce64e4b3e6736dd2f"},
+    {file = "rapidfuzz-3.9.6-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:a9ed7ad9adb68d0fe63a156fe752bbf5f1403ed66961551e749641af2874da92"},
+    {file = "rapidfuzz-3.9.6-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:39ffe48ffbeedf78d120ddfb9d583f2ca906712159a4e9c3c743c9f33e7b1775"},
+    {file = "rapidfuzz-3.9.6-cp38-cp38-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:8502ccdea9084d54b6f737d96a3b60a84e3afed9d016686dc979b49cdac71613"},
+    {file = "rapidfuzz-3.9.6-cp38-cp38-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:6a4bec4956e06b170ca896ba055d08d4c457dac745548172443982956a80e118"},
+    {file = "rapidfuzz-3.9.6-cp38-cp38-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:2c0488b1c273be39e109ff885ccac0448b2fa74dea4c4dc676bcf756c15f16d6"},
+    {file = "rapidfuzz-3.9.6-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0542c036cb6acf24edd2c9e0411a67d7ba71e29e4d3001a082466b86fc34ff30"},
+    {file = "rapidfuzz-3.9.6-cp38-cp38-musllinux_1_2_aarch64.whl", hash = "sha256:0a96b52c9f26857bf009e270dcd829381e7a634f7ddd585fa29b87d4c82146d9"},
+    {file = "rapidfuzz-3.9.6-cp38-cp38-musllinux_1_2_i686.whl", hash = "sha256:6edd3cd7c4aa8c68c716d349f531bd5011f2ca49ddade216bb4429460151559f"},
+    {file = "rapidfuzz-3.9.6-cp38-cp38-musllinux_1_2_ppc64le.whl", hash = "sha256:50b2fb55d7ed58c66d49c9f954acd8fc4a3f0e9fd0ff708299bd8abb68238d0e"},
+    {file = "rapidfuzz-3.9.6-cp38-cp38-musllinux_1_2_s390x.whl", hash = "sha256:32848dfe54391636b84cda1823fd23e5a6b1dbb8be0e9a1d80e4ee9903820994"},
+    {file = "rapidfuzz-3.9.6-cp38-cp38-musllinux_1_2_x86_64.whl", hash = "sha256:29146cb7a1bf69c87e928b31bffa54f066cb65639d073b36e1425f98cccdebc6"},
+    {file = "rapidfuzz-3.9.6-cp38-cp38-win32.whl", hash = "sha256:aed13e5edacb0ecadcc304cc66e93e7e77ff24f059c9792ee602c0381808e10c"},
+    {file = "rapidfuzz-3.9.6-cp38-cp38-win_amd64.whl", hash = "sha256:af440e36b828922256d0b4d79443bf2cbe5515fc4b0e9e96017ec789b36bb9fc"},
+    {file = "rapidfuzz-3.9.6-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:efa674b407424553024522159296690d99d6e6b1192cafe99ca84592faff16b4"},
+    {file = "rapidfuzz-3.9.6-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:0b40ff76ee19b03ebf10a0a87938f86814996a822786c41c3312d251b7927849"},
+    {file = "rapidfuzz-3.9.6-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:16a6c7997cb5927ced6f617122eb116ba514ec6b6f60f4803e7925ef55158891"},
+    {file = "rapidfuzz-3.9.6-cp39-cp39-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:f3f42504bdc8d770987fc3d99964766d42b2a03e4d5b0f891decdd256236bae0"},
+    {file = "rapidfuzz-3.9.6-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:ad9462aa2be9f60b540c19a083471fdf28e7cf6434f068b631525b5e6251b35e"},
+    {file = "rapidfuzz-3.9.6-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:1629698e68f47609a73bf9e73a6da3a4cac20bc710529215cbdf111ab603665b"},
+    {file = "rapidfuzz-3.9.6-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:68bc7621843d8e9a7fd1b1a32729465bf94b47b6fb307d906da168413331f8d6"},
+    {file = "rapidfuzz-3.9.6-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:c6254c50f15bc2fcc33cb93a95a81b702d9e6590f432a7f7822b8c7aba9ae288"},
+    {file = "rapidfuzz-3.9.6-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:7e535a114fa575bc143e175e4ca386a467ec8c42909eff500f5f0f13dc84e3e0"},
+    {file = "rapidfuzz-3.9.6-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:d50acc0e9d67e4ba7a004a14c42d1b1e8b6ca1c515692746f4f8e7948c673167"},
+    {file = "rapidfuzz-3.9.6-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:fa742ec60bec53c5a211632cf1d31b9eb5a3c80f1371a46a23ac25a1fa2ab209"},
+    {file = "rapidfuzz-3.9.6-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:c256fa95d29cbe5aa717db790b231a9a5b49e5983d50dc9df29d364a1db5e35b"},
+    {file = "rapidfuzz-3.9.6-cp39-cp39-win32.whl", hash = "sha256:89acbf728b764421036c173a10ada436ecca22999851cdc01d0aa904c70d362d"},
+    {file = "rapidfuzz-3.9.6-cp39-cp39-win_amd64.whl", hash = "sha256:c608fcba8b14d86c04cb56b203fed31a96e8a1ebb4ce99e7b70313c5bf8cf497"},
+    {file = "rapidfuzz-3.9.6-cp39-cp39-win_arm64.whl", hash = "sha256:d41c00ded0e22e9dba88ff23ebe0dc9d2a5f21ba2f88e185ea7374461e61daa9"},
+    {file = "rapidfuzz-3.9.6-pp310-pypy310_pp73-macosx_10_15_x86_64.whl", hash = "sha256:a65c2f63218ea2dedd56fc56361035e189ca123bd9c9ce63a9bef6f99540d681"},
+    {file = "rapidfuzz-3.9.6-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:680dc78a5f889d3b89f74824b89fe357f49f88ad10d2c121e9c3ad37bac1e4eb"},
+    {file = "rapidfuzz-3.9.6-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b8ca862927a0b05bd825e46ddf82d0724ea44b07d898ef639386530bf9b40f15"},
+    {file = "rapidfuzz-3.9.6-pp310-pypy310_pp73-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2116fa1fbff21fa52cd46f3cfcb1e193ba1d65d81f8b6e123193451cd3d6c15e"},
+    {file = "rapidfuzz-3.9.6-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4dcb7d9afd740370a897c15da61d3d57a8d54738d7c764a99cedb5f746d6a003"},
+    {file = "rapidfuzz-3.9.6-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:1a5bd6401bb489e14cbb5981c378d53ede850b7cc84b2464cad606149cc4e17d"},
+    {file = "rapidfuzz-3.9.6-pp38-pypy38_pp73-macosx_10_9_x86_64.whl", hash = "sha256:29fda70b9d03e29df6fc45cc27cbcc235534b1b0b2900e0a3ae0b43022aaeef5"},
+    {file = "rapidfuzz-3.9.6-pp38-pypy38_pp73-macosx_11_0_arm64.whl", hash = "sha256:88144f5f52ae977df9352029488326afadd7a7f42c6779d486d1f82d43b2b1f2"},
+    {file = "rapidfuzz-3.9.6-pp38-pypy38_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:715aeaabafba2709b9dd91acb2a44bad59d60b4616ef90c08f4d4402a3bbca60"},
+    {file = "rapidfuzz-3.9.6-pp38-pypy38_pp73-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:af26ebd3714224fbf9bebbc27bdbac14f334c15f5d7043699cd694635050d6ca"},
+    {file = "rapidfuzz-3.9.6-pp38-pypy38_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:101bd2df438861a005ed47c032631b7857dfcdb17b82beeeb410307983aac61d"},
+    {file = "rapidfuzz-3.9.6-pp38-pypy38_pp73-win_amd64.whl", hash = "sha256:2185e8e29809b97ad22a7f99281d1669a89bdf5fa1ef4ef1feca36924e675367"},
+    {file = "rapidfuzz-3.9.6-pp39-pypy39_pp73-macosx_10_15_x86_64.whl", hash = "sha256:9e53c72d08f0e9c6e4a369e52df5971f311305b4487690c62e8dd0846770260c"},
+    {file = "rapidfuzz-3.9.6-pp39-pypy39_pp73-macosx_11_0_arm64.whl", hash = "sha256:a0cb157162f0cdd62e538c7bd298ff669847fc43a96422811d5ab933f4c16c3a"},
+    {file = "rapidfuzz-3.9.6-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4bb5ff2bd48132ed5e7fbb8f619885facb2e023759f2519a448b2c18afe07e5d"},
+    {file = "rapidfuzz-3.9.6-pp39-pypy39_pp73-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:6dc37f601865e8407e3a8037ffbc3afe0b0f837b2146f7632bd29d087385babe"},
+    {file = "rapidfuzz-3.9.6-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a657eee4b94668faf1fa2703bdd803654303f7e468eb9ba10a664d867ed9e779"},
+    {file = "rapidfuzz-3.9.6-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:51be6ab5b1d5bb32abd39718f2a5e3835502e026a8272d139ead295c224a6f5e"},
+    {file = "rapidfuzz-3.9.6.tar.gz", hash = "sha256:5cf2a7d621e4515fee84722e93563bf77ff2cbe832a77a48b81f88f9e23b9e8d"},
+]
+
+[package.extras]
+full = ["numpy"]
+
 [[package]]
 name = "redis"
 version = "5.0.8"
@@ -3093,6 +3381,23 @@ files = [
     {file = "rpds_py-0.19.1.tar.gz", hash = "sha256:31dd5794837f00b46f4096aa8ccaa5972f73a938982e32ed817bb520c465e520"},
 ]
 
+[[package]]
+name = "s3transfer"
+version = "0.10.2"
+description = "An Amazon S3 Transfer Manager"
+optional = false
+python-versions = ">=3.8"
+files = [
+    {file = "s3transfer-0.10.2-py3-none-any.whl", hash = "sha256:eca1c20de70a39daee580aef4986996620f365c4e0fda6a86100231d62f1bf69"},
+    {file = "s3transfer-0.10.2.tar.gz", hash = "sha256:0711534e9356d3cc692fdde846b4a1e4b0cb6519971860796e6bc4c7aea00ef6"},
+]
+
+[package.dependencies]
+botocore = ">=1.33.2,<2.0a.0"
+
+[package.extras]
+crt = ["botocore[crt] (>=1.33.2,<2.0a.0)"]
+
 [[package]]
 name = "setuptools"
 version = "73.0.1"
@@ -3641,6 +3946,22 @@ files = [
     {file = "tzdata-2024.1.tar.gz", hash = "sha256:2674120f8d891909751c38abcdfd386ac0a5a1127954fbc332af6b5ceae07efd"},
 ]
 
+[[package]]
+name = "urllib3"
+version = "1.26.19"
+description = "HTTP library with thread-safe connection pooling, file post, and more."
+optional = false
+python-versions = "!=3.0.*,!=3.1.*,!=3.2.*,!=3.3.*,!=3.4.*,!=3.5.*,>=2.7"
+files = [
+    {file = "urllib3-1.26.19-py2.py3-none-any.whl", hash = "sha256:37a0344459b199fce0e80b0d3569837ec6b6937435c5244e7fd73fa6006830f3"},
+    {file = "urllib3-1.26.19.tar.gz", hash = "sha256:3e3d753a8618b86d7de333b4223005f68720bcd6a7d2bcb9fbd2229ec7c1e429"},
+]
+
+[package.extras]
+brotli = ["brotli (==1.0.9)", "brotli (>=1.0.9)", "brotlicffi (>=0.8.0)", "brotlipy (>=0.6.0)"]
+secure = ["certifi", "cryptography (>=1.3.4)", "idna (>=2.0.0)", "ipaddress", "pyOpenSSL (>=0.14)", "urllib3-secure-extra"]
+socks = ["PySocks (>=1.5.6,!=1.5.7,<2.0)"]
+
 [[package]]
 name = "urllib3"
 version = "2.2.2"
@@ -3904,4 +4225,4 @@ typescript = ["tree-sitter-typescript"]
 [metadata]
 lock-version = "2.0"
 python-versions = "^3.9.0"
-content-hash = "b8b1f2e9a8da22e59f874abab8847f1096da0c2782924889778f48722c7f4e44"
+content-hash = "4467d43095e2647060b964f8a8f13b4245d62ebe0f48b947220b2fc0dd115281"
diff --git a/pyproject.toml b/pyproject.toml
index 848c9d62..a1b996a7 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -42,6 +42,8 @@ llama-index-embeddings-litellm = "^0.1.1"
 pre-commit = "^3.8.0"
 qdrant-client = "^1.11.0"
 psycopg2-binary = "^2.9.9"
+boto3 = "^1.35.5"
+python-levenshtein = "^0.25.1"
 
 [tool.poetry.extras]
 python = ["tree-sitter-python"]
@@ -53,3 +55,4 @@ all = ["tree-sitter-python", "tree-sitter-javascript", "tree-sitter-typescript",
 [build-system]
 requires = ["poetry-core"]
 build-backend = "poetry.core.masonry.api"
+
diff --git a/test.txt b/test.txt
new file mode 100644
index 00000000..81b025a4
--- /dev/null
+++ b/test.txt
@@ -0,0 +1,956 @@
+
+Model: gpt-4o
+File: .experiments/code_review/gpt-4o/no_eval/pr_476/issues.json
+Content:
+[
+  {
+    "topic": "Error Handling",
+    "comment": "Generic exception handling without logging specific error details.",
+    "confidence": "critical",
+    "reason": "Using a generic `except Exception` block without logging the specific error details can make debugging difficult.",
+    "solution": "Log the specific error message in the exception block.",
+    "actual_code": "except Exception:\n    print(\"Error\")",
+    "fixed_code": "except Exception as e:\n    print(f\"Error:{e}\")",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 140,
+    "end_line": 141,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "Code Readability",
+    "comment": "Unnecessary print statements left in the code.",
+    "confidence": "important",
+    "reason": "Leaving print statements in production code can clutter the output and is generally not recommended.",
+    "solution": "Remove or replace print statements with proper logging.",
+    "actual_code": "print(\"diff: \", diff_text)\nprint(\"pr_files\", pr_files)",
+    "fixed_code": "",
+    "file_name": "examples/code_review/main.py",
+    "start_line": 21,
+    "end_line": 22,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Function Signature",
+    "comment": "Modified function signature without updating all references.",
+    "confidence": "important",
+    "reason": "Changing a function signature without updating all references can lead to runtime errors.",
+    "solution": "Ensure all references to `post_pull_request` are updated to include the new `tests` parameter.",
+    "actual_code": "def post_pull_request(url, data, installation_id, tests=None):",
+    "fixed_code": "",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 107,
+    "end_line": 107,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 7
+  },
+  {
+    "topic": "Code Maintainability",
+    "comment": "Redundant code for sorting files.",
+    "confidence": "moderate",
+    "reason": "The custom sorting logic can be replaced with Python's built-in sorting functions for better readability and maintainability.",
+    "solution": "Use Python's `sorted` function with a key parameter.",
+    "actual_code": "sorted_files =[]\nfor file in files:\n    min_index = len(sorted_files)\n    file_name = file[\"filename\"]\n    for i, sorted_file in enumerate(sorted_files):\n        if file_name < sorted_file[\"filename\"]:\n            min_index = i\n            break\n    sorted_files.insert(min_index, file)\nreturn sorted_files",
+    "fixed_code": "sorted_files = sorted(files, key=lambda x: x[\"filename\"])\nreturn sorted_files",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 185,
+    "end_line": 194,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 5
+  },
+  {
+    "topic": "Code Quality",
+    "comment": "Unnecessary variable assignment.",
+    "confidence": "low",
+    "reason": "Assigning `issues` in the loop is unnecessary and can be removed.",
+    "solution": "Remove the assignment of `issues` within the loop.",
+    "actual_code": "issues = review",
+    "fixed_code": "",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 153,
+    "end_line": 153,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 3
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
+
+Model: gpt-4o-mini
+File: .experiments/code_review/gpt-4o-mini/no_eval/pr_476/issues.json
+Content:
+[
+  {
+    "topic": "Error Handling",
+    "comment": "Broad exception handling can obscure specific errors.",
+    "confidence": "important",
+    "reason": "Using a generic Exception can make debugging difficult and hide underlying issues.",
+    "solution": "Catch specific exceptions where possible.",
+    "actual_code": "except Exception:",
+    "fixed_code": "except KeyError:",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 140,
+    "end_line": 141,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Function Signature",
+    "comment": "The function 'post_pull_request' has an additional parameter that should be documented.",
+    "confidence": "important",
+    "reason": "New parameters should be documented to ensure clarity for future maintainers.",
+    "solution": "Update the function docstring to include the 'tests' parameter.",
+    "actual_code": "def post_pull_request(url, data, installation_id, tests=None):",
+    "fixed_code": "def post_pull_request(url, data, installation_id, tests=None):  # tests: List of test files",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 106,
+    "end_line": 106,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Code Readability",
+    "comment": "The new function 'sort_files' lacks a docstring.",
+    "confidence": "important",
+    "reason": "Docstrings are essential for understanding the purpose and usage of functions.",
+    "solution": "Add a docstring to describe the function's purpose and parameters.",
+    "actual_code": "def sort_files(files):",
+    "fixed_code": "def sort_files(files):  # Sorts a list of file dictionaries by filename.",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 184,
+    "end_line": 184,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Variable Naming",
+    "comment": "The variable 'tests' could be more descriptive.",
+    "confidence": "moderate",
+    "reason": "Descriptive variable names improve code readability and maintainability.",
+    "solution": "Consider renaming 'tests' to 'generated_tests' for clarity.",
+    "actual_code": "tests = generate_tests(pr_files)",
+    "fixed_code": "generated_tests = generate_tests(pr_files)",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 58,
+    "end_line": 58,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 3
+  },
+  {
+    "topic": "Logging",
+    "comment": "Consider using logging instead of print statements for error reporting.",
+    "confidence": "important",
+    "reason": "Using logging allows for better control over the output and can be configured for different environments.",
+    "solution": "Replace print statements with appropriate logging calls.",
+    "actual_code": "print(\"Error\")",
+    "fixed_code": "logger.error(\"Error occurred\")",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 141,
+    "end_line": 141,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
+
+Model: gpt-4o-try2
+File not found: .experiments/code_review/gpt-4o-try2/no_eval/pr_476/issues.json
+
+Model: haiku
+File: .experiments/code_review/haiku/no_eval/pr_476/issues.json
+Content:
+[
+  {
+    "topic": "Sorting PR Files",
+    "comment": "The PR files are now being sorted before passing them to the description generator. This is a good improvement for maintaining consistent file order in the review.",
+    "confidence": "important",
+    "reason": "Sorting the files ensures a consistent order in the review, making it easier for the reviewer to understand the changes.",
+    "solution": "The `sort_files` function looks good and should effectively sort the files in alphabetical order.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 184,
+    "end_line": 194,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Generating Tests",
+    "comment": "The new `generate_tests` function is a good addition, as it provides a way to generate test cases based on the changed files in the PR.",
+    "confidence": "important",
+    "reason": "Generating tests based on the PR files can help ensure the changes don't break existing functionality.",
+    "solution": "The current implementation of `generate_tests` is simple and effective. It returns a list of file names, which can be used to create test cases.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 199,
+    "end_line": 200,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Printing Diff and PR Files",
+    "comment": "The changes in the `main.py` file to print the `diff_text` and `pr_files` are useful for debugging and understanding the input data.",
+    "confidence": "moderate",
+    "reason": "Printing the diff and PR files can help developers better understand the changes being reviewed.",
+    "solution": "The changes look good and should provide helpful information during the review process.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "examples/code_review/main.py",
+    "start_line": 21,
+    "end_line": 22,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Passing Code Quality to Review Description",
+    "comment": "The change to pass the `code_quality` parameter to the `create_pr_review_text` function is a good improvement, as it allows the review description to include information about the overall code quality.",
+    "confidence": "important",
+    "reason": "Providing information about the code quality in the review description can give the developer a better understanding of the overall state of the codebase.",
+    "solution": "The change looks good and should provide valuable information in the review description.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "examples/code_review/main.py",
+    "start_line": 36,
+    "end_line": 36,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Printing Raw Issues",
+    "comment": "The change to print the `review_data.issues` instead of the `topics` variable is an improvement, as it provides more detailed information about the identified issues.",
+    "confidence": "moderate",
+    "reason": "Printing the raw issues can give the developer a better understanding of the specific problems found during the review.",
+    "solution": "The change looks good and should provide more useful information in the output.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "examples/code_review/main.py",
+    "start_line": 39,
+    "end_line": 39,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Removing Unused Configuration",
+    "comment": "The removal of the `enable_observability_logging` configuration option is a good cleanup, as it removes an unused feature from the configuration file.",
+    "confidence": "moderate",
+    "reason": "Removing unused configuration options helps keep the codebase clean and maintainable.",
+    "solution": "The change looks good and should help simplify the configuration file.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "config.json",
+    "start_line": 4,
+    "end_line": 4,
+    "side": "LEFT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
+
+Model: llama-405b
+File: .experiments/code_review/llama-405b/no_eval/pr_476/issues.json
+Content:
+[
+  {
+    "topic": "Error Handling",
+    "comment": "Broad exception handling can mask bugs and make debugging difficult.",
+    "confidence": "important",
+    "reason": "The `except Exception` block in `github_app/github_helper/pull_requests.py` (line 140) catches all exceptions, which can make it challenging to identify and fix specific issues.",
+    "solution": "Catch specific exceptions that can occur during the execution of the code, and provide meaningful error messages to aid in debugging.",
+    "actual_code": "except Exception:",
+    "fixed_code": "except requests.exceptions.RequestException as e:",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 140,
+    "end_line": 141,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Code Organization",
+    "comment": "The `sort_files` function is not necessary and can be replaced with a built-in sorting function.",
+    "confidence": "moderate",
+    "reason": "The `sort_files` function in `github_app/github_helper/pull_requests.py` (line 184) is not necessary and can be replaced with the built-in `sorted` function.",
+    "solution": "Use the built-in `sorted` function to sort the files, which is more efficient and Pythonic.",
+    "actual_code": "def sort_files(files):",
+    "fixed_code": "sorted_files = sorted(files, key=lambda x: x['filename'])",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 184,
+    "end_line": 194,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Code Quality",
+    "comment": "The `generate_tests` function is not necessary and can be replaced with a list comprehension.",
+    "confidence": "moderate",
+    "reason": "The `generate_tests` function in `github_app/github_helper/pull_requests.py` (line 199) is not necessary and can be replaced with a list comprehension.",
+    "solution": "Use a list comprehension to generate the tests, which is more efficient and Pythonic.",
+    "actual_code": "def generate_tests(pr_files):",
+    "fixed_code": "tests =[f['filename'] for f in pr_files]",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 199,
+    "end_line": 200,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
+
+Model: sonnet-3.5
+File: .experiments/code_review/sonnet-3.5/no_eval/pr_476/issues.json
+Content:
+[
+  {
+    "topic": "Error Handling",
+    "comment": "Exception handling is too broad and prints a generic error message.",
+    "confidence": "important",
+    "reason": "Catching all exceptions and printing a generic error message can hide important errors and make debugging difficult.",
+    "solution": "Catch specific exceptions and provide more informative error messages.",
+    "actual_code": "except Exception:\n    print(\"Error\")",
+    "fixed_code": "except KeyError as e:\n    print(f\"Invalid confidence level:{e}\")\nexcept Exception as e:\n    print(f\"Unexpected error:{e}\")",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 140,
+    "end_line": 141,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Code Efficiency",
+    "comment": "The sort_files function implements a manual insertion sort, which is inefficient for large lists.",
+    "confidence": "important",
+    "reason": "Insertion sort has O(n^2) time complexity, which can be slow for large numbers of files.",
+    "solution": "Use Python's built-in sorted() function with a key function for better performance.",
+    "actual_code": "def sort_files(files):\n    sorted_files =[]\n    for file in files:\n        min_index = len(sorted_files)\n        file_name = file[\"filename\"]\n        for i, sorted_file in enumerate(sorted_files):\n            if file_name < sorted_file[\"filename\"]:\n                min_index = i\n                break\n        sorted_files.insert(min_index, file)\n    return sorted_files",
+    "fixed_code": "def sort_files(files):\n    return sorted(files, key=lambda x: x[\"filename\"])",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 184,
+    "end_line": 194,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Code Simplification",
+    "comment": "The generate_tests function can be simplified using a list comprehension.",
+    "confidence": "moderate",
+    "reason": "The current implementation is unnecessarily verbose for a simple operation.",
+    "solution": "Use a list comprehension to create the list of filenames.",
+    "actual_code": "def generate_tests(pr_files):\n    return[f[\"filename\"] for f in pr_files]",
+    "fixed_code": "def generate_tests(pr_files):\n    return[f[\"filename\"] for f in pr_files]",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 199,
+    "end_line": 200,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 2
+  },
+  {
+    "topic": "Code Consistency",
+    "comment": "Inconsistent use of print statements for debugging.",
+    "confidence": "low",
+    "reason": "Some print statements are commented out while others are added, which may lead to inconsistent debugging output.",
+    "solution": "Decide on a consistent approach for debug logging, preferably using a proper logging system.",
+    "actual_code": "print(\"diff: \", diff_text)\nprint(\"pr_files\", pr_files)",
+    "fixed_code": "import logging\n\nlogging.debug(f\"diff:{diff_text}\")\nlogging.debug(f\"pr_files:{pr_files}\")",
+    "file_name": "examples/code_review/main.py",
+    "start_line": 21,
+    "end_line": 22,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 3
+  },
+  {
+    "topic": "Code Improvement",
+    "comment": "The create_pr_review_text function now includes a code_quality parameter, which is a good improvement.",
+    "confidence": "moderate",
+    "reason": "Including code quality in the review text provides more comprehensive feedback.",
+    "solution": "No change needed, this is a positive improvement.",
+    "actual_code": "review_desc = create_pr_review_text(topics, code_quality=review_data.code_quality)",
+    "fixed_code": "",
+    "file_name": "examples/code_review/main.py",
+    "start_line": 36,
+    "end_line": 36,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 1
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Removal of 'enable_observability_logging' from config.json",
+    "confidence": "moderate",
+    "reason": "Removing configuration options without proper documentation or migration path can lead to issues for existing users.",
+    "solution": "If the feature is no longer supported, provide a migration guide or deprecation notice.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "config.json",
+    "start_line": 4,
+    "end_line": 4,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
+
+Model: gpt-4o
+File: .experiments/code_review/gpt-4o/no_eval/pr_476/issues.json
+Content:
+[
+  {
+    "topic": "Error Handling",
+    "comment": "Generic exception handling without logging specific error details.",
+    "confidence": "critical",
+    "reason": "Using a generic `except Exception` block without logging the specific error details can make debugging difficult.",
+    "solution": "Log the specific error message in the exception block.",
+    "actual_code": "except Exception:\n    print(\"Error\")",
+    "fixed_code": "except Exception as e:\n    print(f\"Error:{e}\")",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 140,
+    "end_line": 141,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 9
+  },
+  {
+    "topic": "Code Readability",
+    "comment": "Unnecessary print statements left in the code.",
+    "confidence": "important",
+    "reason": "Leaving print statements in production code can clutter the output and is generally not recommended.",
+    "solution": "Remove or replace print statements with proper logging.",
+    "actual_code": "print(\"diff: \", diff_text)\nprint(\"pr_files\", pr_files)",
+    "fixed_code": "",
+    "file_name": "examples/code_review/main.py",
+    "start_line": 21,
+    "end_line": 22,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Function Signature",
+    "comment": "Modified function signature without updating all references.",
+    "confidence": "important",
+    "reason": "Changing a function signature without updating all references can lead to runtime errors.",
+    "solution": "Ensure all references to `post_pull_request` are updated to include the new `tests` parameter.",
+    "actual_code": "def post_pull_request(url, data, installation_id, tests=None):",
+    "fixed_code": "",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 107,
+    "end_line": 107,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 7
+  },
+  {
+    "topic": "Code Maintainability",
+    "comment": "Redundant code for sorting files.",
+    "confidence": "moderate",
+    "reason": "The custom sorting logic can be replaced with Python's built-in sorting functions for better readability and maintainability.",
+    "solution": "Use Python's `sorted` function with a key parameter.",
+    "actual_code": "sorted_files =[]\nfor file in files:\n    min_index = len(sorted_files)\n    file_name = file[\"filename\"]\n    for i, sorted_file in enumerate(sorted_files):\n        if file_name < sorted_file[\"filename\"]:\n            min_index = i\n            break\n    sorted_files.insert(min_index, file)\nreturn sorted_files",
+    "fixed_code": "sorted_files = sorted(files, key=lambda x: x[\"filename\"])\nreturn sorted_files",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 185,
+    "end_line": 194,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 5
+  },
+  {
+    "topic": "Code Quality",
+    "comment": "Unnecessary variable assignment.",
+    "confidence": "low",
+    "reason": "Assigning `issues` in the loop is unnecessary and can be removed.",
+    "solution": "Remove the assignment of `issues` within the loop.",
+    "actual_code": "issues = review",
+    "fixed_code": "",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 153,
+    "end_line": 153,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 3
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
+
+Model: gpt-4o-mini
+File: .experiments/code_review/gpt-4o-mini/no_eval/pr_476/issues.json
+Content:
+[
+  {
+    "topic": "Error Handling",
+    "comment": "Broad exception handling can obscure specific errors.",
+    "confidence": "important",
+    "reason": "Using a generic Exception can make debugging difficult and hide underlying issues.",
+    "solution": "Catch specific exceptions where possible.",
+    "actual_code": "except Exception:",
+    "fixed_code": "except KeyError:",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 140,
+    "end_line": 141,
+    "side": "LEFT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Function Signature",
+    "comment": "The function 'post_pull_request' has an additional parameter that should be documented.",
+    "confidence": "important",
+    "reason": "New parameters should be documented to ensure clarity for future maintainers.",
+    "solution": "Update the function docstring to include the 'tests' parameter.",
+    "actual_code": "def post_pull_request(url, data, installation_id, tests=None):",
+    "fixed_code": "def post_pull_request(url, data, installation_id, tests=None):  # tests: List of test files",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 106,
+    "end_line": 106,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 5
+  },
+  {
+    "topic": "Code Readability",
+    "comment": "The new function 'sort_files' lacks a docstring.",
+    "confidence": "important",
+    "reason": "Docstrings are essential for understanding the purpose and usage of functions.",
+    "solution": "Add a docstring to describe the function's purpose and parameters.",
+    "actual_code": "def sort_files(files):",
+    "fixed_code": "def sort_files(files):  # Sorts a list of file dictionaries by filename.",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 184,
+    "end_line": 184,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Variable Naming",
+    "comment": "The variable 'tests' could be more descriptive.",
+    "confidence": "moderate",
+    "reason": "Descriptive variable names improve code readability and maintainability.",
+    "solution": "Consider renaming 'tests' to 'generated_tests' for clarity.",
+    "actual_code": "tests = generate_tests(pr_files)",
+    "fixed_code": "generated_tests = generate_tests(pr_files)",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 58,
+    "end_line": 58,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 3
+  },
+  {
+    "topic": "Logging",
+    "comment": "Consider using logging instead of print statements for error reporting.",
+    "confidence": "important",
+    "reason": "Using logging allows for better control over the output and can be configured for different environments.",
+    "solution": "Replace print statements with appropriate logging calls.",
+    "actual_code": "print(\"Error\")",
+    "fixed_code": "logger.error(\"Error occurred\")",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 141,
+    "end_line": 141,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
+
+Model: gpt-4o-try2
+File not found: .experiments/code_review/gpt-4o-try2/no_eval/pr_476/issues.json
+
+Model: haiku
+File: .experiments/code_review/haiku/no_eval/pr_476/issues.json
+Content:
+[
+  {
+    "topic": "Sorting PR Files",
+    "comment": "The PR files are now being sorted before passing them to the description generator. This is a good improvement for maintaining consistent file order in the review.",
+    "confidence": "important",
+    "reason": "Sorting the files ensures a consistent order in the review, making it easier for the reviewer to understand the changes.",
+    "solution": "The `sort_files` function looks good and should effectively sort the files in alphabetical order.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 184,
+    "end_line": 194,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Generating Tests",
+    "comment": "The new `generate_tests` function is a good addition, as it provides a way to generate test cases based on the changed files in the PR.",
+    "confidence": "important",
+    "reason": "Generating tests based on the PR files can help ensure the changes don't break existing functionality.",
+    "solution": "The current implementation of `generate_tests` is simple and effective. It returns a list of file names, which can be used to create test cases.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 199,
+    "end_line": 200,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Printing Diff and PR Files",
+    "comment": "The changes in the `main.py` file to print the `diff_text` and `pr_files` are useful for debugging and understanding the input data.",
+    "confidence": "moderate",
+    "reason": "Printing the diff and PR files can help developers better understand the changes being reviewed.",
+    "solution": "The changes look good and should provide helpful information during the review process.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "examples/code_review/main.py",
+    "start_line": 21,
+    "end_line": 22,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Passing Code Quality to Review Description",
+    "comment": "The change to pass the `code_quality` parameter to the `create_pr_review_text` function is a good improvement, as it allows the review description to include information about the overall code quality.",
+    "confidence": "important",
+    "reason": "Providing information about the code quality in the review description can give the developer a better understanding of the overall state of the codebase.",
+    "solution": "The change looks good and should provide valuable information in the review description.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "examples/code_review/main.py",
+    "start_line": 36,
+    "end_line": 36,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 4
+  },
+  {
+    "topic": "Printing Raw Issues",
+    "comment": "The change to print the `review_data.issues` instead of the `topics` variable is an improvement, as it provides more detailed information about the identified issues.",
+    "confidence": "moderate",
+    "reason": "Printing the raw issues can give the developer a better understanding of the specific problems found during the review.",
+    "solution": "The change looks good and should provide more useful information in the output.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "examples/code_review/main.py",
+    "start_line": 39,
+    "end_line": 39,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Removing Unused Configuration",
+    "comment": "The removal of the `enable_observability_logging` configuration option is a good cleanup, as it removes an unused feature from the configuration file.",
+    "confidence": "moderate",
+    "reason": "Removing unused configuration options helps keep the codebase clean and maintainable.",
+    "solution": "The change looks good and should help simplify the configuration file.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "config.json",
+    "start_line": 4,
+    "end_line": 4,
+    "side": "LEFT",
+    "sentiment": "positive",
+    "severity_level": 3
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
+
+Model: llama-405b
+File: .experiments/code_review/llama-405b/no_eval/pr_476/issues.json
+Content:
+[
+  {
+    "topic": "Error Handling",
+    "comment": "Broad exception handling can mask bugs and make debugging difficult.",
+    "confidence": "important",
+    "reason": "The `except Exception` block in `github_app/github_helper/pull_requests.py` (line 140) catches all exceptions, which can make it challenging to identify and fix specific issues.",
+    "solution": "Catch specific exceptions that can occur during the execution of the code, and provide meaningful error messages to aid in debugging.",
+    "actual_code": "except Exception:",
+    "fixed_code": "except requests.exceptions.RequestException as e:",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 140,
+    "end_line": 141,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Code Organization",
+    "comment": "The `sort_files` function is not necessary and can be replaced with a built-in sorting function.",
+    "confidence": "moderate",
+    "reason": "The `sort_files` function in `github_app/github_helper/pull_requests.py` (line 184) is not necessary and can be replaced with the built-in `sorted` function.",
+    "solution": "Use the built-in `sorted` function to sort the files, which is more efficient and Pythonic.",
+    "actual_code": "def sort_files(files):",
+    "fixed_code": "sorted_files = sorted(files, key=lambda x: x['filename'])",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 184,
+    "end_line": 194,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Code Quality",
+    "comment": "The `generate_tests` function is not necessary and can be replaced with a list comprehension.",
+    "confidence": "moderate",
+    "reason": "The `generate_tests` function in `github_app/github_helper/pull_requests.py` (line 199) is not necessary and can be replaced with a list comprehension.",
+    "solution": "Use a list comprehension to generate the tests, which is more efficient and Pythonic.",
+    "actual_code": "def generate_tests(pr_files):",
+    "fixed_code": "tests =[f['filename'] for f in pr_files]",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 199,
+    "end_line": 200,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]
+
+Model: sonnet-3.5
+File: .experiments/code_review/sonnet-3.5/no_eval/pr_476/issues.json
+Content:
+[
+  {
+    "topic": "Error Handling",
+    "comment": "Exception handling is too broad and prints a generic error message.",
+    "confidence": "important",
+    "reason": "Catching all exceptions and printing a generic error message can hide important errors and make debugging difficult.",
+    "solution": "Catch specific exceptions and provide more informative error messages.",
+    "actual_code": "except Exception:\n    print(\"Error\")",
+    "fixed_code": "except KeyError as e:\n    print(f\"Invalid confidence level:{e}\")\nexcept Exception as e:\n    print(f\"Unexpected error:{e}\")",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 140,
+    "end_line": 141,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 7
+  },
+  {
+    "topic": "Code Efficiency",
+    "comment": "The sort_files function implements a manual insertion sort, which is inefficient for large lists.",
+    "confidence": "important",
+    "reason": "Insertion sort has O(n^2) time complexity, which can be slow for large numbers of files.",
+    "solution": "Use Python's built-in sorted() function with a key function for better performance.",
+    "actual_code": "def sort_files(files):\n    sorted_files =[]\n    for file in files:\n        min_index = len(sorted_files)\n        file_name = file[\"filename\"]\n        for i, sorted_file in enumerate(sorted_files):\n            if file_name < sorted_file[\"filename\"]:\n                min_index = i\n                break\n        sorted_files.insert(min_index, file)\n    return sorted_files",
+    "fixed_code": "def sort_files(files):\n    return sorted(files, key=lambda x: x[\"filename\"])",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 184,
+    "end_line": 194,
+    "side": "RIGHT",
+    "sentiment": "negative",
+    "severity_level": 6
+  },
+  {
+    "topic": "Code Simplification",
+    "comment": "The generate_tests function can be simplified using a list comprehension.",
+    "confidence": "moderate",
+    "reason": "The current implementation is unnecessarily verbose for a simple operation.",
+    "solution": "Use a list comprehension to create the list of filenames.",
+    "actual_code": "def generate_tests(pr_files):\n    return[f[\"filename\"] for f in pr_files]",
+    "fixed_code": "def generate_tests(pr_files):\n    return[f[\"filename\"] for f in pr_files]",
+    "file_name": "github_app/github_helper/pull_requests.py",
+    "start_line": 199,
+    "end_line": 200,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 2
+  },
+  {
+    "topic": "Code Consistency",
+    "comment": "Inconsistent use of print statements for debugging.",
+    "confidence": "low",
+    "reason": "Some print statements are commented out while others are added, which may lead to inconsistent debugging output.",
+    "solution": "Decide on a consistent approach for debug logging, preferably using a proper logging system.",
+    "actual_code": "print(\"diff: \", diff_text)\nprint(\"pr_files\", pr_files)",
+    "fixed_code": "import logging\n\nlogging.debug(f\"diff:{diff_text}\")\nlogging.debug(f\"pr_files:{pr_files}\")",
+    "file_name": "examples/code_review/main.py",
+    "start_line": 21,
+    "end_line": 22,
+    "side": "RIGHT",
+    "sentiment": "neutral",
+    "severity_level": 3
+  },
+  {
+    "topic": "Code Improvement",
+    "comment": "The create_pr_review_text function now includes a code_quality parameter, which is a good improvement.",
+    "confidence": "moderate",
+    "reason": "Including code quality in the review text provides more comprehensive feedback.",
+    "solution": "No change needed, this is a positive improvement.",
+    "actual_code": "review_desc = create_pr_review_text(topics, code_quality=review_data.code_quality)",
+    "fixed_code": "",
+    "file_name": "examples/code_review/main.py",
+    "start_line": 36,
+    "end_line": 36,
+    "side": "RIGHT",
+    "sentiment": "positive",
+    "severity_level": 1
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Removal of 'enable_observability_logging' from config.json",
+    "confidence": "moderate",
+    "reason": "Removing configuration options without proper documentation or migration path can lead to issues for existing users.",
+    "solution": "If the feature is no longer supported, provide a migration guide or deprecation notice.",
+    "actual_code": "",
+    "fixed_code": "",
+    "file_name": "config.json",
+    "start_line": 4,
+    "end_line": 4,
+    "side": "LEFT",
+    "sentiment": "neutral",
+    "severity_level": 4
+  },
+  {
+    "topic": "Configuration",
+    "comment": "Changes made to sensitive file",
+    "confidence": "critical",
+    "reason": "Changes were made to config.json, which needs review",
+    "solution": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_name": "config.json",
+    "sentiment": "negative",
+    "severity_level": 10
+  }
+]

From d29028dac50ece56e8689aa7d018c392e9693466 Mon Sep 17 00:00:00 2001
From: "dependabot[bot]" <49699333+dependabot[bot]@users.noreply.github.com>
Date: Tue, 27 Aug 2024 21:22:01 +0000
Subject: [PATCH 16/19] build(deps): bump nltk from 3.9b1 to 3.9

Bumps [nltk](https://github.com/nltk/nltk) from 3.9b1 to 3.9.
- [Changelog](https://github.com/nltk/nltk/blob/develop/ChangeLog)
- [Commits](https://github.com/nltk/nltk/commits/3.9)

---
updated-dependencies:
- dependency-name: nltk
  dependency-type: indirect
...

Signed-off-by: dependabot[bot] <support@github.com>
---
 poetry.lock | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/poetry.lock b/poetry.lock
index 9b2c9127..c50ef29d 100644
--- a/poetry.lock
+++ b/poetry.lock
@@ -1,4 +1,4 @@
-# This file is automatically @generated by Poetry 1.8.2 and should not be changed by hand.
+# This file is automatically @generated by Poetry 1.8.3 and should not be changed by hand.
 
 [[package]]
 name = "aiohappyeyeballs"
@@ -1921,13 +1921,13 @@ test = ["pytest (>=7.2)", "pytest-cov (>=4.0)"]
 
 [[package]]
 name = "nltk"
-version = "3.9b1"
+version = "3.9"
 description = "Natural Language Toolkit"
 optional = false
 python-versions = ">=3.8"
 files = [
-    {file = "nltk-3.9b1-py3-none-any.whl", hash = "sha256:2d87d63a93824e6a01fcd42053b26e18e53bf8c22aa9c2c9b19e32c036739ed5"},
-    {file = "nltk-3.9b1.tar.gz", hash = "sha256:52f95a2c3f947a34f78ccff081f31e2c11b6a2037ad66a7ba2093b5d15b4622f"},
+    {file = "nltk-3.9-py3-none-any.whl", hash = "sha256:d17863e861bb33ac617893329d71d06a3dfb7e3eb9ee0b8105281c53944a45a1"},
+    {file = "nltk-3.9.tar.gz", hash = "sha256:e98acac454407fa38b76cccb29208d377731cf7fab68f323754a3681f104531f"},
 ]
 
 [package.dependencies]

From eea4145ee34cd23fc5167286dd93e6ca1158efbe Mon Sep 17 00:00:00 2001
From: Saurav Panda <sgp65@cornell.edu>
Date: Tue, 27 Aug 2024 17:14:27 -0700
Subject: [PATCH 17/19] feat: update to code review prompt

---
 .../no_eval/pr_222/comments.json              | 102 ++++
 .../no_eval/pr_222/issues.json                | 372 ++++++++++++
 .../no_eval/pr_222/review.md                  | 166 ++++++
 .../no_eval/pr_232}/comments.json             |   0
 .../no_eval/pr_232/issues.json                | 227 +++++++
 .../no_eval/pr_232/review.md                  |  41 ++
 .../no_eval/pr_252}/comments.json             |   0
 .../no_eval/pr_252/issues.json                |  77 +++
 .../no_eval/pr_252/review.md                  |  41 ++
 .../no_eval/pr_335}/comments.json             |   0
 .../no_eval/pr_335/issues.json                |  62 ++
 .../no_eval/pr_335/review.md                  |  41 ++
 .../no_eval/pr_400/comments.json              |  16 +
 .../no_eval/pr_400/issues.json                | 421 +++++++++++++
 .../no_eval/pr_400/review.md                  |  64 ++
 .../no_eval/pr_440}/comments.json             |   0
 .../no_eval/pr_440/issues.json                |  47 ++
 .../no_eval/pr_440/review.md                  |  41 ++
 .../no_eval/pr_476/comments.json              |  16 +
 .../no_eval/pr_476/issues.json                | 121 ++++
 .../no_eval/pr_476/review.md                  |  64 ++
 .../no_eval/pr_5/comments.json                |  17 +
 .../no_eval/pr_5/issues.json                  | 107 ++++
 .../no_eval/pr_5/review.md                    |  64 ++
 .../no_eval/pr_222/comments.json              |  87 +++
 .../no_eval/pr_222/issues.json                | 342 +++++++++++
 .../no_eval/pr_222/review.md                  | 136 +++++
 .../no_eval/pr_232}/comments.json             |   0
 .../no_eval/pr_232/issues.json                | 197 ++++++
 .../no_eval/pr_232/review.md                  |  33 ++
 .../no_eval/pr_252}/comments.json             |   0
 .../no_eval/pr_252/issues.json                |  62 ++
 .../no_eval/pr_252/review.md                  |  33 ++
 .../no_eval/pr_335}/comments.json             |   0
 .../no_eval/pr_335/issues.json                |  62 ++
 .../no_eval/pr_335/review.md                  |  33 ++
 .../no_eval/pr_400/comments.json              |  16 +
 .../no_eval/pr_400/issues.json                | 346 +++++++++++
 .../no_eval/pr_400/review.md                  |  56 ++
 .../no_eval/pr_440}/comments.json             |   0
 .../no_eval/pr_440/issues.json                |  47 ++
 .../no_eval/pr_440/review.md                  |  33 ++
 .../no_eval/pr_476/comments.json              |  31 +
 .../no_eval/pr_476/issues.json                |  91 +++
 .../no_eval/pr_476/review.md                  |  72 +++
 .../no_eval/pr_5/comments.json                |  17 +
 .../no_eval/pr_5/issues.json                  | 107 ++++
 .../no_eval/pr_5}/review.md                   |  44 +-
 .../no_eval/pr_222/comments.json              | 100 ++++
 .../no_eval/pr_222/issues.json                | 282 +++++++++
 .../no_eval/pr_222/review.md                  | 152 +++++
 .../no_eval/pr_232}/comments.json             |   0
 .../no_eval/pr_232/issues.json                | 198 +++++++
 .../no_eval/pr_232/review.md                  |  33 ++
 .../no_eval/pr_252}/comments.json             |   0
 .../no_eval/pr_252/issues.json                |  58 ++
 .../no_eval/pr_252/review.md                  |  33 ++
 .../no_eval/pr_335}/comments.json             |   0
 .../no_eval/pr_335/issues.json                |  72 +++
 .../no_eval/pr_335/review.md                  |  33 ++
 .../no_eval/pr_400/comments.json              |  16 +
 .../no_eval/pr_400/issues.json                | 380 ++++++++++++
 .../no_eval/pr_400/review.md                  |  56 ++
 .../no_eval/pr_440}/comments.json             |   0
 .../no_eval/pr_440/issues.json                |  44 ++
 .../no_eval/pr_440/review.md                  |  33 ++
 .../no_eval/pr_476/comments.json              |  30 +
 .../no_eval/pr_476/issues.json                |  86 +++
 .../no_eval/pr_476/review.md                  |  72 +++
 .../no_eval/pr_5/comments.json                |   1 +
 .../no_eval/pr_5/issues.json                  |  86 +++
 .../no_eval/pr_5/review.md                    |  41 ++
 .../no_eval/pr_222/comments.json              |  86 +++
 .../no_eval/pr_222/issues.json                | 282 +++++++++
 .../no_eval/pr_222/review.md                  | 136 +++++
 .../no_eval/pr_232/comments.json              |   1 +
 .../no_eval/pr_232/issues.json                | 156 +++++
 .../no_eval/pr_232/review.md                  |  33 ++
 .../no_eval/pr_252/comments.json              |   1 +
 .../no_eval/pr_252/issues.json                |  58 ++
 .../no_eval/pr_252/review.md                  |  33 ++
 .../no_eval/pr_335/comments.json              |   1 +
 .../no_eval/pr_335/issues.json                |  58 ++
 .../no_eval/pr_335/review.md                  |  33 ++
 .../no_eval/pr_400/comments.json              |  30 +
 .../no_eval/pr_400/issues.json                | 338 +++++++++++
 .../no_eval/pr_400/review.md                  |  81 +++
 .../no_eval/pr_440/comments.json              |   1 +
 .../no_eval/pr_440/issues.json                |  44 ++
 .../no_eval/pr_440/review.md                  |  33 ++
 .../no_eval/pr_476/comments.json              |  16 +
 .../no_eval/pr_476/issues.json                |  72 +++
 .../no_eval/pr_476/review.md                  |  64 ++
 .../no_eval/pr_5/comments.json                |  16 +
 .../no_eval/pr_5/issues.json                  |  72 +++
 .../no_eval/pr_5/review.md                    |  64 ++
 .../code_review/dataset/pr_222/issues.json    | 160 ++---
 .../code_review/dataset/pr_232/issues.json    | 112 ++--
 .../code_review/dataset/pr_252/issues.json    |  96 +--
 .../code_review/dataset/pr_335/issues.json    |  96 +--
 .../code_review/dataset/pr_400/issues.json    |  70 +--
 .../code_review/dataset/pr_476/issues.json    |  78 +--
 .../code_review/dataset/pr_5/issues.json      | 112 ++--
 .experiments/code_review/evaluate.py          |  75 ++-
 .../gpt-4o-mini/no_eval/pr_222/comments.json  | 102 ----
 .../gpt-4o-mini/no_eval/pr_222/issues.json    | 342 -----------
 .../gpt-4o-mini/no_eval/pr_222/review.md      | 402 -------------
 .../gpt-4o-mini/no_eval/pr_335/comments.json  |  17 -
 .../gpt-4o-mini/no_eval/pr_335/issues.json    |  77 ---
 .../gpt-4o-mini/no_eval/pr_335/review.md      | 139 -----
 .../gpt-4o-mini/no_eval/pr_440/issues.json    |  47 --
 .../gpt-4o-mini/no_eval/pr_440/review.md      | 100 ----
 .../gpt-4o-mini/no_eval/pr_476/comments.json  |  16 -
 .../gpt-4o-mini/no_eval/pr_476/issues.json    |  91 ---
 .../gpt-4o-mini/no_eval/pr_476/review.md      | 145 -----
 .../gpt-4o-mini/no_eval/pr_5/comments.json    |  47 --
 .../gpt-4o-mini/no_eval/pr_5/issues.json      | 107 ----
 .../gpt-4o-mini/no_eval/pr_5/review.md        | 155 -----
 .../gpt-4o/no_eval/pr_222/comments.json       | 132 -----
 .../gpt-4o/no_eval/pr_222/issues.json         | 387 ------------
 .../gpt-4o/no_eval/pr_222/review.md           | 560 ------------------
 .../gpt-4o/no_eval/pr_335/comments.json       |  32 -
 .../gpt-4o/no_eval/pr_335/issues.json         |  92 ---
 .../gpt-4o/no_eval/pr_335/review.md           | 151 -----
 .../gpt-4o/no_eval/pr_440/issues.json         |  47 --
 .../gpt-4o/no_eval/pr_440/review.md           |  92 ---
 .../gpt-4o/no_eval/pr_476/comments.json       |  31 -
 .../gpt-4o/no_eval/pr_476/issues.json         |  91 ---
 .../gpt-4o/no_eval/pr_476/review.md           | 144 -----
 .../gpt-4o/no_eval/pr_5/comments.json         |  47 --
 .../gpt-4o/no_eval/pr_5/issues.json           | 107 ----
 .../code_review/gpt-4o/no_eval/pr_5/review.md | 182 ------
 .../haiku/no_eval/pr_222/comments.json        |  72 ---
 .../haiku/no_eval/pr_222/issues.json          | 312 ----------
 .../haiku/no_eval/pr_222/review.md            | 247 --------
 .../haiku/no_eval/pr_252/issues.json          |  32 -
 .../haiku/no_eval/pr_252/review.md            |  70 ---
 .../haiku/no_eval/pr_335/issues.json          |  62 --
 .../haiku/no_eval/pr_335/review.md            |  78 ---
 .../haiku/no_eval/pr_400/comments.json        |  16 -
 .../haiku/no_eval/pr_400/issues.json          | 406 -------------
 .../haiku/no_eval/pr_400/review.md            | 177 ------
 .../haiku/no_eval/pr_440/issues.json          |  62 --
 .../haiku/no_eval/pr_440/review.md            | 118 ----
 .../haiku/no_eval/pr_476/comments.json        |  16 -
 .../haiku/no_eval/pr_476/issues.json          | 106 ----
 .../haiku/no_eval/pr_476/review.md            | 103 ----
 .../haiku/no_eval/pr_5/comments.json          |  32 -
 .../haiku/no_eval/pr_5/issues.json            | 107 ----
 .../code_review/haiku/no_eval/pr_5/review.md  | 211 -------
 .../llama-405b/no_eval/pr_222/comments.json   | 117 ----
 .../llama-405b/no_eval/pr_222/issues.json     | 282 ---------
 .../llama-405b/no_eval/pr_222/review.md       | 195 ------
 .../llama-405b/no_eval/pr_232/issues.json     | 137 -----
 .../llama-405b/no_eval/pr_232/review.md       | 100 ----
 .../llama-405b/no_eval/pr_252/issues.json     |  62 --
 .../llama-405b/no_eval/pr_252/review.md       |  78 ---
 .../llama-405b/no_eval/pr_335/issues.json     |  47 --
 .../llama-405b/no_eval/pr_400/comments.json   |  16 -
 .../llama-405b/no_eval/pr_400/issues.json     | 316 ----------
 .../llama-405b/no_eval/pr_400/review.md       | 197 ------
 .../llama-405b/no_eval/pr_440/issues.json     |  47 --
 .../llama-405b/no_eval/pr_440/review.md       |  92 ---
 .../llama-405b/no_eval/pr_476/comments.json   |  16 -
 .../llama-405b/no_eval/pr_476/issues.json     |  61 --
 .../llama-405b/no_eval/pr_476/review.md       | 115 ----
 .../llama-405b/no_eval/pr_5/comments.json     |  47 --
 .../llama-405b/no_eval/pr_5/issues.json       | 107 ----
 .../llama-405b/no_eval/pr_5/review.md         | 193 ------
 .experiments/code_review/main.py              |   2 +-
 .../sonnet-3.5/no_eval/pr_222/comments.json   |  72 ---
 .../sonnet-3.5/no_eval/pr_222/issues.json     | 312 ----------
 .../sonnet-3.5/no_eval/pr_222/review.md       | 481 ---------------
 .../sonnet-3.5/no_eval/pr_232/issues.json     | 227 -------
 .../sonnet-3.5/no_eval/pr_232/review.md       | 253 --------
 .../sonnet-3.5/no_eval/pr_252/issues.json     |  62 --
 .../sonnet-3.5/no_eval/pr_252/review.md       | 126 ----
 .../sonnet-3.5/no_eval/pr_335/issues.json     |  62 --
 .../sonnet-3.5/no_eval/pr_335/review.md       |  89 ---
 .../sonnet-3.5/no_eval/pr_400/comments.json   |  31 -
 .../sonnet-3.5/no_eval/pr_400/issues.json     | 406 -------------
 .../sonnet-3.5/no_eval/pr_400/review.md       | 435 --------------
 .../sonnet-3.5/no_eval/pr_476/comments.json   |  16 -
 .../sonnet-3.5/no_eval/pr_476/issues.json     | 106 ----
 .../sonnet-3.5/no_eval/pr_476/review.md       | 154 -----
 .../sonnet-3.5/no_eval/pr_5/comments.json     |  47 --
 .../sonnet-3.5/no_eval/pr_5/issues.json       | 107 ----
 .../sonnet-3.5/no_eval/pr_5/review.md         | 234 --------
 .gitignore                                    |   4 +-
 github_app/github_helper/pull_requests.py     |   4 +-
 kaizen/formatters/code_review_formatter.py    |  24 +-
 kaizen/llms/prompts/code_review_prompts.py    | 229 ++++---
 kaizen/reviewer/code_review.py                |  26 +-
 poetry.lock                                   | 100 +++-
 pyproject.toml                                |   5 +-
 195 files changed, 8093 insertions(+), 12121 deletions(-)
 create mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/comments.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/review.md
 rename .experiments/code_review/{gpt-4o-mini/no_eval/pr_440 => code_reviews_20240827_161838/no_eval/pr_232}/comments.json (100%)
 create mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/review.md
 rename .experiments/code_review/{gpt-4o/no_eval/pr_440 => code_reviews_20240827_161838/no_eval/pr_252}/comments.json (100%)
 create mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/review.md
 rename .experiments/code_review/{haiku/no_eval/pr_252 => code_reviews_20240827_161838/no_eval/pr_335}/comments.json (100%)
 create mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/review.md
 create mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/comments.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/review.md
 rename .experiments/code_review/{haiku/no_eval/pr_335 => code_reviews_20240827_161838/no_eval/pr_440}/comments.json (100%)
 create mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/review.md
 create mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/comments.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/review.md
 create mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/comments.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/review.md
 create mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/comments.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/review.md
 rename .experiments/code_review/{haiku/no_eval/pr_440 => code_reviews_20240827_163259/no_eval/pr_232}/comments.json (100%)
 create mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/review.md
 rename .experiments/code_review/{llama-405b/no_eval/pr_232 => code_reviews_20240827_163259/no_eval/pr_252}/comments.json (100%)
 create mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/review.md
 rename .experiments/code_review/{llama-405b/no_eval/pr_252 => code_reviews_20240827_163259/no_eval/pr_335}/comments.json (100%)
 create mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/review.md
 create mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/comments.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/review.md
 rename .experiments/code_review/{llama-405b/no_eval/pr_335 => code_reviews_20240827_163259/no_eval/pr_440}/comments.json (100%)
 create mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/review.md
 create mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/comments.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/review.md
 create mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/comments.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/issues.json
 rename .experiments/code_review/{llama-405b/no_eval/pr_335 => code_reviews_20240827_163259/no_eval/pr_5}/review.md (52%)
 create mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/comments.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/review.md
 rename .experiments/code_review/{llama-405b/no_eval/pr_440 => code_reviews_20240827_163829/no_eval/pr_232}/comments.json (100%)
 create mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/review.md
 rename .experiments/code_review/{sonnet-3.5/no_eval/pr_232 => code_reviews_20240827_163829/no_eval/pr_252}/comments.json (100%)
 create mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/review.md
 rename .experiments/code_review/{sonnet-3.5/no_eval/pr_252 => code_reviews_20240827_163829/no_eval/pr_335}/comments.json (100%)
 create mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/review.md
 create mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/comments.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/review.md
 rename .experiments/code_review/{sonnet-3.5/no_eval/pr_335 => code_reviews_20240827_163829/no_eval/pr_440}/comments.json (100%)
 create mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/review.md
 create mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/comments.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/review.md
 create mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/comments.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/review.md
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/comments.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/review.md
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/comments.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/review.md
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/comments.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/review.md
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/comments.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/review.md
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/comments.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/review.md
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/comments.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/review.md
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/comments.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/review.md
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/comments.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/issues.json
 create mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/review.md
 delete mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_222/comments.json
 delete mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_222/issues.json
 delete mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_222/review.md
 delete mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_335/comments.json
 delete mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_335/issues.json
 delete mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_335/review.md
 delete mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_440/issues.json
 delete mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_440/review.md
 delete mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_476/comments.json
 delete mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_476/issues.json
 delete mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_476/review.md
 delete mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_5/comments.json
 delete mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_5/issues.json
 delete mode 100644 .experiments/code_review/gpt-4o-mini/no_eval/pr_5/review.md
 delete mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_222/comments.json
 delete mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_222/issues.json
 delete mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_222/review.md
 delete mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_335/comments.json
 delete mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_335/issues.json
 delete mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_335/review.md
 delete mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_440/issues.json
 delete mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_440/review.md
 delete mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_476/comments.json
 delete mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_476/issues.json
 delete mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_476/review.md
 delete mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_5/comments.json
 delete mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_5/issues.json
 delete mode 100644 .experiments/code_review/gpt-4o/no_eval/pr_5/review.md
 delete mode 100644 .experiments/code_review/haiku/no_eval/pr_222/comments.json
 delete mode 100644 .experiments/code_review/haiku/no_eval/pr_222/issues.json
 delete mode 100644 .experiments/code_review/haiku/no_eval/pr_222/review.md
 delete mode 100644 .experiments/code_review/haiku/no_eval/pr_252/issues.json
 delete mode 100644 .experiments/code_review/haiku/no_eval/pr_252/review.md
 delete mode 100644 .experiments/code_review/haiku/no_eval/pr_335/issues.json
 delete mode 100644 .experiments/code_review/haiku/no_eval/pr_335/review.md
 delete mode 100644 .experiments/code_review/haiku/no_eval/pr_400/comments.json
 delete mode 100644 .experiments/code_review/haiku/no_eval/pr_400/issues.json
 delete mode 100644 .experiments/code_review/haiku/no_eval/pr_400/review.md
 delete mode 100644 .experiments/code_review/haiku/no_eval/pr_440/issues.json
 delete mode 100644 .experiments/code_review/haiku/no_eval/pr_440/review.md
 delete mode 100644 .experiments/code_review/haiku/no_eval/pr_476/comments.json
 delete mode 100644 .experiments/code_review/haiku/no_eval/pr_476/issues.json
 delete mode 100644 .experiments/code_review/haiku/no_eval/pr_476/review.md
 delete mode 100644 .experiments/code_review/haiku/no_eval/pr_5/comments.json
 delete mode 100644 .experiments/code_review/haiku/no_eval/pr_5/issues.json
 delete mode 100644 .experiments/code_review/haiku/no_eval/pr_5/review.md
 delete mode 100644 .experiments/code_review/llama-405b/no_eval/pr_222/comments.json
 delete mode 100644 .experiments/code_review/llama-405b/no_eval/pr_222/issues.json
 delete mode 100644 .experiments/code_review/llama-405b/no_eval/pr_222/review.md
 delete mode 100644 .experiments/code_review/llama-405b/no_eval/pr_232/issues.json
 delete mode 100644 .experiments/code_review/llama-405b/no_eval/pr_232/review.md
 delete mode 100644 .experiments/code_review/llama-405b/no_eval/pr_252/issues.json
 delete mode 100644 .experiments/code_review/llama-405b/no_eval/pr_252/review.md
 delete mode 100644 .experiments/code_review/llama-405b/no_eval/pr_335/issues.json
 delete mode 100644 .experiments/code_review/llama-405b/no_eval/pr_400/comments.json
 delete mode 100644 .experiments/code_review/llama-405b/no_eval/pr_400/issues.json
 delete mode 100644 .experiments/code_review/llama-405b/no_eval/pr_400/review.md
 delete mode 100644 .experiments/code_review/llama-405b/no_eval/pr_440/issues.json
 delete mode 100644 .experiments/code_review/llama-405b/no_eval/pr_440/review.md
 delete mode 100644 .experiments/code_review/llama-405b/no_eval/pr_476/comments.json
 delete mode 100644 .experiments/code_review/llama-405b/no_eval/pr_476/issues.json
 delete mode 100644 .experiments/code_review/llama-405b/no_eval/pr_476/review.md
 delete mode 100644 .experiments/code_review/llama-405b/no_eval/pr_5/comments.json
 delete mode 100644 .experiments/code_review/llama-405b/no_eval/pr_5/issues.json
 delete mode 100644 .experiments/code_review/llama-405b/no_eval/pr_5/review.md
 delete mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_222/comments.json
 delete mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_222/issues.json
 delete mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_222/review.md
 delete mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_232/issues.json
 delete mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_232/review.md
 delete mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_252/issues.json
 delete mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_252/review.md
 delete mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_335/issues.json
 delete mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_335/review.md
 delete mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_400/comments.json
 delete mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_400/issues.json
 delete mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_400/review.md
 delete mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_476/comments.json
 delete mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_476/issues.json
 delete mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_476/review.md
 delete mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_5/comments.json
 delete mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_5/issues.json
 delete mode 100644 .experiments/code_review/sonnet-3.5/no_eval/pr_5/review.md

diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/comments.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/comments.json
new file mode 100644
index 00000000..259d5518
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/comments.json
@@ -0,0 +1,102 @@
+[
+  {
+    "category": "Security",
+    "description": "Hardcoded API keys in config.json",
+    "impact": "critical",
+    "rationale": "Hardcoding API keys can lead to security vulnerabilities if the code is shared or exposed. It is a best practice to use environment variables for sensitive information.",
+    "recommendation": "Replace hardcoded API keys with environment variables.",
+    "current_code": "\"api_key\": \"os.environ/AZURE_API_KEY\"",
+    "suggested_code": "\"api_key\": \"${AZURE_API_KEY}\"",
+    "file_path": "config.json",
+    "start_line": 13,
+    "end_line": 23,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 9
+  },
+  {
+    "category": "Security",
+    "description": "Potential SQL injection vulnerability in `store_abstraction_and_embedding` method.",
+    "impact": "critical",
+    "rationale": "Directly inserting user input into SQL queries can lead to SQL injection attacks.",
+    "recommendation": "Use parameterized queries to prevent SQL injection.",
+    "current_code": "embedding_query = text(\n    \"\"\"\n    INSERT INTO function_embeddings (function_id, vector)\n    VALUES (:function_id, :vector)\n    ON CONFLICT (function_id) DO UPDATE SET vector = EXCLUDED.vector\n    \"\"\"\n)",
+    "suggested_code": "embedding_query = text(\n    \"\"\"\n    INSERT INTO function_embeddings (function_id, vector)\n    VALUES (:function_id, :vector)\n    ON CONFLICT (function_id) DO UPDATE SET vector = EXCLUDED.vector\n    \"\"\"\n).bindparams(function_id=function_id, vector=embedding)",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 158,
+    "end_line": 163,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 9
+  },
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Docker",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to Dockerfile, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_path": "Dockerfile",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Docker",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to docker-compose.yml, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "15",
+    "end_line": "15",
+    "side": "RIGHT",
+    "file_path": "docker-compose.yml",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Version Control",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to .gitignore, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "164",
+    "end_line": "164",
+    "side": "RIGHT",
+    "file_path": ".gitignore",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Database",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to db_setup/init.sql, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_path": "db_setup/init.sql",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/issues.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/issues.json
new file mode 100644
index 00000000..3c296611
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/issues.json
@@ -0,0 +1,372 @@
+[
+  {
+    "category": "Security",
+    "description": "Hardcoded API keys in config.json",
+    "impact": "critical",
+    "rationale": "Hardcoding API keys can lead to security vulnerabilities if the code is shared or exposed. It is a best practice to use environment variables for sensitive information.",
+    "recommendation": "Replace hardcoded API keys with environment variables.",
+    "current_code": "\"api_key\": \"os.environ/AZURE_API_KEY\"",
+    "suggested_code": "\"api_key\": \"${AZURE_API_KEY}\"",
+    "file_path": "config.json",
+    "start_line": 13,
+    "end_line": 23,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 9
+  },
+  {
+    "category": "Performance",
+    "description": "Inefficient Dockerfile layer management",
+    "impact": "high",
+    "rationale": "Combining multiple RUN commands into a single command can reduce the number of layers in the Docker image, leading to smaller image sizes and faster build times.",
+    "recommendation": "Combine the RUN commands into a single command.",
+    "current_code": "RUN apt-get update && apt-get install -y \\\n    git \\\n    build-essential \\\n    && rm -rf /var/lib/apt/lists/*",
+    "suggested_code": "RUN apt-get update && apt-get install -y \\\n    git \\\n    build-essential \\\n    && rm -rf /var/lib/apt/lists/*",
+    "file_path": "Dockerfile",
+    "start_line": 8,
+    "end_line": 11,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 7
+  },
+  {
+    "category": "Error Handling",
+    "description": "Lack of error handling in install_tree_sitter_languages.sh",
+    "impact": "medium",
+    "rationale": "The script does not check if the git clone or tree-sitter generate commands succeed, which can lead to silent failures.",
+    "recommendation": "Add error handling to check the success of git clone and tree-sitter generate commands.",
+    "current_code": "git clone \"https://github.com/tree-sitter/tree-sitter-$lang\" \"$LANGUAGE_DIR/tree-sitter-$lang\"\ntree-sitter generate",
+    "suggested_code": "git clone \"https://github.com/tree-sitter/tree-sitter-$lang\" \"$LANGUAGE_DIR/tree-sitter-$lang\" ||{echo 'git clone failed'; exit 1;}\ntree-sitter generate ||{echo 'tree-sitter generate failed'; exit 1;}",
+    "file_path": "install_tree_sitter_languages.sh",
+    "start_line": 24,
+    "end_line": 34,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Readability",
+    "description": "Commented-out code in provider.py",
+    "impact": "low",
+    "rationale": "Commented-out code can clutter the codebase and make it harder to read and maintain.",
+    "recommendation": "Remove commented-out code if it is no longer needed.",
+    "current_code": "# for model in self.config[\"language_model\"][\"models\"]:\n#     if model[\"model_name\"] == \"embedding\":\n#         break",
+    "suggested_code": "",
+    "file_path": "kaizen/llms/provider.py",
+    "start_line": 239,
+    "end_line": 241,
+    "change_type": "deletion",
+    "sentiment": "neutral",
+    "severity": 3
+  },
+  {
+    "category": "Best Practices",
+    "description": "No newline at end of file",
+    "impact": "trivial",
+    "rationale": "A newline at the end of a file is a POSIX standard and helps with version control diffs.",
+    "recommendation": "Add a newline at the end of the file.",
+    "current_code": "",
+    "suggested_code": "\n",
+    "file_path": "install_tree_sitter_languages.sh",
+    "start_line": 48,
+    "end_line": 48,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 1
+  },
+  {
+    "category": "Error Handling",
+    "description": "Lack of error handling in database query execution.",
+    "impact": "high",
+    "rationale": "If the database query fails, the code does not handle the exception, which could lead to application crashes or unhandled exceptions.",
+    "recommendation": "Add try-except blocks to handle potential exceptions during the database query execution.",
+    "current_code": "cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))",
+    "suggested_code": "try:\n    cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))\nexcept Exception as e:\n    # Handle exception (e.g., log error, raise custom exception)\n    print(f\"Database query failed:{e}\")",
+    "file_path": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 41,
+    "end_line": 41,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 8
+  },
+  {
+    "category": "Performance",
+    "description": "Inefficient normalization of query embedding.",
+    "impact": "medium",
+    "rationale": "Using numpy for normalization is efficient, but it can be optimized further by avoiding redundant operations.",
+    "recommendation": "Normalize the query embedding in-place to avoid creating unnecessary copies.",
+    "current_code": "query_embedding_normalized = query_embedding_np / np.linalg.norm(query_embedding_np)",
+    "suggested_code": "query_embedding_np /= np.linalg.norm(query_embedding_np)",
+    "file_path": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 16,
+    "end_line": 16,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Code Readability",
+    "description": "SQL query string formatting using f-strings.",
+    "impact": "low",
+    "rationale": "Using f-strings for SQL queries can lead to SQL injection vulnerabilities if not handled properly.",
+    "recommendation": "Use parameterized queries to prevent SQL injection and improve readability.",
+    "current_code": "query = f\"\"\"\nSELECT \n    e.node_id,\n    e.text,\n    e.metadata,\n    1 - (e.embedding <=> %s::vector) as similarity\nFROM \n{self.table_name}e\nJOIN \n    function_abstractions fa ON e.node_id = fa.function_id::text\nJOIN \n    files f ON fa.file_id = f.file_id\nWHERE \n    f.repo_id = %s\nORDER BY \n    similarity DESC\nLIMIT \n    %s\n\"\"\"",
+    "suggested_code": "query = \"\"\"\nSELECT \n    e.node_id,\n    e.text,\n    e.metadata,\n    1 - (e.embedding <=> %s::vector) as similarity\nFROM \n    %s e\nJOIN \n    function_abstractions fa ON e.node_id = fa.function_id::text\nJOIN \n    files f ON fa.file_id = f.file_id\nWHERE \n    f.repo_id = %s\nORDER BY \n    similarity DESC\nLIMIT \n    %s\n\"\"\"",
+    "file_path": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 19,
+    "end_line": 36,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 4
+  },
+  {
+    "category": "Type Hinting",
+    "description": "Missing type hint for the `correction` parameter in `add_feedback` method.",
+    "impact": "low",
+    "rationale": "Providing type hints improves code readability and helps with static analysis.",
+    "recommendation": "Add type hint for the `correction` parameter.",
+    "current_code": "def add_feedback(\n    self, code_id: str, abstraction: str, rating: int, correction: str = None\n) -> None:",
+    "suggested_code": "def add_feedback(\n    self, code_id: str, abstraction: str, rating: int, correction: Optional[str] = None\n) -> None:",
+    "file_path": "kaizen/retriever/feedback_system.py",
+    "start_line": 9,
+    "end_line": 10,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 3
+  },
+  {
+    "category": "Documentation",
+    "description": "Lack of docstrings for methods.",
+    "impact": "low",
+    "rationale": "Docstrings provide essential information about the purpose and usage of methods, improving code maintainability.",
+    "recommendation": "Add docstrings to all methods to describe their functionality, parameters, and return values.",
+    "current_code": "def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:",
+    "suggested_code": "\"\"\"\nExecutes a custom query to retrieve similar embeddings.\n\n:param query_embedding: The embedding to query against.\n:param repo_id: The repository ID to filter results.\n:param similarity_top_k: The number of top similar results to return.\n:return: A list of dictionaries containing the results.\n\"\"\"\ndef custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:",
+    "file_path": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 13,
+    "end_line": 13,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 2
+  },
+  {
+    "category": "Error Handling",
+    "description": "Exception handling in `generate_abstraction` method is too generic.",
+    "impact": "medium",
+    "rationale": "Catching all exceptions with a generic `except Exception as e` can obscure the root cause of errors and make debugging more difficult.",
+    "recommendation": "Handle specific exceptions where possible and log meaningful error messages.",
+    "current_code": "except Exception as e:\n    raise e",
+    "suggested_code": "except SomeSpecificException as e:\n    logger.error(f\"Specific error occurred:{str(e)}\")\n    raise e\nexcept Exception as e:\n    logger.error(f\"Unexpected error:{str(e)}\")\n    raise e",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 218,
+    "end_line": 219,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Resource Management",
+    "description": "Potential resource leak in `store_abstraction_and_embedding` method.",
+    "impact": "high",
+    "rationale": "The database connection is not explicitly closed, which can lead to resource leaks.",
+    "recommendation": "Use a context manager to ensure the connection is properly closed.",
+    "current_code": "with self.engine.begin() as connection:\n    connection.execute(...)",
+    "suggested_code": "with self.engine.connect() as connection:\n    with connection.begin():\n        connection.execute(...)",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 157,
+    "end_line": 171,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 7
+  },
+  {
+    "category": "Performance",
+    "description": "Inefficient string concatenation in `generate_abstraction` method.",
+    "impact": "low",
+    "rationale": "Using f-strings for multi-line strings can be inefficient and harder to read.",
+    "recommendation": "Use triple-quoted strings for multi-line text.",
+    "current_code": "prompt = f\"\"\"Generate a concise yet comprehensive abstract description of the following{language}code block. \nInclude information about:\n1. The purpose or functionality of the code\n2. Input parameters and return values (if applicable)\n3. Any important algorithms or data structures used\n4. Key dependencies or external libraries used\n5. Any notable design patterns or architectural choices\n6. Potential edge cases or error handling\n\nCode:\n```{language}\n{code_block}\n```\n\"\"\"",
+    "suggested_code": "prompt = f\"\"\"\nGenerate a concise yet comprehensive abstract description of the following{language}code block. \nInclude information about:\n1. The purpose or functionality of the code\n2. Input parameters and return values (if applicable)\n3. Any important algorithms or data structures used\n4. Key dependencies or external libraries used\n5. Any notable design patterns or architectural choices\n6. Potential edge cases or error handling\n\nCode:\n```{language}\n{code_block}\n```\n\"\"\"",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 186,
+    "end_line": 198,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 3
+  },
+  {
+    "category": "Security",
+    "description": "Potential SQL injection vulnerability in `store_abstraction_and_embedding` method.",
+    "impact": "critical",
+    "rationale": "Directly inserting user input into SQL queries can lead to SQL injection attacks.",
+    "recommendation": "Use parameterized queries to prevent SQL injection.",
+    "current_code": "embedding_query = text(\n    \"\"\"\n    INSERT INTO function_embeddings (function_id, vector)\n    VALUES (:function_id, :vector)\n    ON CONFLICT (function_id) DO UPDATE SET vector = EXCLUDED.vector\n    \"\"\"\n)",
+    "suggested_code": "embedding_query = text(\n    \"\"\"\n    INSERT INTO function_embeddings (function_id, vector)\n    VALUES (:function_id, :vector)\n    ON CONFLICT (function_id) DO UPDATE SET vector = EXCLUDED.vector\n    \"\"\"\n).bindparams(function_id=function_id, vector=embedding)",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 158,
+    "end_line": 163,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 9
+  },
+  {
+    "category": "Readability",
+    "description": "Logging configuration is hardcoded.",
+    "impact": "low",
+    "rationale": "Hardcoding logging configuration makes it difficult to change log settings without modifying the code.",
+    "recommendation": "Move logging configuration to a configuration file or environment variables.",
+    "current_code": "logging.basicConfig(\n    level=logging.INFO, format=\"%(asctime)s - %(levelname)s - %(message)s\"\n)",
+    "suggested_code": "logging.basicConfig(\n    level=os.getenv('LOG_LEVEL', logging.INFO),\n    format=os.getenv('LOG_FORMAT', \"%(asctime)s - %(levelname)s - %(message)s\")\n)",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 23,
+    "end_line": 25,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 2
+  },
+  {
+    "category": "Error Handling",
+    "description": "General exception caught without specific handling",
+    "impact": "high",
+    "rationale": "Catching a general exception can obscure the root cause of the error and make debugging difficult. It is better to catch specific exceptions.",
+    "recommendation": "Catch specific exceptions where possible and handle them appropriately.",
+    "current_code": "except Exception as e:",
+    "suggested_code": "except ImportError as e:\n    logger.error(f\"ImportError:{str(e)}\")\n    raise\nexcept ValueError as e:\n    logger.error(f\"ValueError:{str(e)}\")\n    raise",
+    "file_path": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 28,
+    "end_line": 30,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 7
+  },
+  {
+    "category": "Logging",
+    "description": "Logging configuration should be placed in a main guard",
+    "impact": "medium",
+    "rationale": "Setting up logging configuration at the module level can lead to unexpected behavior when the module is imported.",
+    "recommendation": "Move logging configuration inside a main guard.",
+    "current_code": "logging.basicConfig(level=logging.INFO)",
+    "suggested_code": "if __name__ == \"__main__\":\n    logging.basicConfig(level=logging.INFO)",
+    "file_path": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 8,
+    "end_line": 8,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Performance",
+    "description": "Inefficient string concatenation in logging",
+    "impact": "low",
+    "rationale": "Using f-strings in logging can lead to unnecessary string concatenation even if the log level is not enabled.",
+    "recommendation": "Use lazy formatting in logging.",
+    "current_code": "logger.error(f\"Failed to load language{language}:{str(e)}\")",
+    "suggested_code": "logger.error(\"Failed to load language %s: %s\", language, str(e))",
+    "file_path": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 29,
+    "end_line": 29,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 3
+  },
+  {
+    "category": "Code Duplication",
+    "description": "Duplicate code for printing chunks",
+    "impact": "medium",
+    "rationale": "Duplicated code can lead to maintenance issues and inconsistencies.",
+    "recommendation": "Refactor the code to avoid duplication.",
+    "current_code": "print_chunks(\"JavaScript\", chunk_code(javascript_code, \"javascript\"))",
+    "suggested_code": "print_chunks(\"JavaScript\", chunk_code(javascript_code, \"javascript\"))",
+    "file_path": "tests/retriever/test_chunker.py",
+    "start_line": 99,
+    "end_line": 100,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 4
+  },
+  {
+    "category": "Code Style",
+    "description": "Unused import statement",
+    "impact": "trivial",
+    "rationale": "Unused imports can clutter the code and affect readability.",
+    "recommendation": "Remove the unused import statement.",
+    "current_code": "import json",
+    "suggested_code": "",
+    "file_path": "tests/retriever/test_chunker.py",
+    "start_line": 2,
+    "end_line": 2,
+    "change_type": "deletion",
+    "sentiment": "neutral",
+    "severity": 1
+  },
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Docker",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to Dockerfile, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_path": "Dockerfile",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Docker",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to docker-compose.yml, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "15",
+    "end_line": "15",
+    "side": "RIGHT",
+    "file_path": "docker-compose.yml",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Version Control",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to .gitignore, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "164",
+    "end_line": "164",
+    "side": "RIGHT",
+    "file_path": ".gitignore",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Database",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to db_setup/init.sql, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_path": "db_setup/init.sql",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/review.md b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/review.md
new file mode 100644
index 00000000..06ec2b07
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/review.md
@@ -0,0 +1,166 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/222
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 25
+- Critical: 7
+- Important: 0
+- Minor: 0
+- Files Affected: 12
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Security (7 issues)</strong></summary>
+
+### 1. Hardcoded API keys in config.json
+📁 **File:** `config.json:13`
+⚖️ **Severity:** 9/10
+🔍 **Description:** Hardcoded API keys in config.json
+💡 **Solution:** 
+
+**Current Code:**
+```python
+"api_key": "os.environ/AZURE_API_KEY"
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 2. Potential SQL injection vulnerability in `store_abstraction_and_embedding` method.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:158`
+⚖️ **Severity:** 9/10
+🔍 **Description:** Potential SQL injection vulnerability in `store_abstraction_and_embedding` method.
+💡 **Solution:** 
+
+**Current Code:**
+```python
+embedding_query = text(
+    """
+    INSERT INTO function_embeddings (function_id, vector)
+    VALUES (:function_id, :vector)
+    ON CONFLICT (function_id) DO UPDATE SET vector = EXCLUDED.vector
+    """
+)
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 3. Changes made to sensitive file
+📁 **File:** `config.json:4`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 4. Changes made to sensitive file
+📁 **File:** `Dockerfile:4`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 5. Changes made to sensitive file
+📁 **File:** `docker-compose.yml:15`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 6. Changes made to sensitive file
+📁 **File:** `.gitignore:164`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 7. Changes made to sensitive file
+📁 **File:** `db_setup/init.sql:1`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (9 issues)</strong></summary>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-2024-05-13)
+{"prompt_tokens": 22541, "completion_tokens": 4406, "total_tokens": 26947}
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_440/comments.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/comments.json
similarity index 100%
rename from .experiments/code_review/gpt-4o-mini/no_eval/pr_440/comments.json
rename to .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/comments.json
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/issues.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/issues.json
new file mode 100644
index 00000000..d2e57187
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/issues.json
@@ -0,0 +1,227 @@
+[
+  {
+    "category": "readability",
+    "description": "Unnecessary comment removed",
+    "impact": "low",
+    "rationale": "The comment explaining the merging of lists was removed. While the code is self-explanatory, comments can help future developers understand the intent.",
+    "recommendation": "Consider adding a concise comment to explain the merging process.",
+    "current_code": "// merge spaces & memories to{item: \"memory\" | \"space\", date: Date, data: Content | StoredSpace}",
+    "suggested_code": "// Merge spaces and memories into a unified list with item type and date",
+    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 98,
+    "end_line": 98,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 2
+  },
+  {
+    "category": "maintainability",
+    "description": "Type alias introduced for props",
+    "impact": "medium",
+    "rationale": "Introducing a type alias for component props improves maintainability by centralizing the type definition.",
+    "recommendation": "Ensure that the type alias is updated if the props change.",
+    "current_code": "}: TMemoriesPage)",
+    "suggested_code": "}: TMemoriesPage)",
+    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 51,
+    "end_line": 51,
+    "change_type": "addition",
+    "sentiment": "positive",
+    "severity": 4
+  },
+  {
+    "category": "error_handling",
+    "description": "Error message improved for clarity",
+    "impact": "low",
+    "rationale": "The error message was slightly modified for better readability.",
+    "recommendation": "No further changes needed.",
+    "current_code": "toast.error(\"Failed to delete space\")",
+    "suggested_code": "toast.error(\"Failed to delete the space\")",
+    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 73,
+    "end_line": 73,
+    "change_type": "addition",
+    "sentiment": "positive",
+    "severity": 1
+  },
+  {
+    "category": "performance",
+    "description": "Removed unused imports",
+    "impact": "low",
+    "rationale": "Removing unused imports can slightly improve performance and reduce bundle size.",
+    "recommendation": "Continue to ensure that imports are only added when necessary.",
+    "current_code": "import{MemoriesIcon, NextIcon, SearchIcon, UrlIcon}from \"@repo/ui/icons\";",
+    "suggested_code": "import{MemoriesIcon, NextIcon, UrlIcon}from \"@repo/ui/icons\";",
+    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 4,
+    "end_line": 4,
+    "change_type": "modification",
+    "sentiment": "positive",
+    "severity": 2
+  },
+  {
+    "category": "usability",
+    "description": "Button styling improved",
+    "impact": "medium",
+    "rationale": "Improving button styling enhances the user interface and user experience.",
+    "recommendation": "Ensure consistent styling across all buttons in the application.",
+    "current_code": "className={`transition px-6 py-2 rounded-xl hover:text-[#369DFD]\" text-[#B3BCC5] bg-secondary hover:bg-secondary hover:text-[#76a3cc]`}",
+    "suggested_code": "className={`transition px-4 py-2 rounded-lg text-[#B3BCC5] bg-secondary hover:bg-secondary hover:text-[#76a3cc]`}",
+    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 214,
+    "end_line": 214,
+    "change_type": "addition",
+    "sentiment": "positive",
+    "severity": 3
+  },
+  {
+    "category": "code_style",
+    "description": "Consistent spacing in JSX",
+    "impact": "trivial",
+    "rationale": "Ensuring consistent spacing in JSX improves code readability.",
+    "recommendation": "Adopt a consistent code style guide and use linters to enforce it.",
+    "current_code": "{title.slice(0, 2).toUpperCase()}{id}",
+    "suggested_code": "{title.slice(0, 2).toUpperCase()}{id}",
+    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 292,
+    "end_line": 292,
+    "change_type": "addition",
+    "sentiment": "positive",
+    "severity": 1
+  },
+  {
+    "category": "performance",
+    "description": "Removed unnecessary useEffect import",
+    "impact": "low",
+    "rationale": "Removing unused imports can slightly improve performance and reduce bundle size.",
+    "recommendation": "Continue to ensure that imports are only added when necessary.",
+    "current_code": "import React,{useEffect, useState}from \"react\";",
+    "suggested_code": "import React,{useState}from \"react\";",
+    "file_path": "apps/web/app/(dash)/home/queryinput.tsx",
+    "start_line": 3,
+    "end_line": 3,
+    "change_type": "modification",
+    "sentiment": "positive",
+    "severity": 2
+  },
+  {
+    "category": "Code Duplication",
+    "description": "Duplicated logic for fetching spaces and handling form submission.",
+    "impact": "high",
+    "rationale": "Duplicated code increases maintenance overhead and the risk of inconsistencies. Any future changes will need to be made in multiple places, increasing the chance of errors.",
+    "recommendation": "Extract the duplicated logic into reusable hooks or utility functions.",
+    "current_code": "useEffect(() =>{(async () =>{let spaces = await getSpaces(); if (!spaces.success || !spaces.data){toast.warning('Unable to get spaces',{richColors: true}); setSpaces([]); return;}setSpaces(spaces.data);})();},[]);",
+    "suggested_code": "const useFetchSpaces = () =>{const[spaces, setSpaces] = useState<StoredSpace[]>([]); useEffect(() =>{(async () =>{let spaces = await getSpaces(); if (!spaces.success || !spaces.data){toast.warning('Unable to get spaces',{richColors: true}); setSpaces([]); return;}setSpaces(spaces.data);})();},[]); return spaces;};",
+    "file_path": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 173,
+    "end_line": 186,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 7
+  },
+  {
+    "category": "Error Handling",
+    "description": "Improper error handling in form submission.",
+    "impact": "medium",
+    "rationale": "Throwing errors without proper handling can lead to unhandled promise rejections and a poor user experience.",
+    "recommendation": "Use try-catch blocks to handle errors gracefully and provide feedback to the user.",
+    "current_code": "const handleSubmit = async (content?: string, spaces?: number[]) =>{setDialogClose(); if (!content || content.length === 0){throw new Error('Content is required');}const cont = await createMemory({content: content, spaces: spaces ?? undefined}); setContent(''); setSelectedSpaces([]); if (cont.success){toast.success('Memory queued',{richColors: true});}else{toast.error(`Memory creation failed: ${cont.error}`); throw new Error(`Memory creation failed: ${cont.error}`); return cont;}};",
+    "suggested_code": "const handleSubmit = async (content?: string, spaces?: number[]) =>{setDialogClose(); if (!content || content.length === 0){toast.error('Content is required'); return;}try{const cont = await createMemory({content: content, spaces: spaces ?? undefined}); setContent(''); setSelectedSpaces([]); if (cont.success){toast.success('Memory queued',{richColors: true});}else{toast.error(`Memory creation failed: ${cont.error}`);}}catch (error){toast.error(`An error occurred: ${error.message}`);}};",
+    "file_path": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 213,
+    "end_line": 233,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Code Readability",
+    "description": "Long and complex inline functions.",
+    "impact": "low",
+    "rationale": "Long inline functions can make the code harder to read and understand.",
+    "recommendation": "Extract complex inline functions into named functions.",
+    "current_code": "action={async (e: FormData) =>{const content = e.get('content')?.toString(); toast.promise(handleSubmit(content, selectedSpaces),{loading: (<span><PlusCircleIcon className='w-4 h-4 inline mr-2 text-white animate-spin' /> Creating memory...</span>), success: (data) => 'Memory queued', error: (error) => error.message, richColors: true,});}}",
+    "suggested_code": "const handleFormSubmit = async (e: FormData) =>{const content = e.get('content')?.toString(); toast.promise(handleSubmit(content, selectedSpaces),{loading: (<span><PlusCircleIcon className='w-4 h-4 inline mr-2 text-white animate-spin' /> Creating memory...</span>), success: (data) => 'Memory queued', error: (error) => error.message, richColors: true,});};",
+    "file_path": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 237,
+    "end_line": 249,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 4
+  },
+  {
+    "category": "Performance",
+    "description": "Inefficient use of useMemo for options.",
+    "impact": "low",
+    "rationale": "Using useMemo for simple transformations can be unnecessary and might not provide a performance benefit.",
+    "recommendation": "Consider removing useMemo if the performance benefit is negligible.",
+    "current_code": "const options = useMemo(() => spaces.map((x) => ({label: x.name, value: x.id.toString()})),[spaces]);",
+    "suggested_code": "const options = spaces.map((x) => ({label: x.name, value: x.id.toString()}));",
+    "file_path": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 208,
+    "end_line": 211,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 3
+  },
+  {
+    "category": "Error Handling",
+    "description": "Missing error handling for onSubmit function.",
+    "impact": "high",
+    "rationale": "The onSubmit function lacks error handling, which could lead to unhandled exceptions if the function fails.",
+    "recommendation": "Wrap the onSubmit call in a try-catch block to handle potential errors gracefully.",
+    "current_code": "onClick={() => onSubmit(inputValue)}",
+    "suggested_code": "onClick={async () =>{try{await onSubmit(inputValue);}catch (error){console.error('Submission failed', error);}}}",
+    "file_path": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 99,
+    "end_line": 99,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 8
+  },
+  {
+    "category": "Performance",
+    "description": "Inefficient filtering of options.",
+    "impact": "medium",
+    "rationale": "The filtering of options inside the render method can be optimized to avoid unnecessary computations on each render.",
+    "recommendation": "Memoize the filteredOptions using useMemo to avoid recalculating on each render.",
+    "current_code": "const filteredOptions = options.filter((option) => !selectedSpaces.includes(parseInt(option.value)));",
+    "suggested_code": "const filteredOptions = useMemo(() => options.filter((option) => !selectedSpaces.includes(parseInt(option.value))),[options, selectedSpaces]);",
+    "file_path": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 55,
+    "end_line": 57,
+    "change_type": "modification",
+    "sentiment": "positive",
+    "severity": 5
+  },
+  {
+    "category": "Code Readability",
+    "description": "Inline styles and class names are hard to read.",
+    "impact": "low",
+    "rationale": "Inline styles and complex class names reduce readability and maintainability of the code.",
+    "recommendation": "Extract class names into a separate variable or use a CSS-in-JS solution for better readability.",
+    "current_code": "className={`group flex bg-[#2F353C] h-min rounded-md ${selectedSpaces.length > 0 && \"p-2\"}transition-all mt-4 mb-4`}",
+    "suggested_code": "const commandClassNames = `group flex bg-[#2F353C] h-min rounded-md ${selectedSpaces.length > 0 && \"p-2\"}transition-all mt-4 mb-4`;\n...\n<Command className={commandClassNames}>",
+    "file_path": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 61,
+    "end_line": 61,
+    "change_type": "modification",
+    "sentiment": "positive",
+    "severity": 3
+  },
+  {
+    "category": "Code Consistency",
+    "description": "Inconsistent use of single and double quotes.",
+    "impact": "trivial",
+    "rationale": "Inconsistent use of quotes can lead to confusion and reduce code readability.",
+    "recommendation": "Standardize the use of quotes throughout the file, preferably using single quotes for consistency.",
+    "current_code": "className={`group flex bg-[#2F353C] h-min rounded-md ${selectedSpaces.length > 0 && \"p-2\"}transition-all mt-4 mb-4`}",
+    "suggested_code": "className={`group flex bg-[#2F353C] h-min rounded-md ${selectedSpaces.length > 0 && 'p-2'}transition-all mt-4 mb-4`}",
+    "file_path": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 61,
+    "end_line": 61,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 2
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/review.md b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/review.md
new file mode 100644
index 00000000..8cd757ad
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/review.md
@@ -0,0 +1,41 @@
+PR URL: https://github.com/supermemoryai/supermemory/pull/232
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 15
+- Critical: 0
+- Important: 0
+- Minor: 0
+- Files Affected: 4
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (9 issues)</strong></summary>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-2024-05-13)
+{"prompt_tokens": 16642, "completion_tokens": 3356, "total_tokens": 19998}
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_440/comments.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/comments.json
similarity index 100%
rename from .experiments/code_review/gpt-4o/no_eval/pr_440/comments.json
rename to .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/comments.json
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/issues.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/issues.json
new file mode 100644
index 00000000..12fe862d
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/issues.json
@@ -0,0 +1,77 @@
+[
+  {
+    "category": "error_handling",
+    "description": "Lack of error handling for JSON parsing",
+    "impact": "high",
+    "rationale": "If the JSON data is malformed or not as expected, the `json.loads` call will raise an exception, which is not currently handled.",
+    "recommendation": "Add a try-except block to handle JSON parsing errors gracefully.",
+    "current_code": "parsed_data = json.loads(json_data)",
+    "suggested_code": "try:\n    parsed_data = json.loads(json_data)\nexcept json.JSONDecodeError as e:\n    logging.error(f\"Failed to parse JSON:{e}\")\n    return None",
+    "file_path": "kaizen/helpers/parser.py",
+    "start_line": 47,
+    "end_line": 47,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 8
+  },
+  {
+    "category": "performance",
+    "description": "Unnecessary multiple calls to `generate_twitter_post` and `generate_linkedin_post`",
+    "impact": "medium",
+    "rationale": "Generating posts multiple times with the same summary and user is redundant and can impact performance.",
+    "recommendation": "Store the generated posts in variables and reuse them.",
+    "current_code": "twitter_post = work_summary_generator.generate_twitter_post(summary, user=\"oss_example\")\nlinkedin_post = work_summary_generator.generate_linkedin_post(summary, user=\"oss_example\")",
+    "suggested_code": "twitter_post = work_summary_generator.generate_twitter_post(summary, user=\"oss_example\")\nlinkedin_post = work_summary_generator.generate_linkedin_post(summary, user=\"oss_example\")",
+    "file_path": "examples/work_summarizer/main.py",
+    "start_line": 59,
+    "end_line": 60,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "readability",
+    "description": "Long lines exceeding 80 characters",
+    "impact": "low",
+    "rationale": "Long lines can affect readability and maintainability of the code.",
+    "recommendation": "Break long lines into multiple lines for better readability.",
+    "current_code": "linkedin_post = work_summary_generator.generate_linkedin_post(summary, user=\"oss_example\")",
+    "suggested_code": "linkedin_post = work_summary_generator.generate_linkedin_post(\n    summary, user=\"oss_example\"\n)",
+    "file_path": "examples/work_summarizer/main.py",
+    "start_line": 60,
+    "end_line": 62,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 3
+  },
+  {
+    "category": "documentation",
+    "description": "Missing docstrings for new functions",
+    "impact": "medium",
+    "rationale": "Docstrings are important for understanding the purpose and usage of functions.",
+    "recommendation": "Add docstrings to the `generate_twitter_post` and `generate_linkedin_post` functions.",
+    "current_code": "",
+    "suggested_code": "def generate_twitter_post(\n    self,\n    summary: Dict,\n    user: Optional[str] = None,\n) -> str:\n    \"\"\"Generate a Twitter post based on the summary.\"\"\"\n    prompt = TWITTER_POST_PROMPT.format(SUMMARY=summary)\n    response, _ = self.provider.chat_completion(prompt, user=user)\n    return parser.extract_markdown_content(response)\n\n\ndef generate_linkedin_post(\n    self,\n    summary: Dict,\n    user: Optional[str] = None,\n) -> str:\n    \"\"\"Generate a LinkedIn post based on the summary.\"\"\"\n    prompt = LINKEDIN_POST_PROMPT.format(SUMMARY=summary)\n    response, _ = self.provider.chat_completion(prompt, user=user)\n    return parser.extract_markdown_content(response)",
+    "file_path": "kaizen/reviewer/work_summarizer.py",
+    "start_line": 58,
+    "end_line": 74,
+    "change_type": "addition",
+    "sentiment": "positive",
+    "severity": 6
+  },
+  {
+    "category": "syntax",
+    "description": "Missing trailing comma in import statement",
+    "impact": "low",
+    "rationale": "Trailing commas in import statements help to avoid syntax errors when adding new imports.",
+    "recommendation": "Add a trailing comma to the import statement.",
+    "current_code": "PR_REVIEW_EVALUATION_PROMPT",
+    "suggested_code": "PR_REVIEW_EVALUATION_PROMPT,",
+    "file_path": "kaizen/reviewer/code_review.py",
+    "start_line": 12,
+    "end_line": 12,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 2
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/review.md b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/review.md
new file mode 100644
index 00000000..13e4a3e7
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/review.md
@@ -0,0 +1,41 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/252
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 5
+- Critical: 0
+- Important: 0
+- Minor: 0
+- Files Affected: 4
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (2 issues)</strong></summary>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-2024-05-13)
+{"prompt_tokens": 4135, "completion_tokens": 1094, "total_tokens": 5229}
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_252/comments.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/comments.json
similarity index 100%
rename from .experiments/code_review/haiku/no_eval/pr_252/comments.json
rename to .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/comments.json
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/issues.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/issues.json
new file mode 100644
index 00000000..30e16352
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/issues.json
@@ -0,0 +1,62 @@
+[
+  {
+    "category": "Code Readability",
+    "description": "Replaced `reeval_response` parameter removal",
+    "impact": "medium",
+    "rationale": "The removal of the `reeval_response` parameter reduces the flexibility of the method. This parameter might be necessary for certain use cases, and removing it could limit the functionality.",
+    "recommendation": "Reintroduce the `reeval_response` parameter to maintain the method's flexibility and ensure it can handle all anticipated use cases.",
+    "current_code": "def _process_full_diff(self, prompt, user):",
+    "suggested_code": "def _process_full_diff(self, prompt, user, reeval_response):",
+    "file_path": "kaizen/generator/pr_description.py",
+    "start_line": 77,
+    "end_line": 80,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling",
+    "description": "Removed re-evaluation logic",
+    "impact": "high",
+    "rationale": "Removing the re-evaluation logic could lead to less accurate or incomplete responses. The re-evaluation step ensures that the response is thoroughly checked and improved if necessary.",
+    "recommendation": "Reintroduce the re-evaluation logic to ensure that the responses are accurate and complete.",
+    "current_code": "return desc",
+    "suggested_code": "if reeval_response:\n    resp = self._reevaluate_response(prompt, resp, user)\nreturn resp['desc']",
+    "file_path": "kaizen/generator/pr_description.py",
+    "start_line": 83,
+    "end_line": 88,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Code Consistency",
+    "description": "Inconsistent prompt naming",
+    "impact": "low",
+    "rationale": "The naming of the prompts should be consistent to avoid confusion and maintain code readability.",
+    "recommendation": "Ensure that all prompt names follow a consistent naming convention.",
+    "current_code": "PR_DESCRIPTION_SYSTEM_PROMPT",
+    "suggested_code": "CODE_REVIEW_SYSTEM_PROMPT",
+    "file_path": "kaizen/generator/pr_description.py",
+    "start_line": 28,
+    "end_line": 29,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 3
+  },
+  {
+    "category": "Performance",
+    "description": "Removed detailed re-evaluation prompt",
+    "impact": "medium",
+    "rationale": "The detailed re-evaluation prompt ensures that the system can provide more accurate and refined feedback. Removing it could degrade the quality of the feedback.",
+    "recommendation": "Reintroduce the detailed re-evaluation prompt to maintain high-quality feedback.",
+    "current_code": "return desc",
+    "suggested_code": "return resp['desc']",
+    "file_path": "kaizen/generator/pr_description.py",
+    "start_line": 176,
+    "end_line": 201,
+    "change_type": "deletion",
+    "sentiment": "negative",
+    "severity": 6
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/review.md b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/review.md
new file mode 100644
index 00000000..e6170bdc
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/review.md
@@ -0,0 +1,41 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/335
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 4
+- Critical: 0
+- Important: 0
+- Minor: 0
+- Files Affected: 1
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (1 issues)</strong></summary>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-2024-05-13)
+{"prompt_tokens": 7000, "completion_tokens": 724, "total_tokens": 7724}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/comments.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/comments.json
new file mode 100644
index 00000000..1ceabeef
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/comments.json
@@ -0,0 +1,16 @@
+[
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "11",
+    "end_line": "11",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/issues.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/issues.json
new file mode 100644
index 00000000..cca018d1
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/issues.json
@@ -0,0 +1,421 @@
+[
+  {
+    "category": "Code Readability",
+    "description": "Inconsistent formatting in DESC_COLLAPSIBLE_TEMPLATE.",
+    "impact": "medium",
+    "rationale": "Inconsistent formatting can lead to confusion and maintenance issues. Ensuring consistent formatting improves readability and maintainability.",
+    "recommendation": "Ensure consistent formatting for multi-line strings.",
+    "current_code": "DESC_COLLAPSIBLE_TEMPLATE = \"<details><summary>Original Description</summary>{desc}</details>\"",
+    "suggested_code": "DESC_COLLAPSIBLE_TEMPLATE = \"<details><summary>Original Description</summary>\\n\\n{desc}\\n\\n</details>\"",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
+    "start_line": 5,
+    "end_line": 5,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 4
+  },
+  {
+    "category": "Code Readability",
+    "description": "Redundant comments in the test file.",
+    "impact": "low",
+    "rationale": "Comments should add value and not state the obvious. Redundant comments can clutter the code and reduce readability.",
+    "recommendation": "Remove redundant comments that do not add value.",
+    "current_code": "# Mock logger",
+    "suggested_code": "",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_folder.py",
+    "start_line": 6,
+    "end_line": 6,
+    "change_type": "deletion",
+    "sentiment": "neutral",
+    "severity": 2
+  },
+  {
+    "category": "Error Handling",
+    "description": "Lack of error handling for invalid folder paths.",
+    "impact": "high",
+    "rationale": "Proper error handling ensures that the application can gracefully handle unexpected inputs and scenarios.",
+    "recommendation": "Add error handling for invalid folder paths.",
+    "current_code": "create_folder(folder_path)",
+    "suggested_code": "if not folder_path: raise ValueError('Folder path cannot be empty')\ncreate_folder(folder_path)",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_folder.py",
+    "start_line": 62,
+    "end_line": 62,
+    "change_type": "modification",
+    "sentiment": "positive",
+    "severity": 7
+  },
+  {
+    "category": "Test Coverage",
+    "description": "Test cases for boundary conditions are missing execution time assertions.",
+    "impact": "medium",
+    "rationale": "Ensuring that boundary conditions are tested for performance can help identify potential performance bottlenecks.",
+    "recommendation": "Add assertions for execution time in boundary condition tests.",
+    "current_code": "print(f\"Execution time:{execution_time}seconds\")",
+    "suggested_code": "assert execution_time < 1, f\"Execution time exceeded:{execution_time}seconds\"",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
+    "start_line": 61,
+    "end_line": 61,
+    "change_type": "modification",
+    "sentiment": "positive",
+    "severity": 5
+  },
+  {
+    "category": "Code Maintainability",
+    "description": "Repeated code for mocking os.path.exists and os.makedirs.",
+    "impact": "medium",
+    "rationale": "Repeated code can lead to maintenance challenges and increase the risk of bugs. Using fixtures can help reduce redundancy.",
+    "recommendation": "Use pytest fixtures to mock os.path.exists and os.makedirs.",
+    "current_code": "@mock.patch(\"kaizen.helpers.output.os.makedirs\")\n@mock.patch(\"kaizen.helpers.output.os.path.exists\")",
+    "suggested_code": "@pytest.fixture\ndef mock_os_path_exists():\n    with mock.patch('os.path.exists') as mock_exists:\n        yield mock_exists\n\n@pytest.fixture\ndef mock_os_makedirs():\n    with mock.patch('os.makedirs') as mock_makedirs:\n        yield mock_makedirs",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_folder.py",
+    "start_line": 9,
+    "end_line": 17,
+    "change_type": "modification",
+    "sentiment": "positive",
+    "severity": 6
+  },
+  {
+    "category": "Code Readability",
+    "description": "Inconsistent string formatting style.",
+    "impact": "medium",
+    "rationale": "The code uses both triple quotes and parentheses for multi-line strings, which can be confusing and reduce readability.",
+    "recommendation": "Use a consistent string formatting style throughout the code. Prefer triple quotes for multi-line strings for better readability.",
+    "current_code": "PR_COLLAPSIBLE_TEMPLATE = (\n    \"<details>\\n\"\n    \"<summary>Review</summary>\\n\"\n    \"<p>\\n\"\n    \"<b>Comment:</b>{comment}<br>\\n\"\n    \"<b>Reason:</b>{reason}<br>\\n\"\n    \"<b>Solution:</b>{solution}<br>\\n\"\n    \"<b>Confidence:</b>{confidence}<br>\\n\"\n    \"<b>Start Line:</b>{start_line}<br>\\n\"\n    \"<b>End Line:</b>{end_line}<br>\\n\"\n    \"<b>File Name:</b>{file_name}<br>\\n\"\n    \"<b>Severity:</b>{severity}<br>\\n\"\n    \"</p>\\n\"\n    \"</details>\"\n)",
+    "suggested_code": "PR_COLLAPSIBLE_TEMPLATE = \"\"\"\n<details>\n<summary>Review</summary>\n<p>{comment}</p>\n<p><strong>Reason:</strong>{reason}</p>\n<p><strong>Solution:</strong>{solution}</p>\n<p><strong>Confidence:</strong>{confidence}</p>\n<p><strong>Start Line:</strong>{start_line}</p>\n<p><strong>End Line:</strong>{end_line}</p>\n<p><strong>File Name:</strong>{file_name}</p>\n<p><strong>Severity:</strong>{severity}</p>\n</details>\n\"\"\"",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 4,
+    "end_line": 16,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Test Coverage",
+    "description": "Missing test for edge cases.",
+    "impact": "high",
+    "rationale": "The tests do not cover edge cases such as empty strings or null values for the review fields.",
+    "recommendation": "Add tests to cover edge cases like empty strings or null values for the review fields to ensure robustness.",
+    "current_code": "",
+    "suggested_code": "@pytest.fixture\n def setup_edge_case_reviews():\n     return{\n         \"topic1\":[\n{\n                 \"comment\": \"\",\n                 \"reason\": None,\n                 \"solution\": \"\",\n                 \"confidence\": \"\",\n                 \"start_line\": 0,\n                 \"end_line\": 0,\n                 \"file_name\": \"\",\n                 \"severity_level\": 0,\n}\n         ]\n}\n\n def test_edge_case_reviews(setup_edge_case_reviews):\n     topics = setup_edge_case_reviews\n     expected_output = \"## Code Review\\n\\n\u2757 **Attention Required:** This PR has potential issues. \ud83d\udea8\\n\\n### topic1\\n\\n\" + PR_COLLAPSIBLE_TEMPLATE.format(\n         comment=\"\",\n         reason=\"NA\",\n         solution=\"\",\n         confidence=\"NA\",\n         start_line=0,\n         end_line=0,\n         file_name=\"\",\n         severity=0,\n     ) + \"\\n\"\n     assert create_pr_review_text(topics) == expected_output",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 271,
+    "end_line": 276,
+    "change_type": "addition",
+    "sentiment": "positive",
+    "severity": 7
+  },
+  {
+    "category": "Error Handling",
+    "description": "Lack of error handling for missing fields in review data.",
+    "impact": "high",
+    "rationale": "If any field in the review data is missing, it could cause runtime errors or incorrect output.",
+    "recommendation": "Add error handling to check for missing fields in the review data and provide default values if necessary.",
+    "current_code": "",
+    "suggested_code": "def create_pr_review_text(topics):\n     output = \"## Code Review\\n\\n\"\n     if not topics:\n         return output + \"\u2705 **All Clear:** This PR is ready to merge! \ud83d\udc4d\\n\\n\"\n\n     output += \"\u2757 **Attention Required:** This PR has potential issues. \ud83d\udea8\\n\\n\"\n     for topic, reviews in topics.items():\n         output += f\"###{topic}\\n\\n\"\n         for review in reviews:\n             comment = review.get(\"comment\", \"NA\")\n             reason = review.get(\"reason\", \"NA\")\n             solution = review.get(\"solution\", \"NA\")\n             confidence = review.get(\"confidence\", \"NA\")\n             start_line = review.get(\"start_line\", 0)\n             end_line = review.get(\"end_line\", 0)\n             file_name = review.get(\"file_name\", \"\")\n             severity = review.get(\"severity_level\", 0)\n             output += PR_COLLAPSIBLE_TEMPLATE.format(\n                 comment=comment,\n                 reason=reason,\n                 solution=solution,\n                 confidence=confidence,\n                 start_line=start_line,\n                 end_line=end_line,\n                 file_name=file_name,\n                 severity=severity,\n             ) + \"\\n\"\n     return output",
+    "file_path": "kaizen/helpers/output.py",
+    "start_line": 1,
+    "end_line": 20,
+    "change_type": "modification",
+    "sentiment": "positive",
+    "severity": 8
+  },
+  {
+    "category": "Best Practices",
+    "description": "Use of hardcoded values for folder paths and module names.",
+    "impact": "medium",
+    "rationale": "Hardcoding values can make the code less flexible and harder to maintain. Using constants or configuration files can improve maintainability.",
+    "recommendation": "Define constants for folder paths and module names, or use a configuration file to manage these values.",
+    "current_code": "folder_path = \"test_folder\"",
+    "suggested_code": "folder_path = CONFIG['TEST_FOLDER_PATH']",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 27,
+    "end_line": 27,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling",
+    "description": "Lack of error handling for file operations.",
+    "impact": "high",
+    "rationale": "File operations can fail due to various reasons such as permission issues or missing files. Proper error handling ensures the program can handle such situations gracefully.",
+    "recommendation": "Add try-except blocks around file operations to handle potential errors.",
+    "current_code": "with open(file_path, 'r') as f:\n    return f.read()",
+    "suggested_code": "try:\n    with open(file_path, 'r') as f:\n        return f.read()\nexcept IOError as e:\n    logger.error(f\"Error reading file{file_path}:{e}\")\n    return None",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 17,
+    "end_line": 18,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 7
+  },
+  {
+    "category": "Code Readability",
+    "description": "Long lines of code that reduce readability.",
+    "impact": "low",
+    "rationale": "Long lines can make the code harder to read and understand. Breaking them into multiple lines can improve readability.",
+    "recommendation": "Break long lines into multiple lines for better readability.",
+    "current_code": "mock.patch('kaizen.helpers.output.create_folder') as mock_create_folder, mock.patch('kaizen.helpers.output.general.clean_python_code') as mock_clean_python_code, mock.patch('kaizen.helpers.output.logger') as mock_logger:",
+    "suggested_code": "mock.patch('kaizen.helpers.output.create_folder') as mock_create_folder, \nmock.patch('kaizen.helpers.output.general.clean_python_code') as mock_clean_python_code, \nmock.patch('kaizen.helpers.output.logger') as mock_logger:",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 10,
+    "end_line": 12,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 3
+  },
+  {
+    "category": "Security",
+    "description": "Potential security vulnerability with file names containing special characters.",
+    "impact": "high",
+    "rationale": "Special characters in file names can lead to security vulnerabilities such as path traversal attacks. Sanitizing file names can mitigate this risk.",
+    "recommendation": "Ensure that file names are sanitized to remove special characters.",
+    "current_code": "def sanitize_filename(filename):\n    return \"\".join(c if c.isalnum() or c in (' ', '.', '_') else '_' for c in filename)",
+    "suggested_code": "def sanitize_filename(filename):\n    return \"\".join(c if c.isalnum() or c in (' ', '.', '_') else '_' for c in filename).replace(' ', '_')",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 21,
+    "end_line": 22,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 8
+  },
+  {
+    "category": "Testing",
+    "description": "Missing test cases for edge scenarios.",
+    "impact": "medium",
+    "rationale": "Edge scenarios such as empty input or invalid data should be tested to ensure the robustness of the code.",
+    "recommendation": "Add test cases for edge scenarios such as empty input or invalid data.",
+    "current_code": "",
+    "suggested_code": "def test_empty_input(tmp_path, mock_dependencies):\n    mock_create_folder, mock_clean_python_code, mock_logger = mock_dependencies\n    json_tests =[]\n    create_test_files(json_tests, tmp_path)\n    assert os.path.exists(os.path.join(tmp_path, \"tests.json\"))\n    assert os.path.getsize(os.path.join(tmp_path, \"tests.json\")) == 0",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 92,
+    "end_line": 102,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "error_handling",
+    "description": "Inconsistent error messages in test_get_parent_folder_error_handling",
+    "impact": "medium",
+    "rationale": "The error messages in the mock.patch side_effect and pytest.raises do not match, which can lead to confusion and potential test failures.",
+    "recommendation": "Ensure that the error messages in mock.patch side_effect and pytest.raises match exactly.",
+    "current_code": "with mock.patch('os.getcwd', side_effect=OSError(\"Unable to determine current working directory\")):\n    with pytest.raises(OSError, match=\"Unable to determine current working directory\"):\n        get_parent_folder()",
+    "suggested_code": "with mock.patch('os.getcwd', side_effect=OSError(\"Failed to get current working directory\")):\n    with pytest.raises(OSError, match=\"Failed to get current working directory\"):\n        get_parent_folder()",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py",
+    "start_line": 20,
+    "end_line": 22,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "performance",
+    "description": "Redundant import of nest_asyncio",
+    "impact": "low",
+    "rationale": "The import of nest_asyncio is redundant as it is already being patched in the test cases. This can lead to unnecessary imports and slightly increased memory usage.",
+    "recommendation": "Remove the redundant import of nest_asyncio.",
+    "current_code": "import nest_asyncio",
+    "suggested_code": "",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 5,
+    "end_line": 5,
+    "change_type": "deletion",
+    "sentiment": "neutral",
+    "severity": 2
+  },
+  {
+    "category": "readability",
+    "description": "Improper indentation in test_get_web_html_large_content",
+    "impact": "low",
+    "rationale": "Improper indentation can make the code harder to read and maintain.",
+    "recommendation": "Ensure proper indentation for better readability.",
+    "current_code": "expected_output = \"<html>\\n <body>\\n\" + \"  <p>\\n   Test\\n  </p>\\n\" * 10000 + \" </body>\\n</html>\"",
+    "suggested_code": "expected_output = (\n    \"<html>\\n <body>\\n\" + \"  <p>\\n   Test\\n  </p>\\n\" * 10000 + \" </body>\\n</html>\"\n)",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 95,
+    "end_line": 95,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 3
+  },
+  {
+    "category": "best_practices",
+    "description": "Hardcoded API key in config.json",
+    "impact": "high",
+    "rationale": "Hardcoding sensitive information like API keys can lead to security vulnerabilities.",
+    "recommendation": "Use environment variables to store sensitive information.",
+    "current_code": "\"api_key\": \"os.environ/AZURE_API_KEY\"",
+    "suggested_code": "\"api_key\": os.getenv('AZURE_API_KEY')",
+    "file_path": "config.json",
+    "start_line": 13,
+    "end_line": 13,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "maintainability",
+    "description": "Unnecessary comments in main.py",
+    "impact": "trivial",
+    "rationale": "Comments that do not add value or are redundant can clutter the code and make it harder to maintain.",
+    "recommendation": "Remove unnecessary comments to improve code clarity.",
+    "current_code": "# generator.generate_tests(file_path=\"sample.py\", content=code)  # Replace with the actual file path",
+    "suggested_code": "",
+    "file_path": "examples/unittest/main.py",
+    "start_line": 34,
+    "end_line": 34,
+    "change_type": "deletion",
+    "sentiment": "neutral",
+    "severity": 1
+  },
+  {
+    "category": "Import Management",
+    "description": "Redundant import of `tqdm` removed and re-added.",
+    "impact": "low",
+    "rationale": "The import of `tqdm` was removed and then re-added later, which is unnecessary and can cause confusion.",
+    "recommendation": "Remove the redundant import statement.",
+    "current_code": "from tqdm import tqdm",
+    "suggested_code": "",
+    "file_path": "kaizen/generator/unit_test.py",
+    "start_line": 14,
+    "end_line": 14,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 2
+  },
+  {
+    "category": "Code Duplication",
+    "description": "Duplicated dictionary for supported languages.",
+    "impact": "medium",
+    "rationale": "The `SUPPORTED_LANGUAGES` dictionary was duplicated, which can lead to maintenance issues if the list needs to be updated.",
+    "recommendation": "Remove the duplicated dictionary and ensure it's defined in one place.",
+    "current_code": "SUPPORTED_LANGUAGES ={\"py\": \"PythonParser\", \"js\": \"JavaScriptParser\", \"ts\": \"TypeScriptParser\", \"jsx\": \"ReactParser\", \"tsx\": \"ReactTSParser\", \"rs\": \"RustParser\"}",
+    "suggested_code": "",
+    "file_path": "kaizen/generator/unit_test.py",
+    "start_line": 24,
+    "end_line": 31,
+    "change_type": "addition",
+    "sentiment": "negative",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling",
+    "description": "Lack of error handling in file operations.",
+    "impact": "high",
+    "rationale": "File operations such as reading and writing files should include error handling to manage potential I/O errors.",
+    "recommendation": "Add try-except blocks around file operations to handle potential I/O errors.",
+    "current_code": "with open(file_path, \"r\") as file: return file.read()",
+    "suggested_code": "try: with open(file_path, \"r\") as file: return file.read() except IOError as e: self.logger.error(f\"Failed to read file{file_path}:{e}\") return None",
+    "file_path": "kaizen/generator/unit_test.py",
+    "start_line": 109,
+    "end_line": 110,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 7
+  },
+  {
+    "category": "Logging",
+    "description": "Logging should be consistent and meaningful.",
+    "impact": "medium",
+    "rationale": "Logging messages should provide meaningful information and be consistent throughout the codebase.",
+    "recommendation": "Ensure that all logging messages are consistent and provide useful information.",
+    "current_code": "print(f\"  \u2713 File will be saved as:{test_file_path}\")",
+    "suggested_code": "self.logger.info(f\"  \u2713 File will be saved as:{test_file_path}\")",
+    "file_path": "kaizen/generator/unit_test.py",
+    "start_line": 140,
+    "end_line": 140,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 4
+  },
+  {
+    "category": "Code Structure",
+    "description": "Refactor long methods into smaller, more manageable methods.",
+    "impact": "high",
+    "rationale": "Long methods can be difficult to read and maintain. Refactoring them into smaller methods improves readability and maintainability.",
+    "recommendation": "Refactor long methods into smaller, more manageable methods.",
+    "current_code": "def generate_test_files(self, parsed_data, file_extension, file_path): ...",
+    "suggested_code": "def generate_test_files(self, parsed_data, file_extension, file_path): for item in tqdm(parsed_data, desc=\"Processing Items\", unit=\"item\"): self._process_item(item, file_extension, file_path)",
+    "file_path": "kaizen/generator/unit_test.py",
+    "start_line": 112,
+    "end_line": 117,
+    "change_type": "addition",
+    "sentiment": "positive",
+    "severity": 6
+  },
+  {
+    "category": "Performance",
+    "description": "Repeated calls to `os.makedirs`.",
+    "impact": "medium",
+    "rationale": "Repeatedly calling `os.makedirs` can be inefficient. It's better to ensure the directory exists once.",
+    "recommendation": "Ensure the directory exists once at the beginning of the process.",
+    "current_code": "os.makedirs(self.log_dir, exist_ok=True)",
+    "suggested_code": "self._setup_directories()",
+    "file_path": "kaizen/generator/unit_test.py",
+    "start_line": 58,
+    "end_line": 58,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "error_handling",
+    "description": "Potential issue with setting log levels",
+    "impact": "high",
+    "rationale": "The function `set_all_loggers_to_ERROR` sets all loggers to ERROR level, which might override specific loggers' configurations that are necessary for debugging or monitoring. This could lead to a loss of important log information.",
+    "recommendation": "Add a parameter to the function to allow excluding certain loggers from being set to ERROR level.",
+    "current_code": "def set_all_loggers_to_ERROR():\n    print(\"All Loggers and their levels:\")\n    for name, logger in logging.Logger.manager.loggerDict.items():\n        if isinstance(logger, logging.Logger):\n            print(f\"Logger:{name}, Level:{logging.getLevelName(logger.level)}\")\n            logging.getLogger(name).setLevel(logging.ERROR)\n        else:\n            print(f\"PlaceHolder:{name}\")",
+    "suggested_code": "def set_all_loggers_to_ERROR(exclude_loggers=None):\n    if exclude_loggers is None:\n        exclude_loggers =[]\n    print(\"All Loggers and their levels:\")\n    for name, logger in logging.Logger.manager.loggerDict.items():\n        if isinstance(logger, logging.Logger) and name not in exclude_loggers:\n            print(f\"Logger:{name}, Level:{logging.getLevelName(logger.level)}\")\n            logging.getLogger(name).setLevel(logging.ERROR)\n        else:\n            print(f\"PlaceHolder:{name}\")",
+    "file_path": "kaizen/llms/provider.py",
+    "start_line": 13,
+    "end_line": 21,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 7
+  },
+  {
+    "category": "performance",
+    "description": "Unnecessary repeated calls to `logging.getLogger`",
+    "impact": "medium",
+    "rationale": "Calling `logging.getLogger(name)` multiple times for the same logger is redundant and can be optimized.",
+    "recommendation": "Store the logger instance in a variable and reuse it.",
+    "current_code": "for name, logger in logging.Logger.manager.loggerDict.items():\n    if isinstance(logger, logging.Logger):\n        print(f\"Logger:{name}, Level:{logging.getLevelName(logger.level)}\")\n        logging.getLogger(name).setLevel(logging.ERROR)",
+    "suggested_code": "for name, logger in logging.Logger.manager.loggerDict.items():\n    if isinstance(logger, logging.Logger):\n        print(f\"Logger:{name}, Level:{logging.getLevelName(logger.level)}\")\n        logger.setLevel(logging.ERROR)",
+    "file_path": "kaizen/llms/provider.py",
+    "start_line": 15,
+    "end_line": 18,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "readability",
+    "description": "Inline comments for setting specific loggers to ERROR",
+    "impact": "low",
+    "rationale": "The inline comments provide context but could be more descriptive for maintainability.",
+    "recommendation": "Expand the comments to explain why these specific loggers are being set to ERROR level.",
+    "current_code": "# Set litellm log level to ERROR\nlogging.getLogger(\"LiteLLM\").setLevel(logging.ERROR)\nlogging.getLogger(\"LiteLLM Router\").setLevel(logging.ERROR)\nlogging.getLogger(\"LiteLLM Proxy\").setLevel(logging.ERROR)",
+    "suggested_code": "# Set specific LiteLLM loggers to ERROR level to minimize log noise and focus on critical issues\nlogging.getLogger(\"LiteLLM\").setLevel(logging.ERROR)\nlogging.getLogger(\"LiteLLM Router\").setLevel(logging.ERROR)\nlogging.getLogger(\"LiteLLM Proxy\").setLevel(logging.ERROR)",
+    "file_path": "kaizen/llms/provider.py",
+    "start_line": 25,
+    "end_line": 27,
+    "change_type": "modification",
+    "sentiment": "positive",
+    "severity": 3
+  },
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "11",
+    "end_line": "11",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/review.md b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/review.md
new file mode 100644
index 00000000..0644c0bf
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/review.md
@@ -0,0 +1,64 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/400
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 28
+- Critical: 1
+- Important: 0
+- Minor: 0
+- Files Affected: 11
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Configuration (1 issues)</strong></summary>
+
+### 1. Changes made to sensitive file
+📁 **File:** `config.json:11`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (7 issues)</strong></summary>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-2024-05-13)
+{"prompt_tokens": 38388, "completion_tokens": 6513, "total_tokens": 44901}
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_335/comments.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/comments.json
similarity index 100%
rename from .experiments/code_review/haiku/no_eval/pr_335/comments.json
rename to .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/comments.json
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/issues.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/issues.json
new file mode 100644
index 00000000..162fbd69
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/issues.json
@@ -0,0 +1,47 @@
+[
+  {
+    "category": "Code Clarity",
+    "description": "Uncommented setup_repository call",
+    "impact": "medium",
+    "rationale": "Uncommenting the setup_repository call without context or explanation can lead to confusion for future maintainers. It's important to provide context or comments explaining why this line is uncommented.",
+    "recommendation": "Add a comment explaining why the setup_repository call is uncommented and when it should be used.",
+    "current_code": "analyzer.setup_repository(\"./github_app/\")",
+    "suggested_code": "# Uncommented for initial setup or updates\nanalyzer.setup_repository(\"./github_app/\")",
+    "file_path": "examples/ragify_codebase/main.py",
+    "start_line": 7,
+    "end_line": 7,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 4
+  },
+  {
+    "category": "Code Commenting",
+    "description": "Incomplete TODO comment",
+    "impact": "low",
+    "rationale": "The TODO comment 'DONT PUSH DUPLICATE' is too vague and does not provide enough information on what needs to be done. This can lead to confusion and potential errors in the future.",
+    "recommendation": "Provide a more detailed comment explaining what needs to be done to avoid pushing duplicates.",
+    "current_code": "# TODO: DONT PUSH DUPLICATE",
+    "suggested_code": "# TODO: Implement logic to check for and avoid pushing duplicate embeddings",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 157,
+    "end_line": 157,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 3
+  },
+  {
+    "category": "Dependency Management",
+    "description": "Updated dependency versions",
+    "impact": "medium",
+    "rationale": "Updating dependency versions without proper testing can introduce new bugs or incompatibilities. It's important to ensure that all tests pass and that the new versions are compatible with the rest of the codebase.",
+    "recommendation": "Run all tests and perform thorough testing to ensure that the updated dependencies do not introduce any issues.",
+    "current_code": "llama-index-core = \"0.10.65\"",
+    "suggested_code": "",
+    "file_path": "pyproject.toml",
+    "start_line": 27,
+    "end_line": 27,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/review.md b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/review.md
new file mode 100644
index 00000000..e9f81733
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/review.md
@@ -0,0 +1,41 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/440
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 3
+- Critical: 0
+- Important: 0
+- Minor: 0
+- Files Affected: 3
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (1 issues)</strong></summary>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-2024-05-13)
+{"prompt_tokens": 1613, "completion_tokens": 581, "total_tokens": 2194}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/comments.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/comments.json
new file mode 100644
index 00000000..9cbf8d49
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/comments.json
@@ -0,0 +1,16 @@
+[
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/issues.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/issues.json
new file mode 100644
index 00000000..b84036f9
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/issues.json
@@ -0,0 +1,121 @@
+[
+  {
+    "category": "Error Handling",
+    "description": "Broad exception handling",
+    "impact": "high",
+    "rationale": "Using a broad `except Exception` can mask other errors and make debugging difficult. It is better to catch specific exceptions.",
+    "recommendation": "Replace `except Exception` with `except KeyError` to handle the specific error.",
+    "current_code": "except Exception:\n    print(\"Error\")",
+    "suggested_code": "except KeyError:\n    raise ValueError(f\"Invalid confidence level:{min_confidence}\")",
+    "file_path": "github_app/github_helper/pull_requests.py",
+    "start_line": 140,
+    "end_line": 141,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Code Quality",
+    "description": "Unnecessary print statements",
+    "impact": "medium",
+    "rationale": "Print statements can clutter the output and are not suitable for production code. Use logging instead.",
+    "recommendation": "Replace `print` statements with appropriate logging calls.",
+    "current_code": "print(\"issues: \", issues)",
+    "suggested_code": "logger.info(\"issues: %s\", issues)",
+    "file_path": "github_app/github_helper/pull_requests.py",
+    "start_line": 157,
+    "end_line": 157,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Code Quality",
+    "description": "Unnecessary print statements",
+    "impact": "medium",
+    "rationale": "Print statements can clutter the output and are not suitable for production code. Use logging instead.",
+    "recommendation": "Replace `print` statements with appropriate logging calls.",
+    "current_code": "print(\"diff: \", diff_text)\nprint(\"pr_files\", pr_files)",
+    "suggested_code": "logger.info(\"diff: %s\", diff_text)\nlogger.info(\"pr_files: %s\", pr_files)",
+    "file_path": "examples/code_review/main.py",
+    "start_line": 21,
+    "end_line": 22,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Code Quality",
+    "description": "Unnecessary print statements",
+    "impact": "medium",
+    "rationale": "Print statements can clutter the output and are not suitable for production code. Use logging instead.",
+    "recommendation": "Replace `print` statements with appropriate logging calls.",
+    "current_code": "print(f\"Raw Topics: \\n{json.dumps(review_data.issues, indent=2)}\\n\")",
+    "suggested_code": "logger.info(\"Raw Topics: \\n %s\\n\", json.dumps(review_data.issues, indent=2))",
+    "file_path": "examples/code_review/main.py",
+    "start_line": 39,
+    "end_line": 39,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Functionality",
+    "description": "Potential issue with test generation",
+    "impact": "medium",
+    "rationale": "The `generate_tests` function is added but not properly integrated. This could lead to unexpected behavior.",
+    "recommendation": "Ensure `generate_tests` is called correctly and its output is used appropriately.",
+    "current_code": "tests = generate_tests(pr_files)",
+    "suggested_code": "tests = generate_tests(sorted_files)",
+    "file_path": "github_app/github_helper/pull_requests.py",
+    "start_line": 58,
+    "end_line": 58,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Functionality",
+    "description": "Unnecessary return statement",
+    "impact": "low",
+    "rationale": "The return statement at the end of the function is unnecessary and can be removed.",
+    "recommendation": "Remove the return statement.",
+    "current_code": "return True, review_desc.decode(\"utf-8\")",
+    "suggested_code": "",
+    "file_path": "github_app/github_helper/pull_requests.py",
+    "start_line": 67,
+    "end_line": 67,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 3
+  },
+  {
+    "category": "Code Quality",
+    "description": "Unnecessary print statements",
+    "impact": "medium",
+    "rationale": "Print statements can clutter the output and are not suitable for production code. Use logging instead.",
+    "recommendation": "Replace `print` statements with appropriate logging calls.",
+    "current_code": "print(f\"Raw Topics: \\n{json.dumps(review_data.issues, indent=2)}\\n\")",
+    "suggested_code": "logger.info(\"Raw Topics: \\n %s\\n\", json.dumps(review_data.issues, indent=2))",
+    "file_path": "examples/code_review/main.py",
+    "start_line": 39,
+    "end_line": 39,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/review.md b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/review.md
new file mode 100644
index 00000000..71a2795c
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/review.md
@@ -0,0 +1,64 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/476
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 8
+- Critical: 1
+- Important: 0
+- Minor: 0
+- Files Affected: 3
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Configuration (1 issues)</strong></summary>
+
+### 1. Changes made to sensitive file
+📁 **File:** `config.json:1`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (1 issues)</strong></summary>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-2024-05-13)
+{"prompt_tokens": 4256, "completion_tokens": 1196, "total_tokens": 5452}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/comments.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/comments.json
new file mode 100644
index 00000000..d23463e6
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/comments.json
@@ -0,0 +1,17 @@
+[
+  {
+    "category": "Error Handling",
+    "description": "Division by zero potential if total_tokens is zero.",
+    "impact": "critical",
+    "rationale": "Division by zero will cause the program to crash. This needs to be handled to ensure robustness.",
+    "recommendation": "Add a check to handle the case when total_tokens is zero.",
+    "current_code": "print(f\"Total tokens used:{total_tokens:,}\")",
+    "suggested_code": "if total_tokens == 0:\n    print(\"Total tokens used: 0\")\nelse:\n    print(f\"Total tokens used:{total_tokens:,}\")",
+    "file_path": "main.py",
+    "start_line": 161,
+    "end_line": 161,
+    "change_type": "addition",
+    "sentiment": "positive",
+    "severity": 9
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/issues.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/issues.json
new file mode 100644
index 00000000..68ddc5bb
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/issues.json
@@ -0,0 +1,107 @@
+[
+  {
+    "category": "Unused Import",
+    "description": "The 'random' module is imported but not used anywhere in the code.",
+    "impact": "trivial",
+    "rationale": "Unused imports can clutter the code and may lead to confusion. They also slightly increase the memory footprint.",
+    "recommendation": "Remove the unused import statement.",
+    "current_code": "import random  # Unused import",
+    "suggested_code": "",
+    "file_path": "main.py",
+    "start_line": 8,
+    "end_line": 8,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 1
+  },
+  {
+    "category": "Error Handling",
+    "description": "Potential for API call to fail without retry mechanism.",
+    "impact": "high",
+    "rationale": "If the API call fails, the function will not handle the failure gracefully, which may lead to incomplete processing of applicant data.",
+    "recommendation": "Implement a retry mechanism with exponential backoff for the API call.",
+    "current_code": "response = completion(\n    model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n)",
+    "suggested_code": "import time\n\nfor attempt in range(3):\n    try:\n        response = completion(\n            model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n        )\n        break\n    except Exception as e:\n        if attempt < 2:\n            time.sleep(2 ** attempt)\n        else:\n            raise e",
+    "file_path": "main.py",
+    "start_line": 66,
+    "end_line": 68,
+    "change_type": "addition",
+    "sentiment": "positive",
+    "severity": 8
+  },
+  {
+    "category": "Error Handling",
+    "description": "Silent failure without logging in JSONDecodeError exception.",
+    "impact": "medium",
+    "rationale": "Silent failures can make debugging difficult. Logging the error provides visibility into what went wrong.",
+    "recommendation": "Add logging for the JSONDecodeError exception.",
+    "current_code": "except json.JSONDecodeError:\n    result ={key: \"\" for key in[\"feedback\", \"review\", \"should_interview\", \"rating\", \"input_tokens\", \"output_tokens\"]}",
+    "suggested_code": "except json.JSONDecodeError as e:\n    print(f\"Failed to parse content for applicant:{e}\")\n    result ={key: \"\" for key in[\"feedback\", \"review\", \"should_interview\", \"rating\", \"input_tokens\", \"output_tokens\"]}",
+    "file_path": "main.py",
+    "start_line": 82,
+    "end_line": 94,
+    "change_type": "addition",
+    "sentiment": "positive",
+    "severity": 6
+  },
+  {
+    "category": "Performance",
+    "description": "Inefficient way to print progress.",
+    "impact": "low",
+    "rationale": "Printing progress in this manner can be inefficient and slow down the processing, especially for large datasets.",
+    "recommendation": "Consider using a more efficient method for progress reporting, such as updating the progress bar less frequently.",
+    "current_code": "print(f\"\\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}\", end=\"\", flush=True)",
+    "suggested_code": "if index % 10 == 0 or index == total - 1:\n    print(f\"\\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}\", end=\"\", flush=True)",
+    "file_path": "main.py",
+    "start_line": 121,
+    "end_line": 121,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 3
+  },
+  {
+    "category": "Redundant Code",
+    "description": "The check for empty DataFrame is unnecessary.",
+    "impact": "trivial",
+    "rationale": "The DataFrame length check before processing is redundant as the subsequent code will handle an empty DataFrame gracefully.",
+    "recommendation": "Remove the redundant check for an empty DataFrame.",
+    "current_code": "if len(df) == 0:\n    return",
+    "suggested_code": "",
+    "file_path": "main.py",
+    "start_line": 141,
+    "end_line": 143,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 1
+  },
+  {
+    "category": "Error Handling",
+    "description": "No error handling for file not found.",
+    "impact": "high",
+    "rationale": "If the file does not exist, the program will crash without providing a meaningful error message.",
+    "recommendation": "Add error handling for file not found.",
+    "current_code": "main(input_file)",
+    "suggested_code": "if not os.path.exists(input_file):\n    print(f\"File not found:{input_file}\")\nelse:\n    main(input_file)",
+    "file_path": "main.py",
+    "start_line": 174,
+    "end_line": 175,
+    "change_type": "addition",
+    "sentiment": "positive",
+    "severity": 7
+  },
+  {
+    "category": "Error Handling",
+    "description": "Division by zero potential if total_tokens is zero.",
+    "impact": "critical",
+    "rationale": "Division by zero will cause the program to crash. This needs to be handled to ensure robustness.",
+    "recommendation": "Add a check to handle the case when total_tokens is zero.",
+    "current_code": "print(f\"Total tokens used:{total_tokens:,}\")",
+    "suggested_code": "if total_tokens == 0:\n    print(\"Total tokens used: 0\")\nelse:\n    print(f\"Total tokens used:{total_tokens:,}\")",
+    "file_path": "main.py",
+    "start_line": 161,
+    "end_line": 161,
+    "change_type": "addition",
+    "sentiment": "positive",
+    "severity": 9
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/review.md b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/review.md
new file mode 100644
index 00000000..eee73c57
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/review.md
@@ -0,0 +1,64 @@
+PR URL: https://github.com/sauravpanda/applicant-screening/pull/5
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 7
+- Critical: 1
+- Important: 0
+- Minor: 0
+- Files Affected: 1
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Error Handling (1 issues)</strong></summary>
+
+### 1. Division by zero potential if total_tokens is zero.
+📁 **File:** `main.py:161`
+⚖️ **Severity:** 9/10
+🔍 **Description:** Division by zero potential if total_tokens is zero.
+💡 **Solution:** 
+
+**Current Code:**
+```python
+print(f"Total tokens used:{total_tokens:,}")
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (3 issues)</strong></summary>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-2024-05-13)
+{"prompt_tokens": 6403, "completion_tokens": 1420, "total_tokens": 7823}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/comments.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/comments.json
new file mode 100644
index 00000000..ef33088e
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/comments.json
@@ -0,0 +1,87 @@
+[
+  {
+    "category": "Security",
+    "description": "Hardcoded API keys in config.json.",
+    "impact": "critical",
+    "rationale": "Storing sensitive information like API keys in the source code can lead to security vulnerabilities if the code is exposed.",
+    "recommendation": "Use environment variables or a secure vault service to manage sensitive information instead of hardcoding them in the configuration file.",
+    "current_code": "\"api_key\": \"os.environ/AZURE_API_KEY\"",
+    "suggested_code": "\"api_key\": \"${AZURE_API_KEY}\"",
+    "file_path": "config.json",
+    "start_line": 13,
+    "end_line": 13,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 9
+  },
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Docker",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to Dockerfile, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_path": "Dockerfile",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Docker",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to docker-compose.yml, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "15",
+    "end_line": "15",
+    "side": "RIGHT",
+    "file_path": "docker-compose.yml",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Version Control",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to .gitignore, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "164",
+    "end_line": "164",
+    "side": "RIGHT",
+    "file_path": ".gitignore",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Database",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to db_setup/init.sql, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_path": "db_setup/init.sql",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/issues.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/issues.json
new file mode 100644
index 00000000..9ce8a512
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/issues.json
@@ -0,0 +1,342 @@
+[
+  {
+    "category": "Documentation",
+    "description": "Lack of comments in critical sections.",
+    "impact": "high",
+    "rationale": "While the code is mostly self-explanatory, certain sections, especially in the Dockerfile and SQL scripts, would benefit from additional comments to clarify their purpose and usage.",
+    "recommendation": "Add comments to explain the purpose of each section, especially in the Dockerfile and SQL setup scripts.",
+    "current_code": "RUN apt-get update && apt-get install -y \\\n    git \\\n    build-essential \\\n    && rm -rf /var/lib/apt/lists/*",
+    "suggested_code": "# Install system dependencies\nRUN apt-get update && apt-get install -y \\\n    git \\\n    build-essential \\\n    && rm -rf /var/lib/apt/lists/*",
+    "file_path": "Dockerfile",
+    "start_line": 7,
+    "end_line": 12,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Security",
+    "description": "Hardcoded API keys in config.json.",
+    "impact": "critical",
+    "rationale": "Storing sensitive information like API keys in the source code can lead to security vulnerabilities if the code is exposed.",
+    "recommendation": "Use environment variables or a secure vault service to manage sensitive information instead of hardcoding them in the configuration file.",
+    "current_code": "\"api_key\": \"os.environ/AZURE_API_KEY\"",
+    "suggested_code": "\"api_key\": \"${AZURE_API_KEY}\"",
+    "file_path": "config.json",
+    "start_line": 13,
+    "end_line": 13,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 9
+  },
+  {
+    "category": "Error Handling",
+    "description": "Missing error handling in the install script.",
+    "impact": "medium",
+    "rationale": "The install_tree_sitter_languages.sh script lacks error handling, which could lead to silent failures if a language fails to install.",
+    "recommendation": "Add error handling to the installation process to ensure that failures are logged and handled appropriately.",
+    "current_code": "git clone \"https://github.com/tree-sitter/tree-sitter-$lang\" \"$LANGUAGE_DIR/tree-sitter-$lang\"",
+    "suggested_code": "git clone \"https://github.com/tree-sitter/tree-sitter-$lang\" \"$LANGUAGE_DIR/tree-sitter-$lang\" ||{echo \"Failed to clone $lang\"; exit 1;}",
+    "file_path": "install_tree_sitter_languages.sh",
+    "start_line": 24,
+    "end_line": 24,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 7
+  },
+  {
+    "category": "Performance",
+    "description": "Potential inefficiency in the chunk_code function.",
+    "impact": "medium",
+    "rationale": "The function processes nodes recursively, which could lead to performance issues with larger codebases. Consider optimizing the traversal method.",
+    "recommendation": "Implement an iterative approach or optimize the recursive function to handle larger codebases more efficiently.",
+    "current_code": "def process_node(node):\n    result = parse_code(code, language)",
+    "suggested_code": "def process_node(node):\n    result = parse_code(code, language)\n    if not result:\n        return",
+    "file_path": "kaizen/retriever/code_chunker.py",
+    "start_line": 20,
+    "end_line": 22,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Readability",
+    "description": "Lack of docstrings for classes and methods.",
+    "impact": "medium",
+    "rationale": "Docstrings provide essential context for understanding the purpose and usage of classes and methods, which enhances maintainability and usability.",
+    "recommendation": "Add docstrings to all classes and methods to describe their functionality and parameters.",
+    "current_code": "",
+    "suggested_code": "\"\"\"\nClass to handle feedback abstraction.\n\nAttributes:\n    feedback_store (Dict[str, Dict[str, Any]]): A dictionary to store feedback.\n\"\"\"",
+    "file_path": "kaizen/retriever/feedback_system.py",
+    "start_line": 4,
+    "end_line": 5,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling",
+    "description": "Potential lack of error handling in database operations.",
+    "impact": "high",
+    "rationale": "Database operations can fail for various reasons (e.g., connection issues, SQL errors). Not handling these exceptions can lead to crashes or unhandled errors in production.",
+    "recommendation": "Wrap database operations in try-except blocks to handle potential exceptions gracefully.",
+    "current_code": "cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))",
+    "suggested_code": "try:\n    cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))\nexcept Exception as e:\n    # Handle exception (log it, re-raise, etc.)\n    raise RuntimeError(\"Database query failed\") from e",
+    "file_path": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 41,
+    "end_line": 41,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Performance",
+    "description": "Normalization of query embedding could be optimized.",
+    "impact": "medium",
+    "rationale": "Using numpy for normalization is efficient, but additional checks can be added to handle edge cases (e.g., zero-length vectors) to avoid potential runtime errors.",
+    "recommendation": "Add a check to ensure the query embedding is not zero-length before normalization.",
+    "current_code": "query_embedding_normalized = query_embedding_np / np.linalg.norm(query_embedding_np)",
+    "suggested_code": "norm = np.linalg.norm(query_embedding_np)\nif norm == 0:\n    raise ValueError(\"Query embedding cannot be zero-length\")\nquery_embedding_normalized = query_embedding_np / norm",
+    "file_path": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 16,
+    "end_line": 16,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Error Handling",
+    "description": "Lack of specific error handling in critical sections.",
+    "impact": "high",
+    "rationale": "While there are general exception handlers in place, specific exceptions should be caught to provide more meaningful error messages and to handle different failure scenarios appropriately.",
+    "recommendation": "Implement more granular exception handling to catch specific exceptions where they may occur, such as FileNotFoundError or SQLAlchemy specific exceptions.",
+    "current_code": "except Exception as e:",
+    "suggested_code": "except FileNotFoundError as e:\n    logger.error(f'File not found:{file_path}')\nexcept SQLAlchemyError as e:\n    logger.error(f'Database error:{str(e)}')",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 108,
+    "end_line": 110,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 7
+  },
+  {
+    "category": "Logging",
+    "description": "Inconsistent logging levels used.",
+    "impact": "medium",
+    "rationale": "The code uses different logging levels (INFO, DEBUG, ERROR) inconsistently. This can lead to confusion regarding the importance of messages and may cause critical issues to be overlooked.",
+    "recommendation": "Standardize logging levels across the codebase to ensure that critical information is logged at the appropriate level.",
+    "current_code": "logger.debug(f'Parsing file:{file_path}')",
+    "suggested_code": "logger.info(f'Parsing file:{file_path}')",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 90,
+    "end_line": 90,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Code Readability",
+    "description": "Long methods with multiple responsibilities.",
+    "impact": "high",
+    "rationale": "The methods are quite long and handle multiple responsibilities, which makes them harder to read and maintain. This violates the Single Responsibility Principle.",
+    "recommendation": "Refactor long methods into smaller, more focused methods to improve readability and maintainability.",
+    "current_code": "def parse_file(self, file_path: str):\n    ...",
+    "suggested_code": "def parse_file(self, file_path: str):\n    self.read_file(file_path)\n    self.process_file_content(content)\n    ...",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 89,
+    "end_line": 110,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Performance",
+    "description": "Potential inefficiency in database operations.",
+    "impact": "high",
+    "rationale": "The code performs multiple database operations in a loop which can lead to performance issues, especially with a large number of records.",
+    "recommendation": "Batch database operations where possible to reduce the number of transactions and improve performance.",
+    "current_code": "connection.execute(query,{...})",
+    "suggested_code": "with connection.begin():\n    connection.execute(batch_query, batch_data)",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 298,
+    "end_line": 312,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 8
+  },
+  {
+    "category": "Documentation",
+    "description": "Insufficient comments and documentation.",
+    "impact": "medium",
+    "rationale": "While the code is largely self-explanatory, additional comments and documentation would enhance understanding, especially for complex methods.",
+    "recommendation": "Add docstrings to all methods and inline comments where necessary to explain complex logic.",
+    "current_code": "def store_code_in_db(self, code: str, abstraction: str, file_path: str, section: str, name: str):",
+    "suggested_code": "def store_code_in_db(self, code: str, abstraction: str, file_path: str, section: str, name: str):\n    \"\"\"\n    Stores the provided code and its abstraction in the database.\n    Args:\n        code (str): The code to store.\n        abstraction (str): The abstract description of the code.\n        file_path (str): The path of the file containing the code.\n        section (str): The section of the code (e.g., functions).\n        name (str): The name of the function or code block.\n    \"\"\"",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 246,
+    "end_line": 248,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 4
+  },
+  {
+    "category": "Error Handling",
+    "description": "General exception handling in language loading and parsing.",
+    "impact": "high",
+    "rationale": "Using a broad exception catch can obscure specific errors, making debugging difficult. It's better to catch specific exceptions where possible.",
+    "recommendation": "Refine exception handling to catch specific exceptions instead of using a generic Exception.",
+    "current_code": "except Exception as e:",
+    "suggested_code": "except (ImportError, ValueError) as e:",
+    "file_path": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 28,
+    "end_line": 30,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 7
+  },
+  {
+    "category": "Logging",
+    "description": "Inconsistent logging levels for errors and warnings.",
+    "impact": "medium",
+    "rationale": "Using different logging levels can lead to confusion about the severity of issues. It's important to use logging levels consistently.",
+    "recommendation": "Use logger.warning for recoverable errors and logger.error for critical failures.",
+    "current_code": "logger.error(f\"Failed to load language{language}:{str(e)}\")",
+    "suggested_code": "logger.warning(f\"Failed to load language{language}:{str(e)}\")",
+    "file_path": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 108,
+    "end_line": 108,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Code Readability",
+    "description": "Lack of comments in complex sections.",
+    "impact": "medium",
+    "rationale": "While the code is mostly clear, adding comments in complex sections (like dynamic imports and tree traversal) would enhance readability for future developers.",
+    "recommendation": "Add comments explaining the purpose and functionality of complex code blocks.",
+    "current_code": "",
+    "suggested_code": "# Dynamically import the language module\n# Traverse the tree structure to extract relevant information",
+    "file_path": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 20,
+    "end_line": 21,
+    "change_type": "addition",
+    "sentiment": "positive",
+    "severity": 4
+  },
+  {
+    "category": "Performance",
+    "description": "Caching of language loading is not explicitly documented.",
+    "impact": "medium",
+    "rationale": "The use of lru_cache is a good performance optimization, but it's important to document its purpose and implications for maintainability.",
+    "recommendation": "Add a docstring to the load_language method explaining the caching behavior.",
+    "current_code": "",
+    "suggested_code": "\"\"\"Load and cache the specified language module using LRU caching.\"\"\"",
+    "file_path": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 15,
+    "end_line": 15,
+    "change_type": "addition",
+    "sentiment": "positive",
+    "severity": 3
+  },
+  {
+    "category": "Dependency Management",
+    "description": "Updating Python version in pyproject.toml.",
+    "impact": "medium",
+    "rationale": "Updating the Python version to a more recent one can improve compatibility with newer libraries and features.",
+    "recommendation": "Ensure all dependencies are compatible with the updated Python version.",
+    "current_code": "python = \"^3.8.1\"",
+    "suggested_code": "python = \"^3.9.0\"",
+    "file_path": "pyproject.toml",
+    "start_line": 13,
+    "end_line": 13,
+    "change_type": "modification",
+    "sentiment": "positive",
+    "severity": 5
+  },
+  {
+    "category": "Testing",
+    "description": "Lack of unit tests for new functionality.",
+    "impact": "high",
+    "rationale": "New features should have corresponding unit tests to ensure functionality and prevent regressions.",
+    "recommendation": "Add unit tests for the new language loading and parsing functionalities.",
+    "current_code": "",
+    "suggested_code": "# Add unit tests for LanguageLoader and ParserFactory",
+    "file_path": "tests/retriever/test_chunker.py",
+    "start_line": 1,
+    "end_line": 1,
+    "change_type": "addition",
+    "sentiment": "positive",
+    "severity": 8
+  },
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Docker",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to Dockerfile, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_path": "Dockerfile",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Docker",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to docker-compose.yml, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "15",
+    "end_line": "15",
+    "side": "RIGHT",
+    "file_path": "docker-compose.yml",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Version Control",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to .gitignore, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "164",
+    "end_line": "164",
+    "side": "RIGHT",
+    "file_path": ".gitignore",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Database",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to db_setup/init.sql, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_path": "db_setup/init.sql",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/review.md b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/review.md
new file mode 100644
index 00000000..907b5595
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/review.md
@@ -0,0 +1,136 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/222
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 23
+- Critical: 6
+- Important: 0
+- Minor: 0
+- Files Affected: 13
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Security (6 issues)</strong></summary>
+
+### 1. Hardcoded API keys in config.json.
+📁 **File:** `config.json:13`
+⚖️ **Severity:** 9/10
+🔍 **Description:** Hardcoded API keys in config.json.
+💡 **Solution:** 
+
+**Current Code:**
+```python
+"api_key": "os.environ/AZURE_API_KEY"
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 2. Changes made to sensitive file
+📁 **File:** `config.json:4`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 3. Changes made to sensitive file
+📁 **File:** `Dockerfile:4`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 4. Changes made to sensitive file
+📁 **File:** `docker-compose.yml:15`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 5. Changes made to sensitive file
+📁 **File:** `.gitignore:164`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 6. Changes made to sensitive file
+📁 **File:** `db_setup/init.sql:1`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 22541, "completion_tokens": 3669, "total_tokens": 26210}
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_440/comments.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/comments.json
similarity index 100%
rename from .experiments/code_review/haiku/no_eval/pr_440/comments.json
rename to .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/comments.json
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/issues.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/issues.json
new file mode 100644
index 00000000..6c136153
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/issues.json
@@ -0,0 +1,197 @@
+[
+  {
+    "category": "Import Optimization",
+    "description": "Redundant imports removed.",
+    "impact": "medium",
+    "rationale": "Removing unused imports helps in reducing bundle size and improving readability.",
+    "recommendation": "Ensure that all imports are necessary and remove any that are not used in the code.",
+    "current_code": "import{MemoriesIcon, NextIcon, SearchIcon, UrlIcon}from \"@repo/ui/icons\";",
+    "suggested_code": "import{MemoriesIcon, NextIcon, UrlIcon}from \"@repo/ui/icons\";",
+    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 4,
+    "end_line": 4,
+    "change_type": "modification",
+    "sentiment": "positive",
+    "severity": 5
+  },
+  {
+    "category": "Type Definition",
+    "description": "Type definition for TMemoriesPage is added.",
+    "impact": "high",
+    "rationale": "Providing a type definition improves type safety and makes the code easier to understand and maintain.",
+    "recommendation": "Consider adding more detailed comments to explain the purpose of each property in the type definition.",
+    "current_code": "",
+    "suggested_code": "type TMemoriesPage ={\n  memoriesAndSpaces:{memories: Content[]; spaces: StoredSpace[]};\n  title?: string;\n  currentSpace?: StoredSpace;\n  usersWithAccess?: string[];\n};",
+    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 40,
+    "end_line": 44,
+    "change_type": "addition",
+    "sentiment": "positive",
+    "severity": 7
+  },
+  {
+    "category": "Error Handling",
+    "description": "Improved error message for space deletion.",
+    "impact": "medium",
+    "rationale": "Providing a more specific error message can help users understand the issue better.",
+    "recommendation": "Consider logging the error or providing more context in the error message.",
+    "current_code": "toast.error(\"Failed to delete space\");",
+    "suggested_code": "toast.error(\"Failed to delete the space\");",
+    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 75,
+    "end_line": 75,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 4
+  },
+  {
+    "category": "Code Readability",
+    "description": "Improved readability of JSX structure.",
+    "impact": "medium",
+    "rationale": "Improving the structure of JSX can enhance readability and maintainability.",
+    "recommendation": "Consider using more descriptive class names and breaking down complex components into smaller ones.",
+    "current_code": "<div className=\"px-2 md:px-32 py-36 h-full flex mx-auto w-full flex-col gap-6\">",
+    "suggested_code": "<div className=\"px-2 md:px-32 py-36 h-full flex mx-auto w-full flex-col gap-6\">",
+    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 140,
+    "end_line": 140,
+    "change_type": "modification",
+    "sentiment": "positive",
+    "severity": 3
+  },
+  {
+    "category": "Component Naming",
+    "description": "Renamed TabComponent to SpaceComponent for clarity.",
+    "impact": "medium",
+    "rationale": "Using descriptive names for components improves code readability and understanding.",
+    "recommendation": "Ensure that component names reflect their functionality.",
+    "current_code": "function TabComponent({",
+    "suggested_code": "function SpaceComponent({",
+    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 275,
+    "end_line": 275,
+    "change_type": "modification",
+    "sentiment": "positive",
+    "severity": 5
+  },
+  {
+    "category": "Code Removal",
+    "description": "Removal of key state variables and imports without clear justification.",
+    "impact": "high",
+    "rationale": "The removal of state variables like 'spaces', 'content', and 'selectedSpaces' can lead to loss of functionality in the Menu component. Additionally, removing imports like 'useMeasure' may indicate missing functionality that could be required for layout measurements.",
+    "recommendation": "Reassess the necessity of removing these variables and imports. If they are not needed, ensure that the functionality relying on them is also removed or refactored appropriately.",
+    "current_code": "import useMeasure from \"react-use-measure\";\nconst[spaces, setSpaces] = useState<StoredSpace[]>([]);\nconst[content, setContent] = useState(\"\");\nconst[selectedSpaces, setSelectedSpaces] = useState<number[]>([]);",
+    "suggested_code": "import useMeasure from \"react-use-measure\";\nconst[spaces, setSpaces] = useState<StoredSpace[]>([]);\nconst[content, setContent] = useState(\"\");\nconst[selectedSpaces, setSelectedSpaces] = useState<number[]>([]);",
+    "file_path": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 32,
+    "end_line": 80,
+    "change_type": "deletion",
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Error Handling",
+    "description": "Inconsistent error handling in async functions.",
+    "impact": "high",
+    "rationale": "The error handling in the 'handleSubmit' function is not consistent. It throws an error after displaying a toast, which could lead to unhandled promise rejections. This can cause the application to crash or behave unexpectedly.",
+    "recommendation": "Ensure that all async functions have consistent error handling. Consider using try-catch blocks to manage errors gracefully and avoid throwing errors after displaying user notifications.",
+    "current_code": "if (cont.success){\n    toast.success(\"Memory queued\",{\n        richColors: true,\n});\n}else{\n    toast.error(`Memory creation failed: ${cont.error}`);\n    throw new Error(`Memory creation failed: ${cont.error}`);\n}",
+    "suggested_code": "try{\n    const cont = await createMemory({\n        content: content,\n        spaces: spaces ?? undefined,\n});\n    if (cont.success){\n        toast.success(\"Memory queued\",{\n            richColors: true,\n});\n}else{\n        throw new Error(`Memory creation failed: ${cont.error}`);\n}\n}catch (error){\n    toast.error(error.message);\n}",
+    "file_path": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 112,
+    "end_line": 130,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Readability",
+    "description": "Complexity in JSX structure can hinder readability.",
+    "impact": "medium",
+    "rationale": "The JSX structure in the 'DialogContentContainer' function is quite complex and may be difficult for future developers to read and maintain. Breaking it down into smaller components can improve readability and maintainability.",
+    "recommendation": "Consider breaking down the JSX structure into smaller, reusable components. This will enhance readability and make it easier to manage the code in the future.",
+    "current_code": "<DialogContent className=\"sm:max-w-[575px] text-[#F2F3F5] rounded-2xl bg-background z-[39]\">\n    <form action={async (e: FormData) =>{...}}>\n{/* Form Fields */}\n    </form>\n</DialogContent>",
+    "suggested_code": "<DialogContent className=\"sm:max-w-[575px] text-[#F2F3F5] rounded-2xl bg-background z-[39]\">\n    <FormComponent onSubmit={handleSubmit}/>\n</DialogContent>",
+    "file_path": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 135,
+    "end_line": 346,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Import Optimization",
+    "description": "Unused import from React.",
+    "impact": "medium",
+    "rationale": "The import statement for useEffect is removed but not used in the code, which can lead to confusion about the necessity of the import.",
+    "recommendation": "Remove unused imports to enhance code clarity and maintainability.",
+    "current_code": "import{useState, useEffect}from \"react\";",
+    "suggested_code": "import{useState}from \"react\";",
+    "file_path": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 3,
+    "end_line": 3,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Interface Definition",
+    "description": "New interface 'StoredSpace' added.",
+    "impact": "high",
+    "rationale": "The addition of the 'StoredSpace' interface improves type safety and clarity in the code, allowing for better maintainability and understanding of the data structure.",
+    "recommendation": "Ensure that all interfaces are documented to provide context for future developers.",
+    "current_code": "",
+    "suggested_code": "interface StoredSpace{\n\tid: number;\n\tname: string;\n}",
+    "file_path": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 19,
+    "end_line": 22,
+    "change_type": "addition",
+    "sentiment": "positive",
+    "severity": 7
+  },
+  {
+    "category": "Props Structure",
+    "description": "Props structure modified to include selectedSpaces and setSelectedSpaces.",
+    "impact": "high",
+    "rationale": "This change enhances the component's functionality by allowing it to manage selected spaces, which is crucial for the component's purpose.",
+    "recommendation": "Consider adding PropTypes or TypeScript documentation for better clarity on prop types.",
+    "current_code": "placeholder?: string;\nemptyMessage?: string;\ncreateNewMessage?: string;\nclassName?: string;",
+    "suggested_code": "selectedSpaces: number[];\nsetSelectedSpaces: React.Dispatch<React.SetStateAction<number[]>>;",
+    "file_path": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 23,
+    "end_line": 28,
+    "change_type": "modification",
+    "sentiment": "positive",
+    "severity": 6
+  },
+  {
+    "category": "Event Handling",
+    "description": "Event handlers for input change and key down events added.",
+    "impact": "high",
+    "rationale": "These handlers improve user interaction by providing feedback and functionality when users type or press keys, enhancing the overall user experience.",
+    "recommendation": "Ensure that event handlers are well-tested to handle edge cases, such as rapid key presses.",
+    "current_code": "",
+    "suggested_code": "const handleInputChange = (e: React.ChangeEvent<HTMLInputElement>) =>{\n\tsetInputValue(e.target.value);\n};\n\nconst handleKeyDown = (e: React.KeyboardEvent<HTMLInputElement>) =>{\n\tif (\n\t\te.key === \"Backspace\" &&\n\t\tinputValue === \"\" &&\n\t\tselectedSpaces.length > 0\n\t\t){\n\t\tsetSelectedSpaces((prev) => prev.slice(0, -1));\n\t}\n};",
+    "file_path": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 41,
+    "end_line": 53,
+    "change_type": "addition",
+    "sentiment": "positive",
+    "severity": 7
+  },
+  {
+    "category": "Filtering Logic",
+    "description": "Filtering logic for options based on selectedSpaces added.",
+    "impact": "high",
+    "rationale": "This logic ensures that the user only sees options that are not already selected, improving usability and preventing confusion.",
+    "recommendation": "Consider adding unit tests to verify that the filtering logic works as expected under various scenarios.",
+    "current_code": "",
+    "suggested_code": "const filteredOptions = options.filter(\n\t(option) => !selectedSpaces.includes(parseInt(option.value)),\n);",
+    "file_path": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 55,
+    "end_line": 57,
+    "change_type": "addition",
+    "sentiment": "positive",
+    "severity": 6
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/review.md b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/review.md
new file mode 100644
index 00000000..ea09b5db
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/review.md
@@ -0,0 +1,33 @@
+PR URL: https://github.com/supermemoryai/supermemory/pull/232
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 13
+- Critical: 0
+- Important: 0
+- Minor: 0
+- Files Affected: 3
+## 🏆 Code Quality
+[███████████████░░░░░] 75% (Fair)
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 16642, "completion_tokens": 2919, "total_tokens": 19561}
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_232/comments.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/comments.json
similarity index 100%
rename from .experiments/code_review/llama-405b/no_eval/pr_232/comments.json
rename to .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/comments.json
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/issues.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/issues.json
new file mode 100644
index 00000000..0f16989d
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/issues.json
@@ -0,0 +1,62 @@
+[
+  {
+    "category": "Code Structure",
+    "description": "Inconsistent parameter formatting in function calls.",
+    "impact": "medium",
+    "rationale": "The change to the `generate_linkedin_post` function call introduces a multi-line format, which could reduce readability and consistency in function calls across the codebase.",
+    "recommendation": "Consider keeping function calls in a single line unless they exceed a reasonable length. This will enhance readability and maintainability.",
+    "current_code": "linkedin_post = work_summary_generator.generate_linkedin_post(\n     summary, user=\"oss_example\"\n )",
+    "suggested_code": "linkedin_post = work_summary_generator.generate_linkedin_post(summary, user=\"oss_example\")",
+    "file_path": "examples/work_summarizer/main.py",
+    "start_line": 60,
+    "end_line": 62,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling",
+    "description": "Missing error handling for JSON parsing.",
+    "impact": "high",
+    "rationale": "The `json.loads` function in `kaizen/helpers/parser.py` does not handle potential exceptions that may arise from malformed JSON, which could lead to runtime crashes.",
+    "recommendation": "Wrap the JSON parsing in a try-except block to handle possible exceptions and provide meaningful error messages.",
+    "current_code": "parsed_data = json.loads(json_data)",
+    "suggested_code": "try:\n    parsed_data = json.loads(json_data)\nexcept json.JSONDecodeError as e:\n    logging.error(f\"Failed to parse JSON:{e}\")\n    return None",
+    "file_path": "kaizen/helpers/parser.py",
+    "start_line": 47,
+    "end_line": 48,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Documentation",
+    "description": "Lack of docstrings for new functions.",
+    "impact": "medium",
+    "rationale": "New functions like `generate_twitter_post` and `generate_linkedin_post` lack docstrings, making it difficult for future developers to understand their purpose and usage.",
+    "recommendation": "Add docstrings to all new functions to describe their parameters, return types, and functionality.",
+    "current_code": "",
+    "suggested_code": "\"\"\"\nGenerates a Twitter post based on the provided summary.\n\nArgs:\n    summary (Dict): The summary data.\n    user (Optional[str]): The user requesting the post.\n\nReturns:\n    str: The generated Twitter post.\n\"\"\"",
+    "file_path": "kaizen/reviewer/work_summarizer.py",
+    "start_line": 58,
+    "end_line": 66,
+    "change_type": "addition",
+    "sentiment": "positive",
+    "severity": 6
+  },
+  {
+    "category": "Code Consistency",
+    "description": "Inconsistent naming conventions for constants.",
+    "impact": "medium",
+    "rationale": "The constants `TWITTER_POST_PROMPT` and `LINKEDIN_POST_PROMPT` are defined without a consistent naming style compared to other constants in the codebase.",
+    "recommendation": "Ensure all constants follow a consistent naming convention, such as using uppercase letters with underscores.",
+    "current_code": "TWITTER_POST_PROMPT = \"\"\"\nGiven the following work summary, create a concise and engaging Twitter post (max 280 characters) that highlights the key changes or improvements. Format the post as markdown, enclosed in triple backticks:\n\nSummary:\n{SUMMARY}\n\nTwitter Post:\n```\n<Your Twitter post here>\n```\n\"\"\"",
+    "suggested_code": "TWITTER_POST_PROMPT = \"\"\"\nGiven the following work summary, create a concise and engaging Twitter post (max 280 characters) that highlights the key changes or improvements. Format the post as markdown, enclosed in triple backticks:\n\nSummary:\n{SUMMARY}\n\nTwitter Post:\n```\n<Your Twitter post here>\n```\n\"\"\"",
+    "file_path": "kaizen/llms/prompts/work_summary_prompts.py",
+    "start_line": 44,
+    "end_line": 54,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 4
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/review.md b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/review.md
new file mode 100644
index 00000000..3de0640e
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/review.md
@@ -0,0 +1,33 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/252
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 4
+- Critical: 0
+- Important: 0
+- Minor: 0
+- Files Affected: 4
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 4135, "completion_tokens": 989, "total_tokens": 5124}
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_252/comments.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/comments.json
similarity index 100%
rename from .experiments/code_review/llama-405b/no_eval/pr_252/comments.json
rename to .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/comments.json
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/issues.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/issues.json
new file mode 100644
index 00000000..b6b4d441
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/issues.json
@@ -0,0 +1,62 @@
+[
+  {
+    "category": "Import Changes",
+    "description": "Updated import paths for prompts.",
+    "impact": "medium",
+    "rationale": "Changing import paths can lead to confusion if not documented properly, especially if the previous imports were widely used.",
+    "recommendation": "Ensure that the new import paths are well-documented and tested to avoid runtime errors.",
+    "current_code": "from kaizen.llms.prompts.code_review_prompts import (",
+    "suggested_code": "from kaizen.llms.prompts.pr_desc_prompts import (",
+    "file_path": "kaizen/generator/pr_description.py",
+    "start_line": 8,
+    "end_line": 8,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Parameter Removal",
+    "description": "Removed 'reeval_response' parameter from multiple functions.",
+    "impact": "high",
+    "rationale": "Removing parameters can break existing functionality if the functions are called elsewhere with those parameters.",
+    "recommendation": "Review all usages of these functions to ensure that removing 'reeval_response' does not cause issues. If it's unnecessary, consider marking it as deprecated instead of removing it outright.",
+    "current_code": "reeval_response: bool = False,",
+    "suggested_code": "",
+    "file_path": "kaizen/generator/pr_description.py",
+    "start_line": 43,
+    "end_line": 43,
+    "change_type": "deletion",
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Error Handling",
+    "description": "Potential lack of error handling after removing 'reeval_response'.",
+    "impact": "high",
+    "rationale": "The removal of this parameter could lead to unhandled exceptions if the logic that depended on it is not properly adjusted.",
+    "recommendation": "Add error handling or checks to ensure that the new logic remains robust and handles edge cases appropriately.",
+    "current_code": "desc = self._process_full_diff(prompt, user)",
+    "suggested_code": "if not prompt or not user: raise ValueError('Invalid parameters')\ndesc = self._process_full_diff(prompt, user)",
+    "file_path": "kaizen/generator/pr_description.py",
+    "start_line": 54,
+    "end_line": 54,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Code Clarity",
+    "description": "Improved clarity in prompt descriptions.",
+    "impact": "medium",
+    "rationale": "The new prompts are clearer and more structured, which enhances readability and maintainability.",
+    "recommendation": "Continue to follow this pattern for future prompt definitions to maintain consistency.",
+    "current_code": "PR_DESCRIPTION_PROMPT = \"\"\"",
+    "suggested_code": "PR_DESCRIPTION_SYSTEM_PROMPT = \"\"\"",
+    "file_path": "kaizen/llms/prompts/pr_desc_prompts.py",
+    "start_line": 1,
+    "end_line": 1,
+    "change_type": "addition",
+    "sentiment": "positive",
+    "severity": 3
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/review.md b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/review.md
new file mode 100644
index 00000000..4919cde8
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/review.md
@@ -0,0 +1,33 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/335
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 4
+- Critical: 0
+- Important: 0
+- Minor: 0
+- Files Affected: 2
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 7000, "completion_tokens": 730, "total_tokens": 7730}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/comments.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/comments.json
new file mode 100644
index 00000000..1ceabeef
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/comments.json
@@ -0,0 +1,16 @@
+[
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "11",
+    "end_line": "11",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/issues.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/issues.json
new file mode 100644
index 00000000..d371af1f
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/issues.json
@@ -0,0 +1,346 @@
+[
+  {
+    "category": "Code Clarity",
+    "description": "Improper formatting of multi-line strings.",
+    "impact": "medium",
+    "rationale": "The formatting of the `DESC_COLLAPSIBLE_TEMPLATE` string could lead to confusion regarding its structure and readability. Multi-line strings should be formatted consistently to enhance clarity.",
+    "recommendation": "Use triple quotes for multi-line strings to maintain readability and avoid confusion.",
+    "current_code": "DESC_COLLAPSIBLE_TEMPLATE = \"<details><summary>Original Description</summary>{desc}</details>\"",
+    "suggested_code": "DESC_COLLAPSIBLE_TEMPLATE = \"\"\"\n<details><summary>Original Description</summary>{desc}</details>\n\"\"\"",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
+    "start_line": 5,
+    "end_line": 6,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling",
+    "description": "Lack of type validation for function parameters.",
+    "impact": "high",
+    "rationale": "The function `create_pr_description` should validate input types to prevent runtime errors. Currently, it assumes that inputs are always strings, which may lead to unexpected behavior.",
+    "recommendation": "Implement type checks for the parameters and raise appropriate exceptions if the types are incorrect.",
+    "current_code": "def create_pr_description(desc, original_desc):",
+    "suggested_code": "def create_pr_description(desc: str, original_desc: str):\n    if not isinstance(desc, str):\n        raise TypeError('desc must be a string')\n    if not isinstance(original_desc, str):\n        raise TypeError('original_desc must be a string')",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/output.py",
+    "start_line": 1,
+    "end_line": 1,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Test Coverage",
+    "description": "Insufficient coverage for edge cases in tests.",
+    "impact": "high",
+    "rationale": "While there are tests for normal cases, edge cases such as extremely long strings or special characters are not adequately covered, which could lead to failures in production.",
+    "recommendation": "Add more test cases to cover edge scenarios, including empty strings, very long strings, and special characters.",
+    "current_code": "def test_create_pr_description_normal_and_edge_cases(desc, original_desc, expected):",
+    "suggested_code": "def test_create_pr_description_edge_cases():\n    # Add tests for edge cases here",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
+    "start_line": 28,
+    "end_line": 28,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 7
+  },
+  {
+    "category": "Performance",
+    "description": "Potential performance issue with repeated string concatenation.",
+    "impact": "medium",
+    "rationale": "Using repeated string concatenation can lead to performance degradation, especially with large strings. It is more efficient to use `str.join()` for concatenation.",
+    "recommendation": "Refactor string concatenation to use `str.join()` for better performance.",
+    "current_code": "result = desc + original_desc",
+    "suggested_code": "result = ''.join([desc, original_desc])",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/output.py",
+    "start_line": 10,
+    "end_line": 10,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Code Structure",
+    "description": "Redundant code removal and template improvement.",
+    "impact": "high",
+    "rationale": "The original PR_COLLAPSIBLE_TEMPLATE was overly verbose and could lead to maintenance challenges. The new format is cleaner and easier to read.",
+    "recommendation": "Ensure that all templates follow a consistent structure to improve readability and maintainability.",
+    "current_code": "PR_COLLAPSIBLE_TEMPLATE = (\n    \"<details>\\n\"\n    \"<summary>Review</summary>\\n\"\n    \"<p>\\n\"\n    \"<b>Comment:</b>{comment}<br>\\n\"\n    \"<b>Reason:</b>{reason}<br>\\n\"\n    \"<b>Solution:</b>{solution}<br>\\n\"\n    \"<b>Confidence:</b>{confidence}<br>\\n\"\n    \"<b>Start Line:</b>{start_line}<br>\\n\"\n    \"<b>End Line:</b>{end_line}<br>\\n\"\n    \"<b>File Name:</b>{file_name}<br>\\n\"\n    \"<b>Severity:</b>{severity}<br>\\n\"\n    \"</p>\\n\"\n    \"</details>\"\n)",
+    "suggested_code": "PR_COLLAPSIBLE_TEMPLATE = \"\"\"\n<details>\n<summary>Review Comment</summary>\n<p>{comment}</p>\n<p><strong>Reason:</strong>{reason}</p>\n<p><strong>Solution:</strong>{solution}</p>\n<p><strong>Confidence:</strong>{confidence}</p>\n<p><strong>Start Line:</strong>{start_line}</p>\n<p><strong>End Line:</strong>{end_line}</p>\n<p><strong>File Name:</strong>{file_name}</p>\n<p><strong>Severity:</strong>{severity}</p>\n</details>\n\"\"\"",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 4,
+    "end_line": 16,
+    "change_type": "modification",
+    "sentiment": "positive",
+    "severity": 7
+  },
+  {
+    "category": "Error Handling",
+    "description": "Missing validation for required fields in review comments.",
+    "impact": "high",
+    "rationale": "The tests should validate that all required fields are provided to avoid runtime errors when generating review texts.",
+    "recommendation": "Implement checks to ensure that all required fields (comment, reason, solution, confidence, start_line, end_line, file_name, severity_level) are present before processing.",
+    "current_code": "topics ={\n    \"topic1\":[\n{\n            \"comment\": \"This is a test comment.\",\n            \"reason\": \"This is a test reason.\",\n            \"solution\": \"This is a test solution.\",\n            \"confidence\": \"critical\",\n            \"start_line\": 10,\n            \"end_line\": 20,\n            \"file_name\": \"test_file.py\",\n            \"severity_level\": 9,\n}\n    ]\n}",
+    "suggested_code": "def validate_review(review):\n    required_fields =[\"comment\", \"reason\", \"solution\", \"confidence\", \"start_line\", \"end_line\", \"file_name\", \"severity_level\"]\n    for field in required_fields:\n        if field not in review:\n            raise ValueError(f'Missing required field:{field}')",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 18,
+    "end_line": 18,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 8
+  },
+  {
+    "category": "Test Coverage",
+    "description": "Insufficient tests for edge cases.",
+    "impact": "medium",
+    "rationale": "While there are tests for normal cases, edge cases such as missing fields or empty topics are not adequately covered.",
+    "recommendation": "Add unit tests that specifically check for edge cases, such as missing fields or empty lists, to ensure robustness.",
+    "current_code": "def test_empty_topics():\n    topics ={}\n    expected_output = \"## Code Review\\n\\n\u2705 **All Clear:** This PR is ready to merge! \ud83d\udc4d\\n\\n\"\n    assert create_pr_review_text(topics) == expected_output",
+    "suggested_code": "def test_reviews_with_missing_fields():\n    topics ={\n        \"topic1\":[{\n            \"comment\": \"This is a test comment.\",\n            \"confidence\": \"critical\",\n            \"start_line\": 10,\n            \"end_line\": 20,\n            \"file_name\": \"test_file.py\",\n            \"severity_level\": 9,\n}]\n}\n    with pytest.raises(ValueError):\n        create_pr_review_text(topics)",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 238,
+    "end_line": 238,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Readability",
+    "description": "Inconsistent naming conventions for importance values.",
+    "impact": "medium",
+    "rationale": "The importance values are sometimes written in lowercase ('high') and sometimes in title case ('High'). This inconsistency can lead to confusion and errors when comparing or processing these values.",
+    "recommendation": "Standardize the naming convention for importance values across the codebase. Choose either all lowercase or title case and apply it consistently.",
+    "current_code": "            \"importance\": \"high\",",
+    "suggested_code": "            \"importance\": \"High\",",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 56,
+    "end_line": 56,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling",
+    "description": "Lack of error handling for file operations.",
+    "impact": "high",
+    "rationale": "The code does not handle potential exceptions that may arise from file operations, such as file not found or permission errors. This could lead to unhandled exceptions during runtime.",
+    "recommendation": "Wrap file operations in try-except blocks to handle potential exceptions gracefully and provide meaningful error messages.",
+    "current_code": "    with open(file_path, 'r') as f:",
+    "suggested_code": "    try:\n        with open(file_path, 'r') as f:\n            return f.read()\n    except FileNotFoundError:\n        raise Exception(f\"File not found:{file_path}\")\n    except PermissionError:\n        raise Exception(f\"Permission denied:{file_path}\")",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 17,
+    "end_line": 17,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Performance",
+    "description": "Inefficient handling of long test names.",
+    "impact": "medium",
+    "rationale": "The code currently constructs long test names by concatenating strings, which can be inefficient and hard to read. This could be optimized for better performance and clarity.",
+    "recommendation": "Use f-strings for constructing long test names to improve readability and performance.",
+    "current_code": "     long_test_name = \"Test \" + \"Example \" * 50",
+    "suggested_code": "     long_test_name = f\"Test{' '.join(['Example'] * 50)}\"",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 131,
+    "end_line": 131,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 4
+  },
+  {
+    "category": "Maintainability",
+    "description": "Redundant mocking of dependencies in multiple tests.",
+    "impact": "medium",
+    "rationale": "Mocking dependencies is repeated across multiple test functions, which can lead to code duplication and make maintenance harder.",
+    "recommendation": "Consider using a fixture for mocking dependencies to reduce code duplication and improve maintainability.",
+    "current_code": "@mock.patch(\"kaizen.helpers.output.create_folder\")\n@mock.patch(\"kaizen.helpers.output.general.clean_python_code\")\n@mock.patch(\"kaizen.helpers.output.logger\")\ndef test_create_test_files_normal_case(",
+    "suggested_code": "@pytest.fixture\ndef mock_dependencies():\n    with mock.patch('kaizen.helpers.output.create_folder') as mock_create_folder,\n         mock.patch('kaizen.helpers.output.general.clean_python_code') as mock_clean_python_code,\n         mock.patch('kaizen.helpers.output.logger') as mock_logger:\n        yield mock_create_folder, mock_clean_python_code, mock_logger",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 7,
+    "end_line": 13,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Test Coverage",
+    "description": "Missing tests for edge cases and error handling.",
+    "impact": "high",
+    "rationale": "While the tests cover normal cases, edge cases and error handling scenarios are crucial for ensuring the robustness of the code. Without these tests, potential bugs may go unnoticed.",
+    "recommendation": "Add tests for edge cases such as invalid inputs and unexpected behaviors.",
+    "current_code": "def test_get_web_html_normal_case(mock_nest_asyncio_apply, mock_get_html):",
+    "suggested_code": "def test_get_web_html_edge_cases(mock_nest_asyncio_apply, mock_get_html):",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 14,
+    "end_line": 14,
+    "change_type": "addition",
+    "sentiment": "neutral",
+    "severity": 7
+  },
+  {
+    "category": "Error Handling",
+    "description": "Inconsistent error messages in exception handling.",
+    "impact": "medium",
+    "rationale": "The error messages in the exception handling do not provide clear context for debugging. Consistent and descriptive error messages are important for understanding failures.",
+    "recommendation": "Use consistent error messages that include context about the operation being performed.",
+    "current_code": "with pytest.raises(OSError, match=\"Unable to determine current working directory\"):",
+    "suggested_code": "with pytest.raises(OSError, match=\"Failed to get current working directory during get_parent_folder call\"):",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py",
+    "start_line": 21,
+    "end_line": 21,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Code Readability",
+    "description": "Lack of comments and documentation for new functions.",
+    "impact": "medium",
+    "rationale": "New functions introduced in the tests lack comments explaining their purpose and usage, which can hinder understanding for future developers.",
+    "recommendation": "Add docstrings to new functions to describe their purpose and expected behavior.",
+    "current_code": "def test_get_web_html_normal():",
+    "suggested_code": "def test_get_web_html_normal():  # Test case for normal HTML input",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 13,
+    "end_line": 13,
+    "change_type": "modification",
+    "sentiment": "positive",
+    "severity": 4
+  },
+  {
+    "category": "Performance",
+    "description": "Potential performance issues with large HTML content.",
+    "impact": "medium",
+    "rationale": "The test for large HTML content may lead to performance issues if the content size grows significantly. This could slow down test execution.",
+    "recommendation": "Consider limiting the size of the HTML content in tests or using mocks to simulate large content.",
+    "current_code": "large_html_content = \"<html><body>\" + \"<p>Test</p>\" * 10000 + \"</body></html>\"",
+    "suggested_code": "large_html_content = \"<html><body><p>Test</p></body></html>\"  # Simplified for performance",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 93,
+    "end_line": 93,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Code Duplication",
+    "description": "Redundant imports and definitions of supported languages.",
+    "impact": "high",
+    "rationale": "The original code had multiple definitions of supported languages and unnecessary imports, which can lead to confusion and maintenance challenges.",
+    "recommendation": "Consolidate the supported languages into a single definition and ensure imports are only made once.",
+    "current_code": "self.supported_languages ={\n    \"py\": \"PythonParser\",\n    \"js\": \"JavaScriptParser\",\n    \"ts\": \"TypeScriptParser\",\n    \"jsx\": \"ReactParser\",\n    \"tsx\": \"ReactTSParser\",\n    \"rs\": \"RustParser\",\n}",
+    "suggested_code": "SUPPORTED_LANGUAGES ={\n    \"py\": \"PythonParser\",\n    \"js\": \"JavaScriptParser\",\n    \"ts\": \"TypeScriptParser\",\n    \"jsx\": \"ReactParser\",\n    \"tsx\": \"ReactTSParser\",\n    \"rs\": \"RustParser\",\n}",
+    "file_path": "kaizen/generator/unit_test.py",
+    "start_line": 24,
+    "end_line": 30,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 7
+  },
+  {
+    "category": "Error Handling",
+    "description": "Lack of error handling in file operations.",
+    "impact": "high",
+    "rationale": "The code currently does not handle potential exceptions that may arise from file operations, such as file not found or permission errors, which could lead to runtime failures.",
+    "recommendation": "Implement try-except blocks around file operations to gracefully handle errors.",
+    "current_code": "with open(file_path, \"r\") as file:\n    return file.read()",
+    "suggested_code": "try:\n    with open(file_path, \"r\") as file:\n        return file.read()\nexcept (FileNotFoundError, IOError) as e:\n    print(f\"Error reading file{file_path}:{e}\")",
+    "file_path": "kaizen/generator/unit_test.py",
+    "start_line": 108,
+    "end_line": 110,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Code Readability",
+    "description": "Inconsistent use of comments and docstrings.",
+    "impact": "medium",
+    "rationale": "Some methods lack docstrings, making it harder to understand their purpose and usage. Consistent documentation improves maintainability and usability.",
+    "recommendation": "Add docstrings to all methods to describe their purpose, parameters, and return values.",
+    "current_code": "def generate_tests(self, file_path: str, content: str = None, output_path: str = None):",
+    "suggested_code": "def generate_tests(self, file_path: str, content: str = None, output_path: str = None):\n    \"\"\"Generate tests based on the provided file path and content.\"\"\"",
+    "file_path": "kaizen/generator/unit_test.py",
+    "start_line": 75,
+    "end_line": 75,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Performance",
+    "description": "Potential inefficiency in reading file content.",
+    "impact": "medium",
+    "rationale": "The method `_read_file_content` reads the entire file into memory, which may not be efficient for large files.",
+    "recommendation": "Consider processing the file in chunks if large files are expected, or document the expected file sizes.",
+    "current_code": "return file.read()",
+    "suggested_code": "content =[]\nfor line in file:\n    content.append(line)\nreturn ''.join(content)",
+    "file_path": "kaizen/generator/unit_test.py",
+    "start_line": 109,
+    "end_line": 109,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Code Structure",
+    "description": "Function definition should be encapsulated properly.",
+    "impact": "medium",
+    "rationale": "The function `set_all_loggers_to_ERROR` is defined and called at the module level, which can lead to unexpected behavior if the module is imported elsewhere. This could also make unit testing more difficult.",
+    "recommendation": "Encapsulate the logger configuration within a main guard to prevent it from executing on import.",
+    "current_code": "set_all_loggers_to_ERROR()",
+    "suggested_code": "if __name__ == '__main__':\n    set_all_loggers_to_ERROR()",
+    "file_path": "kaizen/llms/provider.py",
+    "start_line": 23,
+    "end_line": 23,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Error Handling",
+    "description": "Lack of error handling when setting logger levels.",
+    "impact": "high",
+    "rationale": "If there are issues with setting the logger levels (e.g., if a logger does not exist), the current implementation will fail silently, which can make debugging difficult.",
+    "recommendation": "Add error handling to log any exceptions that occur when setting logger levels.",
+    "current_code": "logging.getLogger(name).setLevel(logging.ERROR)",
+    "suggested_code": "try:\n    logging.getLogger(name).setLevel(logging.ERROR)\nexcept Exception as e:\n    print(f'Error setting level for logger{name}:{e}')",
+    "file_path": "kaizen/llms/provider.py",
+    "start_line": 18,
+    "end_line": 18,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Readability",
+    "description": "Inconsistent naming conventions for logger names.",
+    "impact": "medium",
+    "rationale": "The logger names such as 'LiteLLM', 'LiteLLM Router', and 'LiteLLM Proxy' should follow a consistent naming convention for better readability and maintainability.",
+    "recommendation": "Standardize the naming convention for loggers, possibly using underscores or camel case consistently.",
+    "current_code": "logging.getLogger(\"LiteLLM\").setLevel(logging.ERROR)",
+    "suggested_code": "logging.getLogger(\"lite_llm\").setLevel(logging.ERROR)",
+    "file_path": "kaizen/llms/provider.py",
+    "start_line": 26,
+    "end_line": 28,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "11",
+    "end_line": "11",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/review.md b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/review.md
new file mode 100644
index 00000000..2e01a332
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/review.md
@@ -0,0 +1,56 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/400
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 23
+- Critical: 1
+- Important: 0
+- Minor: 0
+- Files Affected: 9
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Configuration (1 issues)</strong></summary>
+
+### 1. Changes made to sensitive file
+📁 **File:** `config.json:11`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 38388, "completion_tokens": 5366, "total_tokens": 43754}
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_335/comments.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/comments.json
similarity index 100%
rename from .experiments/code_review/llama-405b/no_eval/pr_335/comments.json
rename to .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/comments.json
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/issues.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/issues.json
new file mode 100644
index 00000000..d8a0bb37
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/issues.json
@@ -0,0 +1,47 @@
+[
+  {
+    "category": "Code Modification",
+    "description": "Redundant comment and potential for confusion with TODO.",
+    "impact": "medium",
+    "rationale": "The comment 'TODO: DONT PUSH DUPLICATE' lacks clarity and may lead to misunderstandings about its purpose. Comments should provide clear guidance or context.",
+    "recommendation": "Clarify the comment to specify what should not be duplicated and why. Consider removing it if it does not add value.",
+    "current_code": "# TODO: DONT PUSH DUPLICATE",
+    "suggested_code": "# TODO: Ensure unique embeddings are stored to avoid duplicates in the database.",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 157,
+    "end_line": 157,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Code Consistency",
+    "description": "Inconsistent use of comments regarding repository setup.",
+    "impact": "medium",
+    "rationale": "The comment above the line where `setup_repository` is called suggests that it should only be done during initial analysis or updates, but the line itself is always executed. This can lead to confusion.",
+    "recommendation": "Update the comment to reflect the actual behavior of the code, or consider removing it if it does not provide significant context.",
+    "current_code": "# Set up the repository (do this when you first analyze a repo or when you want to update it)",
+    "suggested_code": "# Set up the repository to ensure it is ready for analysis.",
+    "file_path": "examples/ragify_codebase/main.py",
+    "start_line": 6,
+    "end_line": 6,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 4
+  },
+  {
+    "category": "Dependency Update",
+    "description": "Updating dependencies without clear documentation.",
+    "impact": "high",
+    "rationale": "The removal of specific versions in favor of a newer version can lead to compatibility issues if the new version introduces breaking changes. It is crucial to document such changes clearly.",
+    "recommendation": "Add a note in the changelog or comments explaining why the dependency versions were updated and any potential impacts.",
+    "current_code": "llama-index-core = \"^0.10.47\"",
+    "suggested_code": "llama-index-core = \"0.10.65\"  # Updated for performance improvements and bug fixes.",
+    "file_path": "pyproject.toml",
+    "start_line": 27,
+    "end_line": 27,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 6
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/review.md b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/review.md
new file mode 100644
index 00000000..d14252ac
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/review.md
@@ -0,0 +1,33 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/440
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 3
+- Critical: 0
+- Important: 0
+- Minor: 0
+- Files Affected: 3
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 1613, "completion_tokens": 626, "total_tokens": 2239}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/comments.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/comments.json
new file mode 100644
index 00000000..507ce250
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/comments.json
@@ -0,0 +1,31 @@
+[
+  {
+    "category": "Security Vulnerability",
+    "description": "Potential exposure of sensitive data in logs.",
+    "impact": "critical",
+    "rationale": "Logging sensitive information, such as access tokens or personal data, can lead to security vulnerabilities. It's essential to sanitize logs to avoid exposing sensitive data.",
+    "recommendation": "Review logging statements to ensure sensitive data is not logged. Implement logging best practices.",
+    "current_code": "logger.debug(f\"Post Review comment response:{response.text}\")",
+    "suggested_code": "logger.debug(\"Post Review comment response received.\")",
+    "file_path": "github_app/github_helper/pull_requests.py",
+    "start_line": 181,
+    "end_line": 182,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 9
+  },
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/issues.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/issues.json
new file mode 100644
index 00000000..1364b6ee
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/issues.json
@@ -0,0 +1,91 @@
+[
+  {
+    "category": "Error Handling",
+    "description": "Broad exception handling without specific error management.",
+    "impact": "high",
+    "rationale": "Using a generic 'except Exception' clause can mask underlying issues and make debugging difficult. It is better to handle specific exceptions to provide clearer error messages and control flow.",
+    "recommendation": "Refine the exception handling to catch specific exceptions and log meaningful error messages.",
+    "current_code": "except Exception:\n    print(\"Error\")",
+    "suggested_code": "except KeyError:\n    print(f\"Key error occurred with min_confidence:{min_confidence}\")",
+    "file_path": "github_app/github_helper/pull_requests.py",
+    "start_line": 140,
+    "end_line": 141,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Code Readability",
+    "description": "Inconsistent naming conventions for variables.",
+    "impact": "medium",
+    "rationale": "Using inconsistent naming conventions can lead to confusion and reduce code readability. It's important to maintain a consistent style throughout the codebase.",
+    "recommendation": "Adopt a consistent naming convention (e.g., snake_case for variables) and ensure all variables follow this convention.",
+    "current_code": "repo_name, pull_request_files=pr_files",
+    "suggested_code": "repo_name, pull_request_files=pr_files",
+    "file_path": "github_app/github_helper/pull_requests.py",
+    "start_line": 55,
+    "end_line": 56,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Performance Optimization",
+    "description": "Inefficient sorting algorithm in sort_files function.",
+    "impact": "high",
+    "rationale": "The current sorting method is O(n^2) due to nested loops. Using Python's built-in sorting functions can significantly improve performance.",
+    "recommendation": "Utilize Python's built-in sorting capabilities to improve performance.",
+    "current_code": "for i, sorted_file in enumerate(sorted_files):\n    if file_name < sorted_file[\"filename\"]:\n        min_index = i\n        break",
+    "suggested_code": "sorted_files = sorted(files, key=lambda x: x['filename'])",
+    "file_path": "github_app/github_helper/pull_requests.py",
+    "start_line": 184,
+    "end_line": 194,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Documentation",
+    "description": "Lack of comments and documentation for newly added functions.",
+    "impact": "medium",
+    "rationale": "New functions like 'generate_tests' lack docstrings, making it difficult for other developers to understand their purpose and usage.",
+    "recommendation": "Add docstrings to all new functions to explain their purpose, parameters, and return values.",
+    "current_code": "def generate_tests(pr_files):\n    return[f[\"filename\"] for f in pr_files]",
+    "suggested_code": "def generate_tests(pr_files):\n    \"\"\"Generate a list of test filenames from PR files.\"\n    return[f[\"filename\"] for f in pr_files]",
+    "file_path": "github_app/github_helper/pull_requests.py",
+    "start_line": 199,
+    "end_line": 200,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 4
+  },
+  {
+    "category": "Security Vulnerability",
+    "description": "Potential exposure of sensitive data in logs.",
+    "impact": "critical",
+    "rationale": "Logging sensitive information, such as access tokens or personal data, can lead to security vulnerabilities. It's essential to sanitize logs to avoid exposing sensitive data.",
+    "recommendation": "Review logging statements to ensure sensitive data is not logged. Implement logging best practices.",
+    "current_code": "logger.debug(f\"Post Review comment response:{response.text}\")",
+    "suggested_code": "logger.debug(\"Post Review comment response received.\")",
+    "file_path": "github_app/github_helper/pull_requests.py",
+    "start_line": 181,
+    "end_line": 182,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 9
+  },
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/review.md b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/review.md
new file mode 100644
index 00000000..d25b577b
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/review.md
@@ -0,0 +1,72 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/476
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 6
+- Critical: 2
+- Important: 0
+- Minor: 0
+- Files Affected: 2
+## 🏆 Code Quality
+[███████████████░░░░░] 75% (Fair)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Security Vulnerability (2 issues)</strong></summary>
+
+### 1. Potential exposure of sensitive data in logs.
+📁 **File:** `github_app/github_helper/pull_requests.py:181`
+⚖️ **Severity:** 9/10
+🔍 **Description:** Potential exposure of sensitive data in logs.
+💡 **Solution:** 
+
+**Current Code:**
+```python
+logger.debug(f"Post Review comment response:{response.text}")
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 2. Changes made to sensitive file
+📁 **File:** `config.json:1`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 4256, "completion_tokens": 998, "total_tokens": 5254}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/comments.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/comments.json
new file mode 100644
index 00000000..b3f4d9f3
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/comments.json
@@ -0,0 +1,17 @@
+[
+  {
+    "category": "Division by Zero Risk",
+    "description": "Potential division by zero if total_tokens is zero.",
+    "impact": "critical",
+    "rationale": "Dividing by zero will cause a runtime error, leading to application crashes.",
+    "recommendation": "Add a check to ensure total_tokens is not zero before performing division.",
+    "current_code": "print(f\"Total tokens used:{total_tokens:,}\")",
+    "suggested_code": "if total_tokens > 0: print(f\"Total tokens used:{total_tokens:,}\")",
+    "file_path": "main.py",
+    "start_line": 159,
+    "end_line": 159,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 9
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/issues.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/issues.json
new file mode 100644
index 00000000..3f2cd28f
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/issues.json
@@ -0,0 +1,107 @@
+[
+  {
+    "category": "Unused Import",
+    "description": "Unused import statement for 'random'.",
+    "impact": "low",
+    "rationale": "Having unused imports can lead to confusion and clutter in the codebase.",
+    "recommendation": "Remove the unused import statement.",
+    "current_code": "import random  # Unused import",
+    "suggested_code": "",
+    "file_path": "main.py",
+    "start_line": 8,
+    "end_line": 8,
+    "change_type": "deletion",
+    "sentiment": "neutral",
+    "severity": 2
+  },
+  {
+    "category": "Error Handling",
+    "description": "Potential failure of API call without retry mechanism.",
+    "impact": "high",
+    "rationale": "If the API call fails, the application may crash or return unexpected results without a retry mechanism.",
+    "recommendation": "Implement a retry mechanism for the API call.",
+    "current_code": "response = completion(",
+    "suggested_code": "# Implement retry logic here",
+    "file_path": "main.py",
+    "start_line": 66,
+    "end_line": 66,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Silent Failure",
+    "description": "Silent failure in JSON parsing without logging.",
+    "impact": "high",
+    "rationale": "Failing to log errors can make debugging difficult and lead to unnoticed issues.",
+    "recommendation": "Add logging for the exception to capture the error details.",
+    "current_code": "print(f\"Failed to parse content for applicant\")",
+    "suggested_code": "print(f\"Failed to parse content for applicant:{e}\")",
+    "file_path": "main.py",
+    "start_line": 86,
+    "end_line": 86,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Inefficient Progress Reporting",
+    "description": "Inefficient way to print progress during applicant processing.",
+    "impact": "medium",
+    "rationale": "Using print statements in a loop can lead to performance issues and cluttered output.",
+    "recommendation": "Consider using a progress bar or logging framework for better performance and readability.",
+    "current_code": "print(f\"\\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}\", end='', flush=True)",
+    "suggested_code": "# Use a logging framework or a proper progress bar",
+    "file_path": "main.py",
+    "start_line": 121,
+    "end_line": 121,
+    "change_type": "modification",
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Redundant Code",
+    "description": "Redundant check for empty DataFrame before processing.",
+    "impact": "low",
+    "rationale": "The check for an empty DataFrame is unnecessary since processing will naturally handle it.",
+    "recommendation": "Remove the redundant check for empty DataFrame.",
+    "current_code": "if len(df) == 0: return",
+    "suggested_code": "",
+    "file_path": "main.py",
+    "start_line": 142,
+    "end_line": 143,
+    "change_type": "deletion",
+    "sentiment": "neutral",
+    "severity": 3
+  },
+  {
+    "category": "Division by Zero Risk",
+    "description": "Potential division by zero if total_tokens is zero.",
+    "impact": "critical",
+    "rationale": "Dividing by zero will cause a runtime error, leading to application crashes.",
+    "recommendation": "Add a check to ensure total_tokens is not zero before performing division.",
+    "current_code": "print(f\"Total tokens used:{total_tokens:,}\")",
+    "suggested_code": "if total_tokens > 0: print(f\"Total tokens used:{total_tokens:,}\")",
+    "file_path": "main.py",
+    "start_line": 159,
+    "end_line": 159,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 9
+  },
+  {
+    "category": "File Handling",
+    "description": "No error handling for file not found when running main function.",
+    "impact": "high",
+    "rationale": "If the file does not exist, the application will crash without providing user feedback.",
+    "recommendation": "Add error handling to manage file not found exceptions.",
+    "current_code": "main(input_file)",
+    "suggested_code": "try: main(input_file) except FileNotFoundError: print('File not found. Please check the file name.')",
+    "file_path": "main.py",
+    "start_line": 175,
+    "end_line": 175,
+    "change_type": "modification",
+    "sentiment": "negative",
+    "severity": 8
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_335/review.md b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/review.md
similarity index 52%
rename from .experiments/code_review/llama-405b/no_eval/pr_335/review.md
rename to .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/review.md
index 2a15ce48..6bd79ffb 100644
--- a/.experiments/code_review/llama-405b/no_eval/pr_335/review.md
+++ b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/review.md
@@ -1,36 +1,32 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/335
+PR URL: https://github.com/sauravpanda/applicant-screening/pull/5
 
 # 🔍 Code Review Summary
 
-✅ **All Clear:** This commit looks good! 👍
+❗ **Attention Required:** This push has potential issues. 🚨
 
 ## 📊 Stats
-- Total Issues: 3
-- Critical: 0
+- Total Issues: 7
+- Critical: 1
 - Important: 0
-- Minor: 1
-- Files Affected: 2
+- Minor: 0
+- Files Affected: 1
 ## 🏆 Code Quality
-[████████████████░░░░] 80% (Good)
+[███████████████░░░░░] 75% (Fair)
 
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (2 issues)</strong></summary>
+## 🚨 Critical Issues
 
 <details>
-<summary><strong>Import Statements (1 issues)</strong></summary>
+<summary><strong>Division by Zero Risk (1 issues)</strong></summary>
 
-### 1. Unused import statements are present in the code.
-📁 **File:** `kaizen/generator/pr_description.py:6`
-⚖️ **Severity:** 5/10
-🔍 **Description:** The import statements for 'output' and 'parser' are not used anywhere in the code.
-💡 **Solution:** Remove unused import statements to declutter the code.
+### 1. Potential division by zero if total_tokens is zero.
+📁 **File:** `main.py:159`
+⚖️ **Severity:** 9/10
+🔍 **Description:** Potential division by zero if total_tokens is zero.
+💡 **Solution:** 
 
 **Current Code:**
 ```python
-from kaizen.helpers import output, parser
+print(f"Total tokens used:{total_tokens:,}")
 ```
 
 **Suggested Code:**
@@ -40,6 +36,12 @@ from kaizen.helpers import output, parser
 
 </details>
 
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (2 issues)</strong></summary>
+
 </details>
 
 ---
@@ -58,5 +60,5 @@ from kaizen.helpers import output, parser
 </details>
 
 
------ Cost Usage (azure_ai/Meta-Llama-3-405B-Instruct)
-{"prompt_tokens": 6756, "completion_tokens": 533, "total_tokens": 7289}
\ No newline at end of file
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 6403, "completion_tokens": 1126, "total_tokens": 7529}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/comments.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/comments.json
new file mode 100644
index 00000000..6f3e44d6
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/comments.json
@@ -0,0 +1,100 @@
+[
+  {
+    "category": "Security",
+    "description": "Potential security risk with hardcoded API keys.",
+    "impact": "critical",
+    "rationale": "Hardcoding sensitive information like API keys can lead to security vulnerabilities if the code is exposed. It is essential to use environment variables or secure vaults.",
+    "recommendation": "Replace hardcoded API keys with environment variables and ensure they are not included in version control.",
+    "current_code": "\"api_key\": \"os.environ/AZURE_API_KEY\",",
+    "suggested_code": "\"api_key\": os.getenv('AZURE_API_KEY'),",
+    "file_path": "config.json",
+    "start_line": 13,
+    "end_line": 13,
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Security",
+    "description": "Potential SQL injection risk in dynamic queries.",
+    "impact": "critical",
+    "rationale": "The use of string interpolation in SQL queries can expose the application to SQL injection attacks. Although parameterized queries are used in some places, there are instances where string interpolation is still present.",
+    "recommendation": "Always use parameterized queries instead of string interpolation to prevent SQL injection vulnerabilities.",
+    "current_code": "query = text(f\"SELECT ... WHERE node_content LIKE '%{caller}%' \")",
+    "suggested_code": "query = text(\"SELECT ... WHERE node_content LIKE :caller\").bindparams(caller=f'%{caller}%')",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 303,
+    "end_line": 303,
+    "sentiment": "negative",
+    "severity": 9
+  },
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Docker",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to Dockerfile, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_path": "Dockerfile",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Docker",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to docker-compose.yml, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "15",
+    "end_line": "15",
+    "side": "RIGHT",
+    "file_path": "docker-compose.yml",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Version Control",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to .gitignore, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "164",
+    "end_line": "164",
+    "side": "RIGHT",
+    "file_path": ".gitignore",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Database",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to db_setup/init.sql, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_path": "db_setup/init.sql",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/issues.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/issues.json
new file mode 100644
index 00000000..1a938a3d
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/issues.json
@@ -0,0 +1,282 @@
+[
+  {
+    "category": "Documentation",
+    "description": "Lack of comments and documentation in several files.",
+    "impact": "high",
+    "rationale": "While the code is functional, it lacks sufficient comments and documentation, which can make it difficult for other developers to understand the purpose and functionality of various components.",
+    "recommendation": "Add comments to explain the purpose of functions, classes, and complex logic. Consider using docstrings for public methods and classes.",
+    "current_code": "def chunk_code(code: str, language: str) -> ParsedBody:",
+    "suggested_code": "def chunk_code(code: str, language: str) -> ParsedBody:\n    \"\"\"\n    Chunks the provided code into functions, classes, and other blocks based on the specified language.\n    \"\"\"",
+    "file_path": "kaizen/retriever/code_chunker.py",
+    "start_line": 7,
+    "end_line": 8,
+    "sentiment": "neutral",
+    "severity": 7
+  },
+  {
+    "category": "Error Handling",
+    "description": "Insufficient error handling in database interactions.",
+    "impact": "high",
+    "rationale": "The code interacts with the database without any error handling, which could lead to unhandled exceptions and application crashes.",
+    "recommendation": "Implement try-except blocks around database operations to handle potential exceptions gracefully.",
+    "current_code": "cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))",
+    "suggested_code": "try:\n    cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))\nexcept Exception as e:\n    # Log the error and handle it appropriately\n    print(f\"Database error:{e}\")",
+    "file_path": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 41,
+    "end_line": 41,
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Security",
+    "description": "Potential security risk with hardcoded API keys.",
+    "impact": "critical",
+    "rationale": "Hardcoding sensitive information like API keys can lead to security vulnerabilities if the code is exposed. It is essential to use environment variables or secure vaults.",
+    "recommendation": "Replace hardcoded API keys with environment variables and ensure they are not included in version control.",
+    "current_code": "\"api_key\": \"os.environ/AZURE_API_KEY\",",
+    "suggested_code": "\"api_key\": os.getenv('AZURE_API_KEY'),",
+    "file_path": "config.json",
+    "start_line": 13,
+    "end_line": 13,
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Performance",
+    "description": "Inefficient handling of large code chunks.",
+    "impact": "medium",
+    "rationale": "The current implementation may not efficiently handle large code inputs, leading to performance degradation.",
+    "recommendation": "Consider implementing a streaming approach or chunking the input code to process it in smaller parts.",
+    "current_code": "def chunk_code(code: str, language: str) -> ParsedBody:",
+    "suggested_code": "# Consider breaking down the input code into smaller chunks before processing.",
+    "file_path": "kaizen/retriever/code_chunker.py",
+    "start_line": 7,
+    "end_line": 7,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Code Structure",
+    "description": "Inconsistent naming conventions.",
+    "impact": "medium",
+    "rationale": "Inconsistent naming can lead to confusion and reduce code readability. Following a consistent naming convention improves maintainability.",
+    "recommendation": "Standardize naming conventions across the codebase (e.g., snake_case for variables and functions).",
+    "current_code": "def is_react_hook(name: str) -> bool:",
+    "suggested_code": "def is_react_hook(name: str) -> bool:",
+    "file_path": "kaizen/retriever/code_chunker.py",
+    "start_line": 65,
+    "end_line": 66,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling",
+    "description": "Lack of specific error handling in critical database operations.",
+    "impact": "high",
+    "rationale": "The code currently raises a generic exception in the `generate_abstraction` method without providing context or handling specific error types. This can lead to unhandled exceptions that crash the application.",
+    "recommendation": "Implement specific exception handling to provide more informative error messages and handle different types of exceptions appropriately.",
+    "current_code": "except Exception as e:\n    raise e",
+    "suggested_code": "except (ValueError, TypeError) as e:\n    logger.error(f\"Error generating abstraction:{str(e)}\")\n    raise CustomAbstractionError(f\"Failed to generate abstraction for code block:{code_block}\")",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 218,
+    "end_line": 219,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Code Readability",
+    "description": "Inconsistent logging levels and messages.",
+    "impact": "medium",
+    "rationale": "Using different logging levels inconsistently can make it difficult to track the application's behavior. For example, using `logger.debug` for important events like initialization can obscure critical information.",
+    "recommendation": "Use appropriate logging levels consistently. For instance, use `logger.info` for initialization messages and `logger.debug` for detailed debugging information.",
+    "current_code": "logger.debug(f\"Successfully parsed file:{file_path}\")",
+    "suggested_code": "logger.info(f\"Successfully parsed file:{file_path}\")",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 107,
+    "end_line": 107,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Performance",
+    "description": "Potential performance issues with database operations.",
+    "impact": "high",
+    "rationale": "The current implementation performs multiple database queries within a loop, which can lead to performance bottlenecks, especially with a large number of files or functions.",
+    "recommendation": "Batch database operations where possible to reduce the number of queries. Consider using bulk inserts or updates to improve performance.",
+    "current_code": "connection.execute(query,{...})",
+    "suggested_code": "with connection.begin() as transaction:\n    # Collect all queries and execute them in a batch\n    transaction.execute(...)",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 298,
+    "end_line": 312,
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Security",
+    "description": "Potential SQL injection risk in dynamic queries.",
+    "impact": "critical",
+    "rationale": "The use of string interpolation in SQL queries can expose the application to SQL injection attacks. Although parameterized queries are used in some places, there are instances where string interpolation is still present.",
+    "recommendation": "Always use parameterized queries instead of string interpolation to prevent SQL injection vulnerabilities.",
+    "current_code": "query = text(f\"SELECT ... WHERE node_content LIKE '%{caller}%' \")",
+    "suggested_code": "query = text(\"SELECT ... WHERE node_content LIKE :caller\").bindparams(caller=f'%{caller}%')",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 303,
+    "end_line": 303,
+    "sentiment": "negative",
+    "severity": 9
+  },
+  {
+    "category": "Maintainability",
+    "description": "Hardcoded values for database connection parameters.",
+    "impact": "medium",
+    "rationale": "Hardcoding values like `embed_dim=1536` makes the code less flexible and harder to maintain. If these values need to change, it requires code modification.",
+    "recommendation": "Consider using configuration files or environment variables to manage such parameters, making the code more adaptable.",
+    "current_code": "embed_dim=1536",
+    "suggested_code": "embed_dim=int(os.getenv('EMBED_DIM', 1536))",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 49,
+    "end_line": 49,
+    "sentiment": "neutral",
+    "severity": 4
+  },
+  {
+    "category": "Error Handling",
+    "description": "Broad Exception Catching",
+    "impact": "high",
+    "rationale": "Catching general exceptions can obscure the root cause of errors and make debugging difficult. Specific exceptions should be caught to provide clearer error handling.",
+    "recommendation": "Catch specific exceptions instead of using a general 'Exception'. This will help in identifying the exact issue and provide more meaningful error messages.",
+    "current_code": "except Exception as e:",
+    "suggested_code": "except (ImportError, ValueError) as e:",
+    "file_path": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 28,
+    "end_line": 30,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Logging",
+    "description": "Inconsistent Logging Levels",
+    "impact": "medium",
+    "rationale": "Using different logging levels (error, warning, info) inconsistently can lead to confusion about the severity of issues. It's important to use appropriate logging levels for different situations.",
+    "recommendation": "Use logging levels consistently. For example, use 'logger.warning' for recoverable issues and 'logger.error' for critical failures.",
+    "current_code": "logger.error(f\"Failed to load language{language}:{str(e)}\")",
+    "suggested_code": "logger.warning(f\"Failed to load language{language}:{str(e)}\")",
+    "file_path": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 108,
+    "end_line": 108,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Code Readability",
+    "description": "Complex Conditional Logic",
+    "impact": "medium",
+    "rationale": "The conditional logic in the `traverse_tree` function is complex and could be simplified for better readability and maintainability.",
+    "recommendation": "Consider refactoring the conditional logic into separate functions or using a dictionary mapping to improve clarity.",
+    "current_code": "if node.type in[\"function_definition\", \"function_declaration\", ...]:",
+    "suggested_code": "def is_function(node): return node.type in[\"function_definition\", \"function_declaration\", ...]\n\nif is_function(node):",
+    "file_path": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 48,
+    "end_line": 88,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Performance",
+    "description": "Potential Memory Leak with Caching",
+    "impact": "medium",
+    "rationale": "Using `lru_cache` without a size limit can lead to excessive memory usage if many different languages are loaded. This could affect performance over time.",
+    "recommendation": "Consider setting a reasonable `maxsize` for `lru_cache` to prevent potential memory issues.",
+    "current_code": "@lru_cache(maxsize=None)",
+    "suggested_code": "@lru_cache(maxsize=128)",
+    "file_path": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 14,
+    "end_line": 14,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Unit Testing",
+    "description": "Lack of Unit Tests for New Functionality",
+    "impact": "high",
+    "rationale": "The new functionalities introduced lack corresponding unit tests, which are crucial for verifying correctness and preventing regressions.",
+    "recommendation": "Add unit tests for the new functions, particularly for `load_language` and `get_parser`, to ensure they behave as expected.",
+    "current_code": "No tests present for new functionality.",
+    "suggested_code": "def test_load_language():\n    assert load_language('tree-sitter-python') is not None",
+    "file_path": "tests/retriever/test_chunker.py",
+    "start_line": 1,
+    "end_line": 1,
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Docker",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to Dockerfile, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_path": "Dockerfile",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Docker",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to docker-compose.yml, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "15",
+    "end_line": "15",
+    "side": "RIGHT",
+    "file_path": "docker-compose.yml",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Version Control",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to .gitignore, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "164",
+    "end_line": "164",
+    "side": "RIGHT",
+    "file_path": ".gitignore",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Database",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to db_setup/init.sql, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_path": "db_setup/init.sql",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/review.md b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/review.md
new file mode 100644
index 00000000..37e9cbc5
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/review.md
@@ -0,0 +1,152 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/222
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 20
+- Critical: 7
+- Important: 0
+- Minor: 0
+- Files Affected: 10
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Security (7 issues)</strong></summary>
+
+### 1. Potential security risk with hardcoded API keys.
+📁 **File:** `config.json:13`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Potential security risk with hardcoded API keys.
+💡 **Solution:** 
+
+**Current Code:**
+```python
+"api_key": "os.environ/AZURE_API_KEY",
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 2. Potential SQL injection risk in dynamic queries.
+📁 **File:** `kaizen/retriever/llama_index_retriever.py:303`
+⚖️ **Severity:** 9/10
+🔍 **Description:** Potential SQL injection risk in dynamic queries.
+💡 **Solution:** 
+
+**Current Code:**
+```python
+query = text(f"SELECT ... WHERE node_content LIKE '%{caller}%' ")
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 3. Changes made to sensitive file
+📁 **File:** `config.json:4`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 4. Changes made to sensitive file
+📁 **File:** `Dockerfile:4`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 5. Changes made to sensitive file
+📁 **File:** `docker-compose.yml:15`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 6. Changes made to sensitive file
+📁 **File:** `.gitignore:164`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 7. Changes made to sensitive file
+📁 **File:** `db_setup/init.sql:1`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 19847, "completion_tokens": 2900, "total_tokens": 22747}
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_440/comments.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/comments.json
similarity index 100%
rename from .experiments/code_review/llama-405b/no_eval/pr_440/comments.json
rename to .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/comments.json
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/issues.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/issues.json
new file mode 100644
index 00000000..21f1de13
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/issues.json
@@ -0,0 +1,198 @@
+[
+  {
+    "category": "Code Clarity",
+    "description": "Redundant imports and unused variables.",
+    "impact": "medium",
+    "rationale": "Removing unused imports and variables enhances code readability and reduces confusion for future maintainers. It also slightly improves performance by reducing the bundle size.",
+    "recommendation": "Audit imports and remove any that are not used in the file.",
+    "current_code": "import{MemoriesIcon, NextIcon, SearchIcon, UrlIcon}from \"@repo/ui/icons\";",
+    "suggested_code": "import{MemoriesIcon, NextIcon, UrlIcon}from \"@repo/ui/icons\";",
+    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 4,
+    "end_line": 4,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling",
+    "description": "Lack of error handling in async functions.",
+    "impact": "high",
+    "rationale": "The `handleDeleteSpace` function does not handle potential errors from the `deleteSpace` call, which could lead to unhandled promise rejections and a poor user experience.",
+    "recommendation": "Add a try-catch block around the async call to handle errors gracefully.",
+    "current_code": "const handleDeleteSpace = async (id: number) =>{const response = await deleteSpace(id); ...};",
+    "suggested_code": "const handleDeleteSpace = async (id: number) =>{try{const response = await deleteSpace(id); ...}catch (error){toast.error(\"An error occurred while deleting the space\");}};",
+    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 66,
+    "end_line": 76,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Performance",
+    "description": "Inefficient state updates in `handleDeleteSpace`.",
+    "impact": "medium",
+    "rationale": "Using `spaces.filter` creates a new array on every delete, which can be inefficient. Consider using functional updates to update state based on the previous state.",
+    "recommendation": "Use the functional form of `setSpaces` to avoid unnecessary re-renders.",
+    "current_code": "setSpaces(spaces.filter((space) => space.id !== id));",
+    "suggested_code": "setSpaces((prevSpaces) => prevSpaces.filter((space) => space.id !== id));",
+    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 70,
+    "end_line": 70,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Code Consistency",
+    "description": "Inconsistent naming conventions for components.",
+    "impact": "medium",
+    "rationale": "Inconsistent naming can lead to confusion. For instance, `LinkComponent` and `MemoryComponent` should follow a consistent naming pattern.",
+    "recommendation": "Rename components to follow a consistent naming convention, such as `MemoryLink` and `SpaceLink`.",
+    "current_code": "function LinkComponent({...}){...}",
+    "suggested_code": "function MemoryLink({...}){...}",
+    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 321,
+    "end_line": 321,
+    "sentiment": "neutral",
+    "severity": 4
+  },
+  {
+    "category": "User Experience",
+    "description": "Lack of feedback for loading states.",
+    "impact": "high",
+    "rationale": "Users should receive feedback when actions are being processed, such as during the deletion of a space. This improves user experience and prevents confusion.",
+    "recommendation": "Implement a loading state for actions that involve network requests.",
+    "current_code": "const handleDeleteSpace = async (id: number) =>{...};",
+    "suggested_code": "const[loading, setLoading] = useState(false); const handleDeleteSpace = async (id: number) =>{setLoading(true); ... setLoading(false);};",
+    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 66,
+    "end_line": 76,
+    "sentiment": "negative",
+    "severity": 6
+  },
+  {
+    "category": "Code Structure",
+    "description": "Excessive nesting in the Menu component.",
+    "impact": "high",
+    "rationale": "Deeply nested components can make the code harder to read and maintain. It can also lead to performance issues if not managed properly.",
+    "recommendation": "Consider breaking down the Menu component into smaller, reusable components to improve readability and maintainability.",
+    "current_code": "function Menu(){...}",
+    "suggested_code": "function Menu(){return (<DialogContentContainer setDialogClose={() => setDialogOpen(false)}/>);}",
+    "file_path": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 163,
+    "end_line": 384,
+    "sentiment": "neutral",
+    "severity": 7
+  },
+  {
+    "category": "Error Handling",
+    "description": "Error handling in async functions is insufficient.",
+    "impact": "high",
+    "rationale": "Errors in asynchronous operations are not being caught, which could lead to unhandled promise rejections and application crashes.",
+    "recommendation": "Wrap async calls in try-catch blocks to handle errors gracefully.",
+    "current_code": "const handleSubmit = async (content?: string, spaces?: number[]) =>{...};",
+    "suggested_code": "const handleSubmit = async (content?: string, spaces?: number[]) =>{try{...}catch (error){toast.error(`Error: ${error.message}`);}};",
+    "file_path": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 213,
+    "end_line": 230,
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Performance",
+    "description": "Potential performance issues with useEffect dependency.",
+    "impact": "medium",
+    "rationale": "The useEffect hook does not specify dependencies, which may lead to unnecessary re-renders and API calls.",
+    "recommendation": "Add dependencies to the useEffect hook to control when it runs.",
+    "current_code": "useEffect(() =>{...},[]);",
+    "suggested_code": "useEffect(() =>{...},[setSpaces]);",
+    "file_path": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 38,
+    "end_line": 51,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Readability",
+    "description": "Inconsistent naming conventions.",
+    "impact": "medium",
+    "rationale": "Inconsistent naming can lead to confusion and reduce code readability. For instance, the variable 'spaces' is used both as state and in the function.",
+    "recommendation": "Use more descriptive names for variables and functions to improve clarity.",
+    "current_code": "let spaces = await getSpaces();",
+    "suggested_code": "let fetchedSpaces = await getSpaces();",
+    "file_path": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 40,
+    "end_line": 40,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Security",
+    "description": "Potential security vulnerability with user input.",
+    "impact": "high",
+    "rationale": "User input is not being sanitized before being processed, which could lead to XSS attacks or other vulnerabilities.",
+    "recommendation": "Sanitize user input before processing it to prevent security vulnerabilities.",
+    "current_code": "const content = e.get('content')?.toString();",
+    "suggested_code": "const content = sanitizeInput(e.get('content')?.toString());",
+    "file_path": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 238,
+    "end_line": 238,
+    "sentiment": "negative",
+    "severity": 9
+  },
+  {
+    "category": "Code Readability",
+    "description": "Inconsistent use of destructuring in function parameters.",
+    "impact": "medium",
+    "rationale": "The function `ComboboxWithCreate` uses destructuring for some props but not for others, which can lead to confusion and inconsistency in the codebase.",
+    "recommendation": "Consistently use destructuring for all props in the function parameters.",
+    "current_code": "const ComboboxWithCreate = ({options, onSelect, onSubmit, selectedSpaces, setSelectedSpaces}: ComboboxWithCreateProps) =>{",
+    "suggested_code": "const ComboboxWithCreate: React.FC<ComboboxWithCreateProps> = ({options, onSelect, onSubmit, selectedSpaces, setSelectedSpaces}) =>{",
+    "file_path": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 32,
+    "end_line": 38,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling",
+    "description": "Potential for undefined values in `options.find`.",
+    "impact": "high",
+    "rationale": "Using `?.` operator without a fallback can lead to rendering issues if the option is not found. This could result in a runtime error if the component expects a valid label.",
+    "recommendation": "Provide a fallback value in case the option is not found.",
+    "current_code": "options.find((opt) => opt.value === spaceId.toString())?.label",
+    "suggested_code": "options.find((opt) => opt.value === spaceId.toString())?.label || 'Unknown Space'",
+    "file_path": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 74,
+    "end_line": 74,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Performance",
+    "description": "Unnecessary re-renders due to state updates in `handleKeyDown`.",
+    "impact": "medium",
+    "rationale": "The `handleKeyDown` function updates state on every key press, which can lead to performance issues if the component re-renders frequently.",
+    "recommendation": "Consider using a debouncing technique to limit the frequency of state updates.",
+    "current_code": "setSelectedSpaces((prev) => prev.slice(0, -1));",
+    "suggested_code": "// Implement debouncing logic here",
+    "file_path": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 51,
+    "end_line": 51,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Code Maintainability",
+    "description": "Hardcoded strings in JSX can lead to maintenance issues.",
+    "impact": "medium",
+    "rationale": "Hardcoded strings can make localization and future changes more difficult. It is better to define them as constants or use a localization library.",
+    "recommendation": "Define hardcoded strings as constants at the top of the file.",
+    "current_code": "placeholder=\"Select or create a new space.\"",
+    "suggested_code": "const PLACEHOLDER_TEXT = 'Select or create a new space.'; // Use this constant in JSX",
+    "file_path": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 82,
+    "end_line": 82,
+    "sentiment": "neutral",
+    "severity": 4
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/review.md b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/review.md
new file mode 100644
index 00000000..12226340
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/review.md
@@ -0,0 +1,33 @@
+PR URL: https://github.com/supermemoryai/supermemory/pull/232
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 14
+- Critical: 0
+- Important: 0
+- Minor: 0
+- Files Affected: 3
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 14992, "completion_tokens": 2670, "total_tokens": 17662}
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_232/comments.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/comments.json
similarity index 100%
rename from .experiments/code_review/sonnet-3.5/no_eval/pr_232/comments.json
rename to .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/comments.json
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/issues.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/issues.json
new file mode 100644
index 00000000..b1b25dc2
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/issues.json
@@ -0,0 +1,58 @@
+[
+  {
+    "category": "Code Readability",
+    "description": "Inconsistent formatting in function calls.",
+    "impact": "medium",
+    "rationale": "Consistency in function call formatting enhances readability and maintainability. The change to break the function call into multiple lines improves clarity but should be applied uniformly across the codebase.",
+    "recommendation": "Consider applying consistent formatting for function calls throughout the codebase.",
+    "current_code": "linkedin_post = work_summary_generator.generate_linkedin_post(summary, user=\"oss_example\")",
+    "suggested_code": "linkedin_post = work_summary_generator.generate_linkedin_post(\n    summary, user=\"oss_example\"\n)",
+    "file_path": "examples/work_summarizer/main.py",
+    "start_line": 60,
+    "end_line": 61,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling",
+    "description": "Lack of error handling in JSON parsing.",
+    "impact": "high",
+    "rationale": "The code does not handle potential exceptions that may arise from JSON parsing, which could lead to runtime errors and application crashes.",
+    "recommendation": "Wrap the JSON parsing code in a try-except block to handle potential exceptions gracefully.",
+    "current_code": "parsed_data = json.loads(json_data)",
+    "suggested_code": "try:\n    parsed_data = json.loads(json_data)\nexcept json.JSONDecodeError as e:\n    logging.error(f\"JSON decoding failed:{e}\")\n    return{}",
+    "file_path": "kaizen/helpers/parser.py",
+    "start_line": 47,
+    "end_line": 48,
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Code Efficiency",
+    "description": "Redundant calls to extract markdown content.",
+    "impact": "medium",
+    "rationale": "The extraction of markdown content is done twice in the new methods, which can lead to code duplication and potential inconsistencies.",
+    "recommendation": "Consider refactoring the markdown extraction into a separate utility function to avoid code duplication.",
+    "current_code": "return parser.extract_markdown_content(response)",
+    "suggested_code": "return extract_markdown_content(response)",
+    "file_path": "kaizen/reviewer/work_summarizer.py",
+    "start_line": 65,
+    "end_line": 65,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Documentation",
+    "description": "Missing docstrings for new methods.",
+    "impact": "medium",
+    "rationale": "Newly added methods lack docstrings, making it difficult for other developers to understand their purpose and usage.",
+    "recommendation": "Add docstrings to the new methods to describe their functionality and parameters.",
+    "current_code": "def generate_twitter_post(self, summary: Dict, user: Optional[str] = None) -> str:",
+    "suggested_code": "def generate_twitter_post(self, summary: Dict, user: Optional[str] = None) -> str:\n    \"\"\"Generate a Twitter post based on the given summary.\"\"\"",
+    "file_path": "kaizen/reviewer/work_summarizer.py",
+    "start_line": 58,
+    "end_line": 58,
+    "sentiment": "neutral",
+    "severity": 4
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/review.md b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/review.md
new file mode 100644
index 00000000..63a8b589
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/review.md
@@ -0,0 +1,33 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/252
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 4
+- Critical: 0
+- Important: 0
+- Minor: 0
+- Files Affected: 3
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 3585, "completion_tokens": 758, "total_tokens": 4343}
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_252/comments.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/comments.json
similarity index 100%
rename from .experiments/code_review/sonnet-3.5/no_eval/pr_252/comments.json
rename to .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/comments.json
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/issues.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/issues.json
new file mode 100644
index 00000000..0935869f
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/issues.json
@@ -0,0 +1,72 @@
+[
+  {
+    "category": "Code Consistency",
+    "description": "Inconsistent naming conventions for prompts.",
+    "impact": "medium",
+    "rationale": "The naming of the new prompts does not follow the established naming conventions seen in the previous code, which can lead to confusion and inconsistency.",
+    "recommendation": "Ensure that all prompt variables follow a consistent naming pattern (e.g., using underscores for separation).",
+    "current_code": "PR_DESCRIPTION_SYSTEM_PROMPT",
+    "suggested_code": "PR_DESC_SYSTEM_PROMPT",
+    "file_path": "kaizen/llms/prompts/pr_desc_prompts.py",
+    "start_line": 1,
+    "end_line": 3,
+    "sentiment": "negative",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling",
+    "description": "Lack of error handling in prompt processing.",
+    "impact": "high",
+    "rationale": "The code does not handle potential errors when processing prompts, which could lead to unhandled exceptions during runtime.",
+    "recommendation": "Implement try-except blocks around critical sections of code that process prompts to ensure graceful error handling.",
+    "current_code": "desc = self._process_full_diff(prompt, user)",
+    "suggested_code": "try:\n    desc = self._process_full_diff(prompt, user)\nexcept Exception as e:\n    self.logger.error(f'Error processing full diff:{e}')\n    return 'Error processing request.'",
+    "file_path": "kaizen/generator/pr_description.py",
+    "start_line": 51,
+    "end_line": 55,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Redundant Code",
+    "description": "Redundant parameter in method signatures.",
+    "impact": "medium",
+    "rationale": "The parameter 'reeval_response' is removed in some methods but still appears in others where it is not used, leading to unnecessary complexity.",
+    "recommendation": "Remove unused parameters from method signatures to enhance clarity and maintainability.",
+    "current_code": "def _process_full_diff(self, prompt: str, user: Optional[str], reeval_response: bool) -> str:",
+    "suggested_code": "def _process_full_diff(self, prompt: str, user: Optional[str]) -> str:",
+    "file_path": "kaizen/generator/pr_description.py",
+    "start_line": 40,
+    "end_line": 41,
+    "sentiment": "neutral",
+    "severity": 4
+  },
+  {
+    "category": "Documentation",
+    "description": "Insufficient comments and documentation for new methods.",
+    "impact": "medium",
+    "rationale": "New methods lack docstrings, which makes it difficult for other developers to understand their purpose and usage.",
+    "recommendation": "Add docstrings to all new methods to describe their functionality, parameters, and return values.",
+    "current_code": "def _process_files(self, pull_request_files: List[Dict], pull_request_title: str, pull_request_desc: str, user: Optional[str]):",
+    "suggested_code": "def _process_files(self, pull_request_files: List[Dict], pull_request_title: str, pull_request_desc: str, user: Optional[str]):\n    \"\"\"\n    Process the provided pull request files and generate descriptions.\n    \n    Args:\n        pull_request_files (List[Dict]): List of pull request files.\n        pull_request_title (str): Title of the pull request.\n        pull_request_desc (str): Description of the pull request.\n        user (Optional[str]): User requesting the review.\n    \"\"\"",
+    "file_path": "kaizen/generator/pr_description.py",
+    "start_line": 85,
+    "end_line": 86,
+    "sentiment": "neutral",
+    "severity": 3
+  },
+  {
+    "category": "Performance",
+    "description": "Potential performance issue with JSON handling.",
+    "impact": "medium",
+    "rationale": "Using json.dumps in a loop can lead to performance degradation when processing large datasets.",
+    "recommendation": "Consider using a more efficient method for handling JSON, such as building the dictionary first and then dumping it once.",
+    "current_code": "prompt = MERGE_PR_DESCRIPTION_PROMPT.format(DESCS=json.dumps(file_descs))",
+    "suggested_code": "descs_dict ={'descs': file_descs}\nprompt = MERGE_PR_DESCRIPTION_PROMPT.format(DESCS=json.dumps(descs_dict))",
+    "file_path": "kaizen/generator/pr_description.py",
+    "start_line": 109,
+    "end_line": 110,
+    "sentiment": "neutral",
+    "severity": 6
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/review.md b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/review.md
new file mode 100644
index 00000000..23ae6698
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/review.md
@@ -0,0 +1,33 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/335
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 5
+- Critical: 0
+- Important: 0
+- Minor: 0
+- Files Affected: 2
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 6450, "completion_tokens": 1055, "total_tokens": 7505}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/comments.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/comments.json
new file mode 100644
index 00000000..1ceabeef
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/comments.json
@@ -0,0 +1,16 @@
+[
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "11",
+    "end_line": "11",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/issues.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/issues.json
new file mode 100644
index 00000000..207b04b8
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/issues.json
@@ -0,0 +1,380 @@
+[
+  {
+    "category": "Security",
+    "description": "Sensitive information exposure in .env.example",
+    "impact": "high",
+    "rationale": "The addition of LITELLM_LOG in the .env.example file may expose sensitive logging configurations if this file is shared or pushed to public repositories.",
+    "recommendation": "Consider using a more generic logging level or provide documentation on how to set sensitive configurations without exposing them in version control.",
+    "current_code": "LITELLM_LOG=\"ERROR\"",
+    "suggested_code": "# LITELLM_LOG=\"ERROR\" # Set in production environment only",
+    "file_path": ".env.example",
+    "start_line": 6,
+    "end_line": 6,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Readability",
+    "description": "Inconsistent formatting in test_create_pr_description.py",
+    "impact": "medium",
+    "rationale": "The formatting of the DESC_COLLAPSIBLE_TEMPLATE string is inconsistent with the rest of the code, which can lead to confusion and reduce readability.",
+    "recommendation": "Ensure consistent formatting across all string definitions, especially for multiline strings.",
+    "current_code": "DESC_COLLAPSIBLE_TEMPLATE = \"<details><summary>Original Description</summary>{desc}</details>\"",
+    "suggested_code": "DESC_COLLAPSIBLE_TEMPLATE = \"<details><summary>Original Description</summary>\\n\\n{desc}\\n\\n</details>\"",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
+    "start_line": 5,
+    "end_line": 5,
+    "sentiment": "neutral",
+    "severity": 4
+  },
+  {
+    "category": "Performance",
+    "description": "Potential performance issues with large input sizes in test_create_pr_description.py",
+    "impact": "medium",
+    "rationale": "The boundary condition tests with large strings could lead to performance degradation or memory issues if not handled properly.",
+    "recommendation": "Consider adding checks to limit the size of inputs or to ensure that the function can handle large inputs efficiently without excessive memory usage.",
+    "current_code": "def test_create_pr_description_boundary_conditions(desc, original_desc):",
+    "suggested_code": "# Add input size validation or handling logic here",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
+    "start_line": 46,
+    "end_line": 60,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Maintainability",
+    "description": "Repeated mock patching in test_create_folder.py",
+    "impact": "medium",
+    "rationale": "The repeated use of mock.patch can lead to code duplication and make it harder to maintain tests.",
+    "recommendation": "Consider using fixtures for mocking to reduce redundancy and improve maintainability.",
+    "current_code": "@mock.patch(\"kaizen.helpers.output.os.makedirs\")\n@mock.patch(\"kaizen.helpers.output.os.path.exists\")\n@mock.patch(\"kaizen.helpers.output.logger\")",
+    "suggested_code": "@pytest.fixture\ndef mock_os():\n    with mock.patch(\"kaizen.helpers.output.os\") as mock_os:\n        yield mock_os",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_folder.py",
+    "start_line": 6,
+    "end_line": 8,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Error Handling",
+    "description": "Lack of error handling in create_folder function",
+    "impact": "high",
+    "rationale": "The create_folder function should handle potential errors, such as permission issues or invalid paths, to prevent crashes.",
+    "recommendation": "Implement try-except blocks to catch and handle exceptions gracefully.",
+    "current_code": "os.makedirs(folder_path)",
+    "suggested_code": "try:\n    os.makedirs(folder_path)\nexcept OSError as e:\n    # Handle error (e.g., log it, raise a custom exception)",
+    "file_path": "kaizen/helpers/output.py",
+    "start_line": 10,
+    "end_line": 10,
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Code Readability",
+    "description": "Inconsistent formatting in PR_COLLAPSIBLE_TEMPLATE.",
+    "impact": "medium",
+    "rationale": "The original template uses a multi-line string with parentheses, while the new version uses triple quotes. This inconsistency can lead to confusion and makes the code harder to maintain.",
+    "recommendation": "Choose one style for multi-line strings and apply it consistently throughout the codebase.",
+    "current_code": "PR_COLLAPSIBLE_TEMPLATE = (\n    \"<details>\\n\"\n    \"<summary>Review</summary>\\n\"\n    ...\n)",
+    "suggested_code": "PR_COLLAPSIBLE_TEMPLATE = \"\"\"\n<details>\n<summary>Review Comment</summary>\n<p>{comment}</p>\n...\"\"\"",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 4,
+    "end_line": 16,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling",
+    "description": "Missing validation for required fields in review comments.",
+    "impact": "high",
+    "rationale": "The tests do not validate if required fields (like comment, reason, solution) are present. This could lead to runtime errors or unexpected behavior when these fields are missing.",
+    "recommendation": "Implement checks to ensure that all required fields are present before processing the review comments.",
+    "current_code": "def test_reviews_with_missing_fields():\n    topics ={...}",
+    "suggested_code": "def validate_review(review):\n    required_fields =['comment', 'reason', 'solution', 'confidence', 'start_line', 'end_line', 'file_name', 'severity']\n    for field in required_fields:\n        if field not in review:\n            raise ValueError(f'Missing required field:{field}')\n\n    # Call this validation function in your tests.",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 143,
+    "end_line": 184,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Test Coverage",
+    "description": "Lack of tests for edge cases in review comments.",
+    "impact": "medium",
+    "rationale": "The current tests do not cover scenarios where fields are missing or contain invalid data. This could lead to unhandled exceptions in production.",
+    "recommendation": "Add unit tests that cover various edge cases, such as missing fields, invalid data types, and empty inputs.",
+    "current_code": "def test_reviews_with_missing_fields():\n    topics ={...}",
+    "suggested_code": "def test_reviews_with_missing_fields():\n    # Test with missing fields\n    topics ={...}\n    with pytest.raises(ValueError):\n        create_pr_review_text(topics)",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 143,
+    "end_line": 184,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Performance",
+    "description": "Inefficient handling of review comments.",
+    "impact": "medium",
+    "rationale": "The current implementation processes review comments in a linear fashion. If the number of comments grows, this could lead to performance issues.",
+    "recommendation": "Consider using a more efficient data structure or algorithm to handle review comments, especially if the number of comments can be large.",
+    "current_code": "for topic, reviews in topics.items():\n    for review in reviews:\n        ...",
+    "suggested_code": "from collections import defaultdict\n\ncomments_by_topic = defaultdict(list)\nfor topic, reviews in topics.items():\n    comments_by_topic[topic].extend(reviews)\n# Process comments from comments_by_topic",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 80,
+    "end_line": 100,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Readability",
+    "description": "Inconsistent naming conventions for test cases.",
+    "impact": "medium",
+    "rationale": "Consistency in naming conventions improves readability and maintainability. Some test functions use underscores while others use camel case.",
+    "recommendation": "Adopt a consistent naming convention for all test functions, preferably using snake_case.",
+    "current_code": "def test_get_parent_folder_normal_case():",
+    "suggested_code": "def test_get_parent_folder_normal():",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py",
+    "start_line": 6,
+    "end_line": 6,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling",
+    "description": "Lack of error handling for file operations.",
+    "impact": "high",
+    "rationale": "The `create_test_files` function may fail due to various reasons (e.g., permission issues), and this should be handled gracefully to avoid crashes.",
+    "recommendation": "Wrap file operations in try-except blocks to handle potential exceptions and log appropriate error messages.",
+    "current_code": "with open(file_path, 'r') as f:",
+    "suggested_code": "try:\n    with open(file_path, 'r') as f:\n        return f.read()\nexcept OSError as e:\n    # Log the error or handle it appropriately\n    raise e",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 17,
+    "end_line": 17,
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Performance",
+    "description": "Potential inefficiency in sanitizing filenames.",
+    "impact": "medium",
+    "rationale": "The current implementation of `sanitize_filename` iterates over each character in the filename, which could be optimized for better performance.",
+    "recommendation": "Consider using regular expressions to sanitize filenames, which can be more efficient and concise.",
+    "current_code": "return \"\".join(c if c.isalnum() or c in (' ', '.', '_') else '_' for c in filename)",
+    "suggested_code": "import re\nreturn re.sub(r'[^\\w .]', '_', filename)",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 21,
+    "end_line": 22,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Testing Coverage",
+    "description": "Insufficient testing for edge cases.",
+    "impact": "high",
+    "rationale": "While there are tests for normal cases, edge cases such as very long filenames or special characters are not thoroughly tested.",
+    "recommendation": "Add more test cases to cover edge scenarios, such as filenames with special characters or exceeding typical length limits.",
+    "current_code": "def test_special_characters_in_test_names(tmp_path, mock_dependencies):",
+    "suggested_code": "def test_edge_cases_in_filenames(tmp_path, mock_dependencies):\n    # Add tests for various edge cases here",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 104,
+    "end_line": 106,
+    "sentiment": "neutral",
+    "severity": 7
+  },
+  {
+    "category": "Code Structure",
+    "description": "Redundant mock patches in tests.",
+    "impact": "medium",
+    "rationale": "Repeatedly patching the same functions in multiple tests can lead to code duplication and make the tests harder to maintain.",
+    "recommendation": "Use fixtures to handle common mock patches, reducing redundancy and improving readability.",
+    "current_code": "@mock.patch(\"kaizen.helpers.output.create_folder\")\n@mock.patch(\"kaizen.helpers.output.general.clean_python_code\")\n@mock.patch(\"kaizen.helpers.output.logger\")\ndef test_create_test_files_special_characters(...):",
+    "suggested_code": "@pytest.fixture\ndef mock_dependencies():\n    with mock.patch(...) as mock_create_folder, ...:\n        yield mock_create_folder, ...",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 84,
+    "end_line": 88,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Code Efficiency",
+    "description": "Use of synchronous function in an asynchronous context.",
+    "impact": "high",
+    "rationale": "The `get_web_html` function is being called synchronously within an async test, which can lead to blocking behavior and defeat the purpose of using async functions.",
+    "recommendation": "Ensure that `get_web_html` is awaited in all async contexts to maintain non-blocking behavior.",
+    "current_code": "result = get_web_html(url)",
+    "suggested_code": "result = await get_web_html(url)",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 92,
+    "end_line": 92,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Error Handling",
+    "description": "Lack of error handling in tests for invalid URLs.",
+    "impact": "high",
+    "rationale": "The test for invalid URLs should assert that an exception is raised, ensuring that the function behaves correctly under error conditions.",
+    "recommendation": "Wrap the call to `get_web_html` in a pytest.raises context manager to verify that the expected exception is thrown.",
+    "current_code": "result = get_web_html(url)",
+    "suggested_code": "with pytest.raises(Exception, match='Network error'):\n    await get_web_html(url)",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 89,
+    "end_line": 91,
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Readability",
+    "description": "Inconsistent use of single and double quotes.",
+    "impact": "medium",
+    "rationale": "Inconsistent quoting can lead to confusion and makes the code less readable. It's a good practice to stick to one style throughout the codebase.",
+    "recommendation": "Choose either single or double quotes for string literals and apply it consistently across the code.",
+    "current_code": "with patch('kaizen.helpers.output.get_html', new_callable=AsyncMock) as mock:",
+    "suggested_code": "with patch(\"kaizen.helpers.output.get_html\", new_callable=AsyncMock) as mock:",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 12,
+    "end_line": 12,
+    "sentiment": "neutral",
+    "severity": 4
+  },
+  {
+    "category": "Test Coverage",
+    "description": "Lack of tests for edge cases and large inputs.",
+    "impact": "medium",
+    "rationale": "While there are tests for normal cases, edge cases such as extremely large HTML content or malformed HTML should also be tested to ensure robustness.",
+    "recommendation": "Add tests that specifically handle edge cases and large inputs to ensure that the function can handle unexpected scenarios.",
+    "current_code": "No specific tests for edge cases.",
+    "suggested_code": "async def test_get_web_html_large_content(mock_get_html):\n    large_html_content = '<html>' + '<p>Test</p>' * 10000 + '</html>'\n    expected_output = '<html><body>' + '<p>Test</p>' * 10000 + '</body></html>'\n    mock_get_html.return_value = large_html_content\n    result = await get_web_html(url)\n    assert result.strip() == expected_output.strip()",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 93,
+    "end_line": 95,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Code Structure",
+    "description": "Inconsistent use of class attributes for supported languages and prompts.",
+    "impact": "high",
+    "rationale": "The code uses both instance attributes and class attributes for similar purposes, which can lead to confusion and inconsistency in how the data is accessed and modified.",
+    "recommendation": "Standardize the use of either class attributes or instance attributes for supported languages and prompts to improve clarity.",
+    "current_code": "self.supported_languages ={...}",
+    "suggested_code": "SUPPORTED_LANGUAGES ={...}",
+    "file_path": "kaizen/generator/unit_test.py",
+    "start_line": 24,
+    "end_line": 30,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Error Handling",
+    "description": "Lack of error handling when reading files.",
+    "impact": "high",
+    "rationale": "If the file does not exist or cannot be read, the current implementation will raise an unhandled exception, which could crash the application.",
+    "recommendation": "Implement error handling using try-except blocks when reading files to gracefully handle potential issues.",
+    "current_code": "with open(file_path, 'r') as file: content = file.read()",
+    "suggested_code": "try:\n    with open(file_path, 'r') as file:\n        return file.read()\nexcept IOError as e:\n    print(f'Error reading file:{e}')\n    return None",
+    "file_path": "kaizen/generator/unit_test.py",
+    "start_line": 108,
+    "end_line": 110,
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Readability",
+    "description": "Long methods with multiple responsibilities.",
+    "impact": "medium",
+    "rationale": "Methods like `generate_tests_from_dir` and `generate_ai_tests` are doing too much, making them harder to read and maintain.",
+    "recommendation": "Refactor long methods into smaller, single-responsibility methods to enhance readability and maintainability.",
+    "current_code": "def generate_tests_from_dir(self, ...): ...",
+    "suggested_code": "def generate_tests_from_dir(self, ...):\n    self._prepare_tests(...)\n    self._execute_tests(...)",
+    "file_path": "kaizen/generator/unit_test.py",
+    "start_line": 61,
+    "end_line": 61,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Performance",
+    "description": "Repeated calls to `os.makedirs` for the same directory.",
+    "impact": "medium",
+    "rationale": "The code calls `os.makedirs` multiple times for the same directory, which can be inefficient.",
+    "recommendation": "Check if the directory exists before creating it to avoid unnecessary calls.",
+    "current_code": "os.makedirs(self.log_dir, exist_ok=True)",
+    "suggested_code": "if not os.path.exists(self.log_dir):\n    os.makedirs(self.log_dir)",
+    "file_path": "kaizen/generator/unit_test.py",
+    "start_line": 58,
+    "end_line": 58,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Logging",
+    "description": "Lack of logging for critical operations.",
+    "impact": "medium",
+    "rationale": "The absence of logging makes it difficult to trace the flow of execution and diagnose issues.",
+    "recommendation": "Add logging statements to critical operations to provide better visibility into the application's behavior.",
+    "current_code": "print(f'Error: Could not generate tests for{file_path}:{e}')",
+    "suggested_code": "self.logger.error(f'Could not generate tests for{file_path}:{e}')",
+    "file_path": "kaizen/generator/unit_test.py",
+    "start_line": 73,
+    "end_line": 73,
+    "sentiment": "neutral",
+    "severity": 4
+  },
+  {
+    "category": "Logging Best Practices",
+    "description": "Improper handling of logger levels.",
+    "impact": "high",
+    "rationale": "Setting all loggers to ERROR without a conditional check can lead to loss of important log information. This could make debugging difficult, especially in production environments where lower log levels may be necessary for tracing issues.",
+    "recommendation": "Consider allowing configurable log levels instead of hardcoding ERROR for all loggers. This can be done via environment variables or configuration files.",
+    "current_code": "logging.getLogger(name).setLevel(logging.ERROR)",
+    "suggested_code": "level = os.environ.get('DEFAULT_LOG_LEVEL', 'ERROR').upper()\nlogging.getLogger(name).setLevel(getattr(logging, level, logging.ERROR))",
+    "file_path": "kaizen/llms/provider.py",
+    "start_line": 13,
+    "end_line": 28,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Code Readability",
+    "description": "Lack of docstrings and comments in the new function.",
+    "impact": "medium",
+    "rationale": "The function `set_all_loggers_to_ERROR` lacks a docstring, making it unclear what its purpose is. Clear documentation is essential for maintainability and understanding of the code.",
+    "recommendation": "Add a docstring to explain the purpose of the function and its behavior.",
+    "current_code": "def set_all_loggers_to_ERROR():",
+    "suggested_code": "def set_all_loggers_to_ERROR():\n    \"\"\"\n    Set all loggers to the ERROR level.\n    This function iterates through all loggers and sets their level to ERROR, which can help in reducing log verbosity.\n    \"\"\"",
+    "file_path": "kaizen/llms/provider.py",
+    "start_line": 13,
+    "end_line": 13,
+    "sentiment": "negative",
+    "severity": 5
+  },
+  {
+    "category": "Performance",
+    "description": "Inefficient logging of all logger names.",
+    "impact": "medium",
+    "rationale": "Printing all logger names and their levels can be inefficient and clutter the output, especially if there are many loggers. This could lead to performance issues in environments with extensive logging.",
+    "recommendation": "Consider logging only the loggers that are set to a certain level or have specific names of interest.",
+    "current_code": "print(f\"Logger:{name}, Level:{logging.getLevelName(logger.level)}\")",
+    "suggested_code": "if logger.level <= logging.WARNING:\n    print(f\"Logger:{name}, Level:{logging.getLevelName(logger.level)}\")",
+    "file_path": "kaizen/llms/provider.py",
+    "start_line": 14,
+    "end_line": 20,
+    "sentiment": "negative",
+    "severity": 6
+  },
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "11",
+    "end_line": "11",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/review.md b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/review.md
new file mode 100644
index 00000000..bdd46645
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/review.md
@@ -0,0 +1,56 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/400
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 27
+- Critical: 1
+- Important: 0
+- Minor: 0
+- Files Affected: 11
+## 🏆 Code Quality
+[███████████████░░░░░] 75% (Fair)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Configuration (1 issues)</strong></summary>
+
+### 1. Changes made to sensitive file
+📁 **File:** `config.json:11`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 35088, "completion_tokens": 5327, "total_tokens": 40415}
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_335/comments.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/comments.json
similarity index 100%
rename from .experiments/code_review/sonnet-3.5/no_eval/pr_335/comments.json
rename to .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/comments.json
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/issues.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/issues.json
new file mode 100644
index 00000000..e9c23d5a
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/issues.json
@@ -0,0 +1,44 @@
+[
+  {
+    "category": "Code Efficiency",
+    "description": "Redundant repository setup call.",
+    "impact": "medium",
+    "rationale": "The repository setup is called every time the script runs, which may not be necessary if the repository has already been set up. This can lead to unnecessary overhead.",
+    "recommendation": "Check if the repository is already set up before calling setup_repository.",
+    "current_code": "analyzer.setup_repository(\"./github_app/\")",
+    "suggested_code": "if not analyzer.is_repository_setup():\n    analyzer.setup_repository(\"./github_app/\")",
+    "file_path": "examples/ragify_codebase/main.py",
+    "start_line": 7,
+    "end_line": 7,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Code Clarity",
+    "description": "Missing comment on potential duplicate embedding storage.",
+    "impact": "high",
+    "rationale": "The TODO comment indicates a potential issue with duplicate embeddings being pushed to the database, which could lead to data integrity issues. Clear documentation is essential for maintainability.",
+    "recommendation": "Implement a check to prevent duplicate embeddings from being stored and document the logic clearly.",
+    "current_code": "# TODO: DONT PUSH DUPLICATE",
+    "suggested_code": "# Check for duplicates before storing the embedding\nif not self.is_duplicate_embedding(embedding):\n    # Store the embedding in the database",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 157,
+    "end_line": 157,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Dependency Management",
+    "description": "Outdated dependency version.",
+    "impact": "medium",
+    "rationale": "Updating dependencies is crucial for security and performance improvements. The change from version 0.10.47 to 0.10.65 should be validated for compatibility.",
+    "recommendation": "Ensure that the new version is compatible with the existing codebase and run tests after updating.",
+    "current_code": "llama-index-core = \"^0.10.47\"",
+    "suggested_code": "llama-index-core = \"0.10.65\"",
+    "file_path": "pyproject.toml",
+    "start_line": 27,
+    "end_line": 27,
+    "sentiment": "neutral",
+    "severity": 4
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/review.md b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/review.md
new file mode 100644
index 00000000..44d9c077
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/review.md
@@ -0,0 +1,33 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/440
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 3
+- Critical: 0
+- Important: 0
+- Minor: 0
+- Files Affected: 3
+## 🏆 Code Quality
+[███████████████░░░░░] 75% (Fair)
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 1063, "completion_tokens": 567, "total_tokens": 1630}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/comments.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/comments.json
new file mode 100644
index 00000000..752f765d
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/comments.json
@@ -0,0 +1,30 @@
+[
+  {
+    "category": "Security",
+    "description": "Potential exposure of sensitive information in logs.",
+    "impact": "critical",
+    "rationale": "Logging sensitive information, such as access tokens or user data, can lead to security vulnerabilities. It is crucial to avoid logging sensitive data.",
+    "recommendation": "Review logging statements to ensure that no sensitive information is being logged. Use obfuscation or redaction where necessary.",
+    "current_code": "logger.debug(f\"Post Review comment response:{response.text}\")",
+    "suggested_code": "logger.debug(f\"Post Review comment response:{response.status_code}\")  # Avoid logging response body",
+    "file_path": "github_app/github_helper/pull_requests.py",
+    "start_line": 181,
+    "end_line": 182,
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/issues.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/issues.json
new file mode 100644
index 00000000..fe021299
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/issues.json
@@ -0,0 +1,86 @@
+[
+  {
+    "category": "Error Handling",
+    "description": "Broad exception handling without specific error types.",
+    "impact": "high",
+    "rationale": "Using a generic Exception in the error handling can mask underlying issues and make debugging difficult. It is better to catch specific exceptions to handle known error cases appropriately.",
+    "recommendation": "Refine the exception handling to catch specific exceptions and log or handle them accordingly.",
+    "current_code": "except Exception:\n    print(\"Error\")",
+    "suggested_code": "except KeyError:\n    raise ValueError(f\"Invalid confidence level:{min_confidence}\")\nexcept ValueError as ve:\n    logger.error(f\"ValueError:{ve}\")",
+    "file_path": "github_app/github_helper/pull_requests.py",
+    "start_line": 140,
+    "end_line": 141,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Code Readability",
+    "description": "Inconsistent naming conventions and lack of comments.",
+    "impact": "medium",
+    "rationale": "Inconsistent naming can lead to confusion and make the code harder to maintain. Comments are essential for explaining complex logic.",
+    "recommendation": "Ensure consistent naming conventions across the codebase and add comments where necessary to explain the purpose of complex logic.",
+    "current_code": "def post_pull_request(url, data, installation_id, tests=None):",
+    "suggested_code": "def post_pull_request(url, data, installation_id, test_files=None):  # Renamed for clarity",
+    "file_path": "github_app/github_helper/pull_requests.py",
+    "start_line": 106,
+    "end_line": 107,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Performance",
+    "description": "Inefficient file sorting algorithm.",
+    "impact": "high",
+    "rationale": "The current sorting algorithm has a time complexity of O(n^2), which can lead to performance issues with larger datasets. Using Python's built-in sorting methods would be more efficient.",
+    "recommendation": "Replace the custom sorting logic with Python's built-in sort() method for better performance.",
+    "current_code": "for file in files:\n    min_index = len(sorted_files)\n    file_name = file[\"filename\"]\n    for i, sorted_file in enumerate(sorted_files):\n        if file_name < sorted_file[\"filename\"]:\n            min_index = i\n            break\n    sorted_files.insert(min_index, file)",
+    "suggested_code": "sorted_files = sorted(files, key=lambda x: x['filename'])",
+    "file_path": "github_app/github_helper/pull_requests.py",
+    "start_line": 184,
+    "end_line": 196,
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Unit Testing",
+    "description": "Lack of unit tests for new functionality.",
+    "impact": "high",
+    "rationale": "New functionality should be accompanied by unit tests to ensure correctness and prevent regressions in the future.",
+    "recommendation": "Add unit tests for the new functions, especially for `generate_tests` and `sort_files`, to validate their behavior.",
+    "current_code": "def generate_tests(pr_files):\n    return[f[\"filename\"] for f in pr_files]",
+    "suggested_code": "# Add unit tests in a separate test file for generate_tests and sort_files",
+    "file_path": "github_app/github_helper/pull_requests.py",
+    "start_line": 199,
+    "end_line": 200,
+    "sentiment": "negative",
+    "severity": 9
+  },
+  {
+    "category": "Security",
+    "description": "Potential exposure of sensitive information in logs.",
+    "impact": "critical",
+    "rationale": "Logging sensitive information, such as access tokens or user data, can lead to security vulnerabilities. It is crucial to avoid logging sensitive data.",
+    "recommendation": "Review logging statements to ensure that no sensitive information is being logged. Use obfuscation or redaction where necessary.",
+    "current_code": "logger.debug(f\"Post Review comment response:{response.text}\")",
+    "suggested_code": "logger.debug(f\"Post Review comment response:{response.status_code}\")  # Avoid logging response body",
+    "file_path": "github_app/github_helper/pull_requests.py",
+    "start_line": 181,
+    "end_line": 182,
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/review.md b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/review.md
new file mode 100644
index 00000000..11cdc033
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/review.md
@@ -0,0 +1,72 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/476
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 6
+- Critical: 2
+- Important: 0
+- Minor: 0
+- Files Affected: 2
+## 🏆 Code Quality
+[███████████████░░░░░] 75% (Fair)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Security (2 issues)</strong></summary>
+
+### 1. Potential exposure of sensitive information in logs.
+📁 **File:** `github_app/github_helper/pull_requests.py:181`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Potential exposure of sensitive information in logs.
+💡 **Solution:** 
+
+**Current Code:**
+```python
+logger.debug(f"Post Review comment response:{response.text}")
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 2. Changes made to sensitive file
+📁 **File:** `config.json:1`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 3706, "completion_tokens": 1023, "total_tokens": 4729}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/comments.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/comments.json
new file mode 100644
index 00000000..0637a088
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/comments.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/issues.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/issues.json
new file mode 100644
index 00000000..d942a77a
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/issues.json
@@ -0,0 +1,86 @@
+[
+  {
+    "category": "Unused Imports",
+    "description": "The import of 'random' is unnecessary and should be removed.",
+    "impact": "low",
+    "rationale": "Unused imports can clutter the code and increase loading time slightly.",
+    "recommendation": "Remove the import statement for 'random'.",
+    "current_code": "import random  # Unused import",
+    "suggested_code": "",
+    "file_path": "main.py",
+    "start_line": 8,
+    "end_line": 8,
+    "sentiment": "negative",
+    "severity": 3
+  },
+  {
+    "category": "Error Handling",
+    "description": "The API call in 'process_applicant' lacks a retry mechanism.",
+    "impact": "high",
+    "rationale": "If the API call fails, the function will throw an error without any recovery, leading to potential data loss.",
+    "recommendation": "Implement a retry mechanism with exponential backoff for the API call.",
+    "current_code": "response = completion(...)",
+    "suggested_code": "import time\n\nfor attempt in range(3):\n    try:\n        response = completion(...)\n        break\n    except Exception as e:\n        if attempt < 2:\n            time.sleep(2 ** attempt)\n        else:\n            raise e",
+    "file_path": "main.py",
+    "start_line": 65,
+    "end_line": 68,
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Silent Failures",
+    "description": "The JSON parsing failure in 'process_applicant' is silent and does not log the error.",
+    "impact": "high",
+    "rationale": "Silent failures can lead to debugging difficulties and loss of important information.",
+    "recommendation": "Log the error message instead of just printing it.",
+    "current_code": "print(f\"Failed to parse content for applicant\")",
+    "suggested_code": "import logging\n\nlogging.error(f\"Failed to parse content for applicant:{e}\")",
+    "file_path": "main.py",
+    "start_line": 86,
+    "end_line": 86,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Redundant Code",
+    "description": "The check for an empty DataFrame in 'main' is unnecessary.",
+    "impact": "low",
+    "rationale": "If the DataFrame is empty, the process_applicants function will handle it gracefully.",
+    "recommendation": "Remove the check for an empty DataFrame.",
+    "current_code": "if len(df) == 0:\n    return",
+    "suggested_code": "",
+    "file_path": "main.py",
+    "start_line": 141,
+    "end_line": 143,
+    "sentiment": "neutral",
+    "severity": 2
+  },
+  {
+    "category": "Division by Zero",
+    "description": "Potential division by zero when calculating total tokens.",
+    "impact": "high",
+    "rationale": "If total_tokens is zero, it will raise an exception.",
+    "recommendation": "Add a check to prevent division by zero.",
+    "current_code": "print(f\"Total tokens used:{total_tokens:,}\")",
+    "suggested_code": "if total_tokens > 0:\n    print(f\"Total tokens used:{total_tokens:,}\")\nelse:\n    print(\"No tokens used.\")",
+    "file_path": "main.py",
+    "start_line": 159,
+    "end_line": 161,
+    "sentiment": "negative",
+    "severity": 6
+  },
+  {
+    "category": "File Handling",
+    "description": "No error handling for file not found in 'main'.",
+    "impact": "high",
+    "rationale": "If the file does not exist, the program will crash without a user-friendly message.",
+    "recommendation": "Add a try-except block to handle file not found errors.",
+    "current_code": "main(input_file)",
+    "suggested_code": "try:\n    main(input_file)\nexcept FileNotFoundError:\n    print(f\"Error: File '{input_file}' not found.\")",
+    "file_path": "main.py",
+    "start_line": 175,
+    "end_line": 175,
+    "sentiment": "negative",
+    "severity": 8
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/review.md b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/review.md
new file mode 100644
index 00000000..112891fd
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/review.md
@@ -0,0 +1,41 @@
+PR URL: https://github.com/sauravpanda/applicant-screening/pull/5
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 6
+- Critical: 0
+- Important: 0
+- Minor: 0
+- Files Affected: 1
+## 🏆 Code Quality
+[███████████████░░░░░] 75% (Fair)
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (2 issues)</strong></summary>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 5853, "completion_tokens": 998, "total_tokens": 6851}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/comments.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/comments.json
new file mode 100644
index 00000000..8b8c06d9
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/comments.json
@@ -0,0 +1,86 @@
+[
+  {
+    "category": "Security Vulnerabilities",
+    "description": "Hard-coded API keys in config.json",
+    "impact": "critical",
+    "rationale": "Storing sensitive information such as API keys directly in the codebase can lead to security breaches if the repository is exposed. This practice violates security best practices and can compromise application integrity.",
+    "recommendation": "Use environment variables or a secure vault to manage sensitive information instead of hard-coding them in configuration files.",
+    "current_code": "\"api_key\": \"os.environ/AZURE_API_KEY\",",
+    "suggested_code": "\"api_key\": \"${AZURE_API_KEY}\",",
+    "file_path": "config.json",
+    "start_line": 13,
+    "end_line": 14,
+    "sentiment": "negative",
+    "severity": 9
+  },
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Docker",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to Dockerfile, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_path": "Dockerfile",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Docker",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to docker-compose.yml, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "15",
+    "end_line": "15",
+    "side": "RIGHT",
+    "file_path": "docker-compose.yml",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Version Control",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to .gitignore, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "164",
+    "end_line": "164",
+    "side": "RIGHT",
+    "file_path": ".gitignore",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Database",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to db_setup/init.sql, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_path": "db_setup/init.sql",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/issues.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/issues.json
new file mode 100644
index 00000000..81b0a7b6
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/issues.json
@@ -0,0 +1,282 @@
+[
+  {
+    "category": "Security Vulnerabilities",
+    "description": "Hard-coded API keys in config.json",
+    "impact": "critical",
+    "rationale": "Storing sensitive information such as API keys directly in the codebase can lead to security breaches if the repository is exposed. This practice violates security best practices and can compromise application integrity.",
+    "recommendation": "Use environment variables or a secure vault to manage sensitive information instead of hard-coding them in configuration files.",
+    "current_code": "\"api_key\": \"os.environ/AZURE_API_KEY\",",
+    "suggested_code": "\"api_key\": \"${AZURE_API_KEY}\",",
+    "file_path": "config.json",
+    "start_line": 13,
+    "end_line": 14,
+    "sentiment": "negative",
+    "severity": 9
+  },
+  {
+    "category": "Code Structure and Design",
+    "description": "Lack of comments in critical sections of code",
+    "impact": "medium",
+    "rationale": "While the code is generally readable, critical sections, especially in the Dockerfile and SQL scripts, lack comments explaining their purpose. This can hinder maintainability and onboarding for new developers.",
+    "recommendation": "Add comments to explain the purpose of each section, especially in complex scripts and Docker configurations.",
+    "current_code": "RUN apt-get update && apt-get install -y \\",
+    "suggested_code": "# Update package list and install necessary dependencies\nRUN apt-get update && apt-get install -y \\",
+    "file_path": "Dockerfile",
+    "start_line": 8,
+    "end_line": 11,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling and Logging",
+    "description": "Lack of error handling in database operations",
+    "impact": "high",
+    "rationale": "The database operations in the custom_query method do not handle potential exceptions that may arise from database connectivity issues or query failures. This can lead to unhandled exceptions and application crashes.",
+    "recommendation": "Implement try-except blocks around database operations to catch exceptions and log errors appropriately.",
+    "current_code": "with client.cursor() as cur:\n    cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))",
+    "suggested_code": "try:\n    with client.cursor() as cur:\n        cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))\nexcept Exception as e:\n    # Log the error\n    print(f\"Database query failed:{e}\")",
+    "file_path": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 39,
+    "end_line": 41,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Performance Issues",
+    "description": "Inefficient handling of query results",
+    "impact": "medium",
+    "rationale": "The current implementation fetches all results from the database and processes them in memory. This can lead to performance issues with large datasets.",
+    "recommendation": "Consider using pagination or streaming results to handle large datasets more efficiently.",
+    "current_code": "results = cur.fetchall()",
+    "suggested_code": "# Consider using a generator to yield results one by one\nfor row in cur:\n    # Process each row as it is fetched",
+    "file_path": "kaizen/retriever/custom_vector_store.py",
+    "start_line": 42,
+    "end_line": 42,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Code Structure and Design",
+    "description": "Code duplication in Dockerfiles",
+    "impact": "medium",
+    "rationale": "The Dockerfile and Dockerfile-postgres have similar sections for installing dependencies. This can lead to maintenance challenges if changes are needed in the future.",
+    "recommendation": "Consider creating a base Dockerfile that includes common dependencies and extend it in specific Dockerfiles.",
+    "current_code": "RUN apt-get update && apt-get install -y \\\n    git \\\n    build-essential \\\n    && rm -rf /var/lib/apt/lists/*",
+    "suggested_code": "# Base Dockerfile\n# Common dependencies can be installed here\nRUN apt-get update && apt-get install -y \\\n    git \\\n    build-essential \\\n    && rm -rf /var/lib/apt/lists/*",
+    "file_path": "Dockerfile",
+    "start_line": 8,
+    "end_line": 11,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling and Logging",
+    "description": "Inconsistent error handling in the `generate_abstraction` method.",
+    "impact": "high",
+    "rationale": "The `generate_abstraction` method raises exceptions directly without logging them, which may lead to loss of context in error tracking.",
+    "recommendation": "Log exceptions before raising them to maintain context for debugging.",
+    "current_code": "raise e",
+    "suggested_code": "logger.error(f'Error generating abstraction:{str(e)}'); raise e",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 218,
+    "end_line": 219,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Code Structure and Design",
+    "description": "Hard-coded values for start and end lines in `store_code_in_db` method.",
+    "impact": "medium",
+    "rationale": "Using hard-coded values for line numbers can lead to inaccuracies and maintenance issues when the code changes.",
+    "recommendation": "Calculate actual start and end lines based on the parsed content.",
+    "current_code": "\"start_line\": 1,  # You might want to calculate actual start and end lines",
+    "suggested_code": "\"start_line\": actual_start_line, \"end_line\": actual_end_line,",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 287,
+    "end_line": 287,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Performance Issues",
+    "description": "Potential N+1 query problem in `store_function_relationships` method.",
+    "impact": "high",
+    "rationale": "The current implementation executes a separate query for each edge in the graph, which can lead to performance bottlenecks.",
+    "recommendation": "Batch insert relationships or use a single query to insert multiple relationships at once.",
+    "current_code": "connection.execute(query,{\"caller\": f\"%{caller}%\", \"callee\": f\"%{callee}%\"})",
+    "suggested_code": "batch_insert_query = ...; connection.execute(batch_insert_query, batch_params)",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 295,
+    "end_line": 312,
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Code Structure and Design",
+    "description": "The `parse_file` method has high cyclomatic complexity.",
+    "impact": "medium",
+    "rationale": "High cyclomatic complexity can make the code harder to understand and maintain. The method has multiple nested conditions.",
+    "recommendation": "Refactor the method to break it into smaller, more manageable functions.",
+    "current_code": "if isinstance(items, dict): ... elif isinstance(items, list): ...",
+    "suggested_code": "def handle_dict(items): ...; def handle_list(items): ...; then call these in parse_file",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 98,
+    "end_line": 102,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Resource Management",
+    "description": "File handles in `parse_file` method are not guaranteed to be closed in case of an exception.",
+    "impact": "high",
+    "rationale": "If an exception occurs before the file is closed, it may lead to resource leaks.",
+    "recommendation": "Use a context manager to ensure the file is properly closed.",
+    "current_code": "with open(file_path, \"r\", encoding=\"utf-8\") as file:",
+    "suggested_code": "with open(file_path, \"r\", encoding=\"utf-8\") as file: ...",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 92,
+    "end_line": 92,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Error Handling and Logging",
+    "description": "Broad exception handling in language loading and parser creation.",
+    "impact": "high",
+    "rationale": "Using a broad exception catch can obscure the root cause of errors and make debugging difficult. Specific exceptions should be caught to provide clearer error messages and handling.",
+    "recommendation": "Catch specific exceptions (e.g., ImportError, ValueError) instead of a generic Exception to improve error clarity.",
+    "current_code": "except Exception as e:",
+    "suggested_code": "except (ImportError, ValueError) as e:",
+    "file_path": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 28,
+    "end_line": 30,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Code Structure and Design",
+    "description": "Inconsistent naming conventions for functions and variables.",
+    "impact": "medium",
+    "rationale": "Inconsistent naming conventions can lead to confusion and reduce code readability. Following a consistent style guide improves maintainability.",
+    "recommendation": "Use snake_case for function names and variables to maintain consistency with Python's PEP 8 style guide.",
+    "current_code": "def check_language_files():",
+    "suggested_code": "def check_language_files():",
+    "file_path": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 101,
+    "end_line": 101,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Performance Issues",
+    "description": "Potential inefficiency in dynamic module loading.",
+    "impact": "medium",
+    "rationale": "Dynamically importing modules can introduce overhead and may lead to performance issues if done repeatedly in a loop. Consider caching or pre-loading modules if they are frequently accessed.",
+    "recommendation": "Evaluate if dynamic imports can be replaced with static imports or cached to improve performance.",
+    "current_code": "module = importlib.import_module(module_name)",
+    "suggested_code": "# Consider caching or pre-loading modules if frequently accessed.",
+    "file_path": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 23,
+    "end_line": 23,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Code Structure and Design",
+    "description": "Redundant code in traverse_tree function for JSX elements.",
+    "impact": "medium",
+    "rationale": "The logic for handling JSX elements is duplicated and can be simplified, improving readability and maintainability.",
+    "recommendation": "Refactor the JSX handling logic to reduce duplication.",
+    "current_code": "return{'type': 'component', 'name': (node.child_by_field_name('opening_element').child_by_field_name('name').text.decode('utf8') if node.type == 'jsx_element' else node.child_by_field_name('name').text.decode('utf8')), 'code': code_bytes[node.start_byte : node.end_byte].decode('utf8')}",
+    "suggested_code": "name = node.child_by_field_name('opening_element').child_by_field_name('name').text.decode('utf8') if node.type == 'jsx_element' else node.child_by_field_name('name').text.decode('utf8'); return{'type': 'component', 'name': name, 'code': code_bytes[node.start_byte : node.end_byte].decode('utf8')}",
+    "file_path": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 71,
+    "end_line": 79,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Resource Management",
+    "description": "Lack of resource cleanup in parser creation.",
+    "impact": "high",
+    "rationale": "Not properly managing resources can lead to memory leaks or unclosed connections, especially in a long-running application.",
+    "recommendation": "Ensure that resources are properly closed or disposed of after use, especially in the ParserFactory.",
+    "current_code": "parser = Parser()",
+    "suggested_code": "# Ensure parser is properly disposed of after use.",
+    "file_path": "kaizen/retriever/tree_sitter_utils.py",
+    "start_line": 38,
+    "end_line": 38,
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Docker",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to Dockerfile, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "4",
+    "end_line": "4",
+    "side": "RIGHT",
+    "file_path": "Dockerfile",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Docker",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to docker-compose.yml, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "15",
+    "end_line": "15",
+    "side": "RIGHT",
+    "file_path": "docker-compose.yml",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Version Control",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to .gitignore, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "164",
+    "end_line": "164",
+    "side": "RIGHT",
+    "file_path": ".gitignore",
+    "sentiment": "negative",
+    "severity": 10
+  },
+  {
+    "category": "Database",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to db_setup/init.sql, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_path": "db_setup/init.sql",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/review.md b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/review.md
new file mode 100644
index 00000000..6183caac
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/review.md
@@ -0,0 +1,136 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/222
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 20
+- Critical: 6
+- Important: 0
+- Minor: 0
+- Files Affected: 8
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Security Vulnerabilities (6 issues)</strong></summary>
+
+### 1. Hard-coded API keys in config.json
+📁 **File:** `config.json:13`
+⚖️ **Severity:** 9/10
+🔍 **Description:** Hard-coded API keys in config.json
+💡 **Solution:** 
+
+**Current Code:**
+```python
+"api_key": "os.environ/AZURE_API_KEY",
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 2. Changes made to sensitive file
+📁 **File:** `config.json:4`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 3. Changes made to sensitive file
+📁 **File:** `Dockerfile:4`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 4. Changes made to sensitive file
+📁 **File:** `docker-compose.yml:15`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 5. Changes made to sensitive file
+📁 **File:** `.gitignore:164`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 6. Changes made to sensitive file
+📁 **File:** `db_setup/init.sql:1`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 20363, "completion_tokens": 2974, "total_tokens": 23337}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/comments.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/comments.json
new file mode 100644
index 00000000..0637a088
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/comments.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/issues.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/issues.json
new file mode 100644
index 00000000..1e3db8ca
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/issues.json
@@ -0,0 +1,156 @@
+[
+  {
+    "category": "Code Structure and Design",
+    "description": "Inconsistent naming conventions for components.",
+    "impact": "medium",
+    "rationale": "The naming of components should be consistent to improve readability and maintainability. For example, 'LinkComponent' and 'MemoryComponent' should follow a uniform naming pattern to indicate their purpose clearly.",
+    "recommendation": "Rename components to maintain consistency in naming conventions.",
+    "current_code": "function LinkComponent({...}){...}",
+    "suggested_code": "function MemoryComponent({...}){...}",
+    "file_path": "apps/web/app/(dash)/home/page.tsx",
+    "start_line": 321,
+    "end_line": 321,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling and Logging",
+    "description": "Lack of error handling in asynchronous operations.",
+    "impact": "high",
+    "rationale": "The function 'handleDeleteSpace' does not handle potential errors from the 'deleteSpace' API call. This could lead to unhandled promise rejections and a poor user experience.",
+    "recommendation": "Implement error handling for the asynchronous operation to ensure that errors are caught and logged appropriately.",
+    "current_code": "const response = await deleteSpace(id);",
+    "suggested_code": "try{const response = await deleteSpace(id);}catch (error){console.error('Error deleting space:', error); toast.error('An error occurred while deleting the space.');}",
+    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 66,
+    "end_line": 70,
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Performance Issues",
+    "description": "Inefficient merging and sorting of items.",
+    "impact": "medium",
+    "rationale": "The current approach merges and sorts items in a way that could be optimized. The use of multiple map operations followed by a sort can lead to performance degradation, especially with large datasets.",
+    "recommendation": "Consider using a single iteration to merge and sort items to improve performance.",
+    "current_code": "const unifiedItems =[...memoriesAndSpaces.memories.map(...), ...spaces.map(...)]",
+    "suggested_code": "const unifiedItems =[...memoriesAndSpaces.memories, ...spaces].map(spaceOrMemory => ({item: spaceOrMemory.type, date: new Date(spaceOrMemory.date), data: spaceOrMemory})).sort((a, b) => a.date - b.date);",
+    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 98,
+    "end_line": 118,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Syntax and Logic Errors",
+    "description": "Potential null pointer dereference.",
+    "impact": "high",
+    "rationale": "The code assumes that 'currentSpace' and 'usersWithAccess' are always defined. If they are null or undefined, it could lead to runtime errors.",
+    "recommendation": "Add null checks for 'currentSpace' and 'usersWithAccess' before accessing their properties.",
+    "current_code": "usersWithAccess.length > 0",
+    "suggested_code": "usersWithAccess?.length > 0",
+    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
+    "start_line": 168,
+    "end_line": 168,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Error Handling and Logging",
+    "description": "Inadequate error handling in asynchronous functions.",
+    "impact": "high",
+    "rationale": "The current implementation of the `handleSubmit` function throws errors without proper handling, which can lead to unhandled promise rejections and poor user experience.",
+    "recommendation": "Implement try-catch blocks around asynchronous calls to handle errors gracefully.",
+    "current_code": "const cont = await createMemory({\n    content: content,\n    spaces: spaces ?? undefined,\n});",
+    "suggested_code": "try{\n    const cont = await createMemory({\n        content: content,\n        spaces: spaces ?? undefined,\n});\n}catch (error){\n    toast.error(`Error: ${error.message}`);\n    return;\n}",
+    "file_path": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 213,
+    "end_line": 223,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Code Structure and Design",
+    "description": "Code duplication in the `handleSubmit` function and the form submission handler.",
+    "impact": "medium",
+    "rationale": "The logic for handling the form submission is repeated in both the `handleSubmit` function and the inline form submission handler, which violates the DRY principle and makes maintenance harder.",
+    "recommendation": "Extract the form submission logic into a separate function to avoid duplication.",
+    "current_code": "toast.promise(handleSubmit(content, selectedSpaces),{\n    loading: (...),\n    success: (data) => \"Memory queued\",\n    error: (error) => error.message,\n});",
+    "suggested_code": "const submitMemory = async () =>{\n    return toast.promise(handleSubmit(content, selectedSpaces),{\n        loading: (...),\n        success: (data) => \"Memory queued\",\n        error: (error) => error.message,\n});\n};\n\n// Use submitMemory in both places.",
+    "file_path": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 186,
+    "end_line": 198,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Performance Issues",
+    "description": "Potential performance issue with frequent state updates in `handleSubmit`.",
+    "impact": "medium",
+    "rationale": "The state updates for `content` and `selectedSpaces` are done sequentially, which can lead to unnecessary re-renders. Using a functional update can optimize this.",
+    "recommendation": "Use functional updates for state setters to ensure they are based on the latest state.",
+    "current_code": "setContent(\"\");\nsetSelectedSpaces([]);",
+    "suggested_code": "setContent((prev) => \"\");\nsetSelectedSpaces((prev) =>[]);",
+    "file_path": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 221,
+    "end_line": 223,
+    "sentiment": "neutral",
+    "severity": 4
+  },
+  {
+    "category": "Code Structure and Design",
+    "description": "Lack of separation of concerns in the `Menu` component.",
+    "impact": "medium",
+    "rationale": "The `Menu` component is handling multiple responsibilities, including state management and UI rendering. This can make the component harder to test and maintain.",
+    "recommendation": "Consider breaking the `Menu` component into smaller components, such as `MemoryForm` and `SpaceSelector`, to improve readability and maintainability.",
+    "current_code": "function Menu(){\n    // ...\n}",
+    "suggested_code": "function Menu(){\n    return <MemoryForm />;\n}\n\nfunction MemoryForm(){\n    // Logic for handling memory submission\n}",
+    "file_path": "apps/web/app/(dash)/menu.tsx",
+    "start_line": 35,
+    "end_line": 36,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Code Structure and Design",
+    "description": "Inconsistent naming conventions for props.",
+    "impact": "medium",
+    "rationale": "The props in the component have inconsistent naming conventions. Some props use camelCase while others use snake_case, which can lead to confusion and reduce readability.",
+    "recommendation": "Standardize the naming convention for props to camelCase to maintain consistency throughout the codebase.",
+    "current_code": "placeholder?: string;\nemptyMessage?: string;\ncreateNewMessage?: string;\nclassName?: string;",
+    "suggested_code": "placeholder?: string;\nemptyMessage?: string;\ncreateNewMessage?: string;\nclassName?: string;",
+    "file_path": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 23,
+    "end_line": 26,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Performance Issues",
+    "description": "Inefficient filtering of options.",
+    "impact": "high",
+    "rationale": "The current implementation filters options on every render, which can lead to performance degradation, especially with a large dataset. This could be optimized by memoizing the filtered options.",
+    "recommendation": "Use `useMemo` to memoize the filtered options based on the selected spaces and the original options.",
+    "current_code": "const filteredOptions = options.filter(\n\t(option) => !selectedSpaces.includes(parseInt(option.value)),\n);",
+    "suggested_code": "const filteredOptions = useMemo(() => options.filter(\n\t(option) => !selectedSpaces.includes(parseInt(option.value)),\n),[options, selectedSpaces]);",
+    "file_path": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 55,
+    "end_line": 57,
+    "sentiment": "neutral",
+    "severity": 7
+  },
+  {
+    "category": "Error Handling and Logging",
+    "description": "Lack of error handling for asynchronous operations.",
+    "impact": "high",
+    "rationale": "The onSubmit function is called asynchronously without any error handling, which can lead to unhandled promise rejections and a poor user experience.",
+    "recommendation": "Wrap the onSubmit call in a try-catch block to handle potential errors gracefully.",
+    "current_code": "onClick={() => onSubmit(inputValue)}",
+    "suggested_code": "onClick={async () =>{\n\ttry{\n\t\tawait onSubmit(inputValue);\n\t}catch (error){\n\t\tconsole.error('Error submitting:', error);\n\t}\n}}",
+    "file_path": "packages/ui/shadcn/combobox.tsx",
+    "start_line": 98,
+    "end_line": 98,
+    "sentiment": "negative",
+    "severity": 8
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/review.md b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/review.md
new file mode 100644
index 00000000..0403dc15
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/review.md
@@ -0,0 +1,33 @@
+PR URL: https://github.com/supermemoryai/supermemory/pull/232
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 11
+- Critical: 0
+- Important: 0
+- Minor: 0
+- Files Affected: 4
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 15508, "completion_tokens": 2390, "total_tokens": 17898}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/comments.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/comments.json
new file mode 100644
index 00000000..0637a088
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/comments.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/issues.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/issues.json
new file mode 100644
index 00000000..296cc3b7
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/issues.json
@@ -0,0 +1,58 @@
+[
+  {
+    "category": "Code Structure and Design",
+    "description": "Inconsistent naming conventions for parameters.",
+    "impact": "medium",
+    "rationale": "Inconsistent naming can lead to confusion and make the code harder to read and maintain. For example, the parameter names in `generate_twitter_post` and `generate_linkedin_post` methods could be more consistent.",
+    "recommendation": "Use a consistent naming convention for parameters across similar methods.",
+    "current_code": "def generate_twitter_post(self, summary: Dict, user: Optional[str] = None):",
+    "suggested_code": "def generate_twitter_post(self, summary: Dict, user: Optional[str] = None):",
+    "file_path": "kaizen/reviewer/work_summarizer.py",
+    "start_line": 58,
+    "end_line": 58,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling and Logging",
+    "description": "Lack of error handling when parsing JSON data.",
+    "impact": "high",
+    "rationale": "If the JSON data is malformed, the application will raise an unhandled exception, which can lead to crashes. Proper error handling is essential to ensure robustness.",
+    "recommendation": "Wrap the JSON parsing in a try-except block to handle potential exceptions gracefully.",
+    "current_code": "parsed_data = json.loads(json_data)",
+    "suggested_code": "try:\n    parsed_data = json.loads(json_data)\nexcept json.JSONDecodeError as e:\n    logging.error(f'Failed to parse JSON:{e}')\n    return None",
+    "file_path": "kaizen/helpers/parser.py",
+    "start_line": 47,
+    "end_line": 48,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Performance Issues",
+    "description": "Potential inefficiency in string formatting.",
+    "impact": "medium",
+    "rationale": "Using f-strings for logging can be inefficient if the logging level is set to a level higher than the one being used. This can lead to unnecessary string formatting operations.",
+    "recommendation": "Use logging methods that include the level check, such as `logging.debug` instead of f-strings directly.",
+    "current_code": "print(f\" Work Summary: \n{summary}\n\")",
+    "suggested_code": "logging.debug(\" Work Summary: \n %s\", summary)",
+    "file_path": "examples/work_summarizer/main.py",
+    "start_line": 62,
+    "end_line": 62,
+    "sentiment": "neutral",
+    "severity": 4
+  },
+  {
+    "category": "Resource Management",
+    "description": "Unclosed resources in the context of file handling.",
+    "impact": "high",
+    "rationale": "If there are any file operations in the code (not shown in the provided patch), failing to close files can lead to resource leaks.",
+    "recommendation": "Ensure that all file operations are wrapped in a context manager (with statement) to guarantee closure.",
+    "current_code": "file = open('example.txt', 'r')",
+    "suggested_code": "with open('example.txt', 'r') as file:\n    data = file.read()",
+    "file_path": "kaizen/reviewer/code_review.py",
+    "start_line": 14,
+    "end_line": 14,
+    "sentiment": "negative",
+    "severity": 6
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/review.md b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/review.md
new file mode 100644
index 00000000..4e6e60f3
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/review.md
@@ -0,0 +1,33 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/252
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 4
+- Critical: 0
+- Important: 0
+- Minor: 0
+- Files Affected: 4
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 3757, "completion_tokens": 786, "total_tokens": 4543}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/comments.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/comments.json
new file mode 100644
index 00000000..0637a088
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/comments.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/issues.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/issues.json
new file mode 100644
index 00000000..67c6897f
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/issues.json
@@ -0,0 +1,58 @@
+[
+  {
+    "category": "Code Structure and Design",
+    "description": "Inconsistent naming conventions and prompt usage.",
+    "impact": "medium",
+    "rationale": "The change from `CODE_REVIEW_SYSTEM_PROMPT` to `PR_DESCRIPTION_SYSTEM_PROMPT` is a good improvement for clarity. However, there are still remnants of the old naming conventions that could lead to confusion.",
+    "recommendation": "Ensure all prompt variables follow a consistent naming convention that clearly reflects their purpose. Consider refactoring any remaining instances of old variable names.",
+    "current_code": "self.provider.system_prompt = CODE_REVIEW_SYSTEM_PROMPT",
+    "suggested_code": "self.provider.system_prompt = PR_DESCRIPTION_SYSTEM_PROMPT",
+    "file_path": "kaizen/generator/pr_description.py",
+    "start_line": 29,
+    "end_line": 29,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling and Logging",
+    "description": "Lack of specific exception handling.",
+    "impact": "high",
+    "rationale": "Raising a generic Exception can obscure the root cause of issues and complicate debugging. Specific exceptions provide clearer context for errors.",
+    "recommendation": "Use more specific exception types or create custom exceptions to provide better context. Additionally, consider logging the error details for easier troubleshooting.",
+    "current_code": "raise Exception(\"Both diff_text and pull_request_files are empty!\")",
+    "suggested_code": "raise ValueError(\"Both diff_text and pull_request_files are empty!\")",
+    "file_path": "kaizen/generator/pr_description.py",
+    "start_line": 51,
+    "end_line": 51,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Performance Issues",
+    "description": "Potential inefficiencies in processing logic.",
+    "impact": "medium",
+    "rationale": "The `_process_full_diff` and `_process_files` methods may be called in scenarios where they are not necessary, leading to unnecessary computations.",
+    "recommendation": "Implement checks to ensure that these methods are only called when required, potentially using early returns to simplify logic.",
+    "current_code": "if diff_text and self.provider.is_inside_token_limit(PROMPT=prompt):",
+    "suggested_code": "if not diff_text or not self.provider.is_inside_token_limit(PROMPT=prompt): return None",
+    "file_path": "kaizen/generator/pr_description.py",
+    "start_line": 53,
+    "end_line": 53,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Code Structure and Design",
+    "description": "Code duplication in prompt generation.",
+    "impact": "medium",
+    "rationale": "There are similar patterns in how prompts are generated across different methods, leading to code duplication.",
+    "recommendation": "Consider creating a utility function to handle prompt generation to reduce duplication and improve maintainability.",
+    "current_code": "prompt = PR_DESCRIPTION_PROMPT.format(...)\nprompt = MERGE_PR_DESCRIPTION_PROMPT.format(...)",
+    "suggested_code": "def generate_prompt(template, **kwargs): return template.format(**kwargs)\nprompt = generate_prompt(PR_DESCRIPTION_PROMPT, ...)",
+    "file_path": "kaizen/generator/pr_description.py",
+    "start_line": 45,
+    "end_line": 45,
+    "sentiment": "neutral",
+    "severity": 5
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/review.md b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/review.md
new file mode 100644
index 00000000..32e3df81
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/review.md
@@ -0,0 +1,33 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/335
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 4
+- Critical: 0
+- Important: 0
+- Minor: 0
+- Files Affected: 1
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 6622, "completion_tokens": 774, "total_tokens": 7396}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/comments.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/comments.json
new file mode 100644
index 00000000..252fbeac
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/comments.json
@@ -0,0 +1,30 @@
+[
+  {
+    "category": "Error Handling and Logging",
+    "description": "Lack of input validation for create_folder function.",
+    "impact": "critical",
+    "rationale": "The create_folder function should validate inputs to prevent unexpected behavior or crashes. The tests now include checks for empty and invalid paths, which is a good start, but the function itself should also handle these cases.",
+    "recommendation": "Implement input validation in the create_folder function to ensure that it handles invalid inputs gracefully.",
+    "current_code": "def create_folder(folder_path):\n    # Function logic here",
+    "suggested_code": "def create_folder(folder_path):\n    if not folder_path:\n        raise ValueError('Folder path cannot be empty')\n    # Function logic here",
+    "file_path": "kaizen/helpers/output.py",
+    "start_line": 1,
+    "end_line": 1,
+    "sentiment": "negative",
+    "severity": 9
+  },
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "11",
+    "end_line": "11",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/issues.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/issues.json
new file mode 100644
index 00000000..548c5215
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/issues.json
@@ -0,0 +1,338 @@
+[
+  {
+    "category": "Code Structure and Design",
+    "description": "Inconsistent use of mocking in tests.",
+    "impact": "high",
+    "rationale": "The original code used decorators for mocking which can lead to confusion and inconsistency in test setup. The new approach uses fixtures, which improves readability and maintainability by clearly defining the mock setup in one place.",
+    "recommendation": "Continue using fixtures for mocking to maintain consistency across tests. This will also help in reducing boilerplate code.",
+    "current_code": "@mock.patch('kaizen.helpers.output.os.makedirs')\n@mock.patch('kaizen.helpers.output.os.path.exists')\ndef test_create_folder_success(mock_logger, mock_exists, mock_makedirs):",
+    "suggested_code": "@pytest.fixture\ndef mock_os_path_exists():\n    with mock.patch('os.path.exists') as mock_exists:\n        yield mock_exists",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_folder.py",
+    "start_line": 6,
+    "end_line": 8,
+    "sentiment": "positive",
+    "severity": 7
+  },
+  {
+    "category": "Error Handling and Logging",
+    "description": "Lack of input validation for create_folder function.",
+    "impact": "critical",
+    "rationale": "The create_folder function should validate inputs to prevent unexpected behavior or crashes. The tests now include checks for empty and invalid paths, which is a good start, but the function itself should also handle these cases.",
+    "recommendation": "Implement input validation in the create_folder function to ensure that it handles invalid inputs gracefully.",
+    "current_code": "def create_folder(folder_path):\n    # Function logic here",
+    "suggested_code": "def create_folder(folder_path):\n    if not folder_path:\n        raise ValueError('Folder path cannot be empty')\n    # Function logic here",
+    "file_path": "kaizen/helpers/output.py",
+    "start_line": 1,
+    "end_line": 1,
+    "sentiment": "negative",
+    "severity": 9
+  },
+  {
+    "category": "Performance Issues",
+    "description": "Potential performance issue with deep folder creation.",
+    "impact": "medium",
+    "rationale": "Creating deeply nested folders can lead to performance degradation if not handled properly. The tests should ensure that the function can handle such cases efficiently.",
+    "recommendation": "Consider implementing a method to create nested directories in a single call if the underlying OS supports it, or handle it in a way that minimizes performance hits.",
+    "current_code": "create_folder(folder_path)",
+    "suggested_code": "os.makedirs(folder_path, exist_ok=True)",
+    "file_path": "kaizen/helpers/output.py",
+    "start_line": 1,
+    "end_line": 1,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Code Structure and Design",
+    "description": "Duplication of test cases.",
+    "impact": "medium",
+    "rationale": "There are multiple test cases that check similar functionality with slight variations. This can lead to maintenance challenges and bloated test suites.",
+    "recommendation": "Consolidate similar test cases into parameterized tests to improve maintainability and clarity.",
+    "current_code": "def test_create_pr_description(desc, original_desc, expected):\n    assert create_pr_description(desc, original_desc) == expected",
+    "suggested_code": "@pytest.mark.parametrize('desc, original_desc, expected',[\n    ('This is a test description.', 'This is the original description.', 'Expected output'),\n    # Other cases\n])\ndef test_create_pr_description_normal_and_edge_cases(desc, original_desc, expected):\n    assert create_pr_description(desc, original_desc) == expected",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
+    "start_line": 8,
+    "end_line": 8,
+    "sentiment": "positive",
+    "severity": 5
+  },
+  {
+    "category": "Code Structure and Design",
+    "description": "Inconsistent naming conventions for severity levels.",
+    "impact": "medium",
+    "rationale": "The code uses both 'severity_level' and 'severity' interchangeably, which can lead to confusion and inconsistency in the codebase.",
+    "recommendation": "Standardize the naming convention for severity across the codebase.",
+    "current_code": "\"severity_level\": 9,",
+    "suggested_code": "\"severity\": 9,",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 30,
+    "end_line": 30,
+    "sentiment": "negative",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling and Logging",
+    "description": "Missing error handling for potential issues in input data.",
+    "impact": "high",
+    "rationale": "The current implementation does not validate the input data structure or handle cases where expected fields are missing, which could lead to runtime errors.",
+    "recommendation": "Implement input validation to ensure all required fields are present before processing.",
+    "current_code": "def test_reviews_with_missing_fields():",
+    "suggested_code": "def test_reviews_with_missing_fields():\n    # Validate input data structure here\n    assert 'comment' in review and 'reason' in review, 'Missing required fields'",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 144,
+    "end_line": 144,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Performance Issues",
+    "description": "Repeated use of PR_COLLAPSIBLE_TEMPLATE can lead to performance overhead.",
+    "impact": "medium",
+    "rationale": "The template is being formatted multiple times within loops, which could be optimized by caching or reusing the formatted string.",
+    "recommendation": "Consider pre-formatting the template outside of loops or using a more efficient string handling method.",
+    "current_code": "PR_COLLAPSIBLE_TEMPLATE.format(...)",
+    "suggested_code": "formatted_template = PR_COLLAPSIBLE_TEMPLATE.format(...)\n# Use formatted_template in the loop",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 85,
+    "end_line": 85,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Readability",
+    "description": "Use of inline comments to explain complex logic.",
+    "impact": "low",
+    "rationale": "While the code is generally readable, adding comments to explain the purpose of complex blocks can improve maintainability.",
+    "recommendation": "Add inline comments to clarify the purpose of key sections of the code.",
+    "current_code": "def test_multiple_topics_multiple_reviews(setup_multiple_topics_multiple_reviews):",
+    "suggested_code": "def test_multiple_topics_multiple_reviews(setup_multiple_topics_multiple_reviews):\n    # Test case for multiple topics and reviews",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
+    "start_line": 99,
+    "end_line": 99,
+    "sentiment": "positive",
+    "severity": 3
+  },
+  {
+    "category": "Code Structure and Design",
+    "description": "Inconsistent naming conventions for importance levels.",
+    "impact": "medium",
+    "rationale": "The importance levels in the JSON structure are sometimes written as 'High' and other times as 'high'. This inconsistency can lead to confusion and bugs when processing these values.",
+    "recommendation": "Standardize the naming convention for importance levels across all tests.",
+    "current_code": "\"importance\": \"High\"",
+    "suggested_code": "\"importance\": \"high\"",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 32,
+    "end_line": 32,
+    "sentiment": "negative",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling and Logging",
+    "description": "Lack of logging for error cases in file operations.",
+    "impact": "high",
+    "rationale": "While the tests handle exceptions, there is no logging to capture the context of the errors, making it harder to debug issues when they arise in production.",
+    "recommendation": "Add logging statements to capture exceptions and provide context for failures.",
+    "current_code": "with pytest.raises(PermissionError):",
+    "suggested_code": "try:\n    create_test_files(json_tests, tmp_path)\nexcept PermissionError as e:\n    mock_logger.error(f'Permission error:{e}')\n    raise",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 194,
+    "end_line": 195,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Performance Issues",
+    "description": "Inefficient handling of long test names.",
+    "impact": "medium",
+    "rationale": "The test for very long test names does not account for potential performance issues when generating and checking file names. This could lead to unnecessary resource consumption.",
+    "recommendation": "Implement a check to limit the length of generated test names before creating files.",
+    "current_code": "long_test_name = \"Test \" + \"Example \" * 50",
+    "suggested_code": "long_test_name = \"Test Example Long Name\"[:250]",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 131,
+    "end_line": 131,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Code Structure and Design",
+    "description": "Redundant mock patches in tests.",
+    "impact": "medium",
+    "rationale": "The same mock patches are repeated across multiple tests, leading to code duplication and increased maintenance effort.",
+    "recommendation": "Use a fixture to apply common mock patches to reduce redundancy.",
+    "current_code": "@mock.patch(\"kaizen.helpers.output.create_folder\")\n@mock.patch(\"kaizen.helpers.output.general.clean_python_code\")\n@mock.patch(\"kaizen.helpers.output.logger\")",
+    "suggested_code": "@pytest.fixture\ndef mock_dependencies():\n    with mock.patch(\"kaizen.helpers.output.create_folder\") as mock_create_folder,\n         mock.patch(\"kaizen.helpers.output.general.clean_python_code\") as mock_clean_python_code,\n         mock.patch(\"kaizen.helpers.output.logger\") as mock_logger:\n        yield mock_create_folder, mock_clean_python_code, mock_logger",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
+    "start_line": 48,
+    "end_line": 50,
+    "sentiment": "neutral",
+    "severity": 4
+  },
+  {
+    "category": "Code Structure and Design",
+    "description": "Inconsistent use of async/await in test functions.",
+    "impact": "high",
+    "rationale": "Some test functions are defined as async but do not use await properly, which can lead to unexpected behavior and test failures.",
+    "recommendation": "Ensure all async functions use await where necessary to handle asynchronous calls correctly.",
+    "current_code": "result = get_web_html(url)",
+    "suggested_code": "result = await get_web_html(url)",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 92,
+    "end_line": 92,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Error Handling and Logging",
+    "description": "Lack of error handling for network errors in async tests.",
+    "impact": "high",
+    "rationale": "The tests do not handle exceptions that may arise from network calls, which can lead to false negatives in test results.",
+    "recommendation": "Add exception handling in the async tests to ensure that network errors are properly caught and asserted.",
+    "current_code": "result = await get_web_html(url)",
+    "suggested_code": "with pytest.raises(Exception, match='Network error'):\n    await get_web_html(url)",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 88,
+    "end_line": 90,
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Performance Issues",
+    "description": "Potential inefficiency in handling large HTML content.",
+    "impact": "medium",
+    "rationale": "The test for large HTML content may lead to performance issues if the content size grows significantly, affecting test execution time.",
+    "recommendation": "Consider using a smaller sample size or mocking the HTML content for performance-sensitive tests.",
+    "current_code": "large_html_content = '<html><body>' + '<p>Test</p>' * 10000 + '</body></html>'",
+    "suggested_code": "large_html_content = '<html><body><p>Test</p></body></html>'",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 94,
+    "end_line": 95,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Code Structure and Design",
+    "description": "Use of hard-coded values in tests.",
+    "impact": "medium",
+    "rationale": "Hard-coded values can lead to maintenance issues and reduce the flexibility of the tests.",
+    "recommendation": "Define constants for commonly used values to improve maintainability and readability.",
+    "current_code": "url = 'https://cloudcode.ai'",
+    "suggested_code": "BASE_URL = 'https://cloudcode.ai'\nurl = BASE_URL",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "start_line": 48,
+    "end_line": 48,
+    "sentiment": "positive",
+    "severity": 4
+  },
+  {
+    "category": "Code Structure and Design",
+    "description": "Inconsistent naming conventions for constants.",
+    "impact": "medium",
+    "rationale": "Constants like SUPPORTED_LANGUAGES and LANGUAGE_PROMPTS should follow a consistent naming convention, typically ALL_CAPS, to improve readability and maintainability.",
+    "recommendation": "Rename constants to follow the ALL_CAPS convention for clarity.",
+    "current_code": "SUPPORTED_LANGUAGES ={...}",
+    "suggested_code": "SUPPORTED_LANGUAGES ={...}",
+    "file_path": "kaizen/generator/unit_test.py",
+    "start_line": 24,
+    "end_line": 40,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Error Handling and Logging",
+    "description": "Lack of error handling in file operations.",
+    "impact": "high",
+    "rationale": "File operations can fail due to various reasons (e.g., file not found, permission issues). Proper error handling should be implemented to avoid crashes and provide meaningful feedback.",
+    "recommendation": "Wrap file operations in try-except blocks to handle potential exceptions gracefully.",
+    "current_code": "with open(file_path, 'r') as file: ...",
+    "suggested_code": "try:\n    with open(file_path, 'r') as file:\n        return file.read()\nexcept IOError as e:\n    self.logger.error(f'Error reading file{file_path}:{e}')\n    return None",
+    "file_path": "kaizen/generator/unit_test.py",
+    "start_line": 108,
+    "end_line": 110,
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Performance Issues",
+    "description": "Potential inefficiency in string operations.",
+    "impact": "medium",
+    "rationale": "Using string concatenation in a loop can lead to performance issues, especially with large datasets. It's better to use a list and join at the end.",
+    "recommendation": "Change string concatenation to use a list and join method for better performance.",
+    "current_code": "formatted_scenarios += f'{category.replace('_', ' ').title()}:\n'\nformatted_scenarios += '\n'.join(f'-{case}' for case in cases)",
+    "suggested_code": "formatted_scenarios =[]\nformatted_scenarios.append(f'{category.replace('_', ' ').title()}:\n')\nformatted_scenarios.append('\n'.join(f'-{case}' for case in cases))\nreturn ''.join(formatted_scenarios)",
+    "file_path": "kaizen/generator/unit_test.py",
+    "start_line": 240,
+    "end_line": 250,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Error Handling and Logging",
+    "description": "Inconsistent logging practices.",
+    "impact": "medium",
+    "rationale": "Some methods log their actions while others do not, leading to inconsistent behavior and difficulty in tracing execution flow.",
+    "recommendation": "Ensure that all significant actions, especially those that could fail, are logged consistently.",
+    "current_code": "print(f'Error: Could not generate tests for{file_path}:{e}')",
+    "suggested_code": "self.logger.error(f'Could not generate tests for{file_path}:{e}')",
+    "file_path": "kaizen/generator/unit_test.py",
+    "start_line": 73,
+    "end_line": 74,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Code Structure and Design",
+    "description": "Hardcoded logger settings and lack of configurability.",
+    "impact": "high",
+    "rationale": "The function `set_all_loggers_to_ERROR` hardcodes the log level for specific loggers, which reduces flexibility and makes it difficult to change logging behavior without modifying the code.",
+    "recommendation": "Consider using a configuration file or environment variables to manage logger settings dynamically.",
+    "current_code": "logging.getLogger(\"LiteLLM\").setLevel(logging.ERROR)",
+    "suggested_code": "logging.getLogger(\"LiteLLM\").setLevel(os.environ.get(\"LITELLM_LOG_LEVEL\", logging.ERROR))",
+    "file_path": "kaizen/llms/provider.py",
+    "start_line": 25,
+    "end_line": 27,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Error Handling and Logging",
+    "description": "Lack of error handling when setting logger levels.",
+    "impact": "medium",
+    "rationale": "If there are issues with setting the logger levels (e.g., invalid logger names), the application will fail without any meaningful error message.",
+    "recommendation": "Add try-except blocks around logger level settings to handle potential exceptions gracefully.",
+    "current_code": "logging.getLogger(name).setLevel(logging.ERROR)",
+    "suggested_code": "try:\n    logging.getLogger(name).setLevel(logging.ERROR)\nexcept Exception as e:\n    print(f\"Failed to set logger level for{name}:{e}\")",
+    "file_path": "kaizen/llms/provider.py",
+    "start_line": 18,
+    "end_line": 18,
+    "sentiment": "negative",
+    "severity": 6
+  },
+  {
+    "category": "Readability",
+    "description": "Inconsistent naming conventions.",
+    "impact": "medium",
+    "rationale": "The function name `set_all_loggers_to_ERROR` uses snake_case, while the logger names use CamelCase. This inconsistency can lead to confusion.",
+    "recommendation": "Standardize naming conventions across the codebase to improve readability and maintainability.",
+    "current_code": "set_all_loggers_to_ERROR()",
+    "suggested_code": "set_all_loggers_to_error()",
+    "file_path": "kaizen/llms/provider.py",
+    "start_line": 23,
+    "end_line": 23,
+    "sentiment": "neutral",
+    "severity": 4
+  },
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "11",
+    "end_line": "11",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/review.md b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/review.md
new file mode 100644
index 00000000..2cffb7a1
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/review.md
@@ -0,0 +1,81 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/400
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 24
+- Critical: 2
+- Important: 0
+- Minor: 0
+- Files Affected: 9
+## 🏆 Code Quality
+[███████████████░░░░░] 75% (Fair)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Error Handling and Logging (2 issues)</strong></summary>
+
+### 1. Lack of input validation for create_folder function.
+📁 **File:** `kaizen/helpers/output.py:1`
+⚖️ **Severity:** 9/10
+🔍 **Description:** Lack of input validation for create_folder function.
+💡 **Solution:** 
+
+**Current Code:**
+```python
+def create_folder(folder_path):
+    # Function logic here
+```
+
+**Suggested Code:**
+```python
+
+```
+
+### 2. Changes made to sensitive file
+📁 **File:** `config.json:11`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (1 issues)</strong></summary>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 36120, "completion_tokens": 4596, "total_tokens": 40716}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/comments.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/comments.json
new file mode 100644
index 00000000..0637a088
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/comments.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/issues.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/issues.json
new file mode 100644
index 00000000..65970832
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/issues.json
@@ -0,0 +1,44 @@
+[
+  {
+    "category": "Code Structure and Design",
+    "description": "Potential for duplicate embeddings in the database.",
+    "impact": "high",
+    "rationale": "The comment 'TODO: DONT PUSH DUPLICATE' indicates that there is a risk of inserting duplicate embeddings into the database. This can lead to data inconsistency and increased storage usage.",
+    "recommendation": "Implement a check before inserting embeddings to ensure duplicates are not added.",
+    "current_code": "embedding_query = text(\n                  \"\"\"\n",
+    "suggested_code": "if not self.is_duplicate_embedding(embedding):\n    # Proceed with insertion logic here\n",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 157,
+    "end_line": 157,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Error Handling and Logging",
+    "description": "Lack of error handling for repository setup.",
+    "impact": "medium",
+    "rationale": "The `setup_repository` method is called without any error handling. If this operation fails, it could lead to unhandled exceptions and application crashes.",
+    "recommendation": "Wrap the `setup_repository` call in a try-except block to handle potential exceptions gracefully.",
+    "current_code": "analyzer.setup_repository(\"./github_app/\")",
+    "suggested_code": "try:\n    analyzer.setup_repository(\"./github_app/\")\nexcept Exception as e:\n    logging.error(f\"Failed to set up repository:{e}\")\n",
+    "file_path": "examples/ragify_codebase/main.py",
+    "start_line": 7,
+    "end_line": 7,
+    "sentiment": "negative",
+    "severity": 6
+  },
+  {
+    "category": "Performance Issues",
+    "description": "Potential inefficiency in embedding retrieval.",
+    "impact": "medium",
+    "rationale": "The line `embedding = embedding[0][\"embedding\"]` assumes that the embedding list will always have at least one element. If it doesn't, this could lead to an index error.",
+    "recommendation": "Check if the embedding list is not empty before accessing its first element.",
+    "current_code": "embedding = embedding[0][\"embedding\"]",
+    "suggested_code": "if embedding:\n    embedding = embedding[0][\"embedding\"]\nelse:\n    raise ValueError(\"No embeddings found\")\n",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
+    "start_line": 155,
+    "end_line": 155,
+    "sentiment": "negative",
+    "severity": 5
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/review.md b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/review.md
new file mode 100644
index 00000000..3219c661
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/review.md
@@ -0,0 +1,33 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/440
+
+# 🔍 Code Review Summary
+
+✅ **All Clear:** This commit looks good! 👍
+
+## 📊 Stats
+- Total Issues: 3
+- Critical: 0
+- Important: 0
+- Minor: 0
+- Files Affected: 2
+## 🏆 Code Quality
+[███████████████░░░░░] 75% (Fair)
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 1235, "completion_tokens": 619, "total_tokens": 1854}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/comments.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/comments.json
new file mode 100644
index 00000000..9cbf8d49
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/comments.json
@@ -0,0 +1,16 @@
+[
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/issues.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/issues.json
new file mode 100644
index 00000000..28b2c5e9
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/issues.json
@@ -0,0 +1,72 @@
+[
+  {
+    "category": "Error Handling and Logging",
+    "description": "Broad exception handling without specific error management.",
+    "impact": "high",
+    "rationale": "Using a generic `except Exception` can mask underlying issues and makes debugging difficult. It\u2019s crucial to handle specific exceptions to ensure that errors are logged and managed appropriately.",
+    "recommendation": "Refine the exception handling to catch specific exceptions and log meaningful error messages.",
+    "current_code": "except Exception:\n    print(\"Error\")",
+    "suggested_code": "except KeyError:\n    raise ValueError(f\"Invalid confidence level:{min_confidence}\")\nexcept Exception as e:\n    logger.error(f\"Unexpected error:{str(e)}\")",
+    "file_path": "github_app/github_helper/pull_requests.py",
+    "start_line": 140,
+    "end_line": 141,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Code Structure and Design",
+    "description": "Potential code duplication in sorting logic.",
+    "impact": "medium",
+    "rationale": "The sorting logic for files appears to be implemented in a way that could lead to duplication if similar logic is used elsewhere. This can lead to maintenance challenges.",
+    "recommendation": "Consider using Python's built-in sorting capabilities or refactor the sorting logic into a separate utility function to promote reusability.",
+    "current_code": "for file in files:\n    min_index = len(sorted_files)\n    file_name = file[\"filename\"]\n    for i, sorted_file in enumerate(sorted_files):\n        if file_name < sorted_file[\"filename\"]:\n            min_index = i\n            break\n    sorted_files.insert(min_index, file)",
+    "suggested_code": "sorted_files = sorted(files, key=lambda f: f[\"filename\"])",
+    "file_path": "github_app/github_helper/pull_requests.py",
+    "start_line": 184,
+    "end_line": 196,
+    "sentiment": "neutral",
+    "severity": 5
+  },
+  {
+    "category": "Performance Issues",
+    "description": "Inefficient file sorting algorithm.",
+    "impact": "medium",
+    "rationale": "The current sorting algorithm has a time complexity of O(n^2) due to the nested loops. This can lead to performance issues with larger datasets.",
+    "recommendation": "Use Python's built-in sorting functions which are optimized and have a time complexity of O(n log n).",
+    "current_code": "for file in files:\n    min_index = len(sorted_files)\n    for i, sorted_file in enumerate(sorted_files):\n        if file_name < sorted_file[\"filename\"]:\n            min_index = i\n            break\n    sorted_files.insert(min_index, file)",
+    "suggested_code": "sorted_files = sorted(files, key=lambda f: f[\"filename\"])",
+    "file_path": "github_app/github_helper/pull_requests.py",
+    "start_line": 184,
+    "end_line": 196,
+    "sentiment": "neutral",
+    "severity": 6
+  },
+  {
+    "category": "Code Structure and Design",
+    "description": "Inconsistent naming conventions.",
+    "impact": "low",
+    "rationale": "The naming of variables such as `pr_files` and `pull_request_files` can lead to confusion. Consistent naming conventions improve readability and maintainability.",
+    "recommendation": "Standardize variable names throughout the codebase to follow a consistent naming convention.",
+    "current_code": "pull_request_files=pr_files",
+    "suggested_code": "pull_request_files=sorted_files",
+    "file_path": "github_app/github_helper/pull_requests.py",
+    "start_line": 55,
+    "end_line": 56,
+    "sentiment": "positive",
+    "severity": 4
+  },
+  {
+    "category": "Configuration",
+    "description": "Changes made to sensitive file",
+    "impact": "critical",
+    "recommendation": "Changes were made to config.json, which needs review",
+    "current_code": "NA",
+    "fixed_code": "",
+    "start_line": "1",
+    "end_line": "1",
+    "side": "RIGHT",
+    "file_path": "config.json",
+    "sentiment": "negative",
+    "severity": 10
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/review.md b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/review.md
new file mode 100644
index 00000000..30369f46
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/review.md
@@ -0,0 +1,64 @@
+PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/476
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 5
+- Critical: 1
+- Important: 0
+- Minor: 0
+- Files Affected: 2
+## 🏆 Code Quality
+[█████████████████░░░] 85% (Good)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Configuration (1 issues)</strong></summary>
+
+### 1. Changes made to sensitive file
+📁 **File:** `config.json:1`
+⚖️ **Severity:** 10/10
+🔍 **Description:** Changes made to sensitive file
+💡 **Solution:** 
+
+**Current Code:**
+```python
+NA
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (1 issues)</strong></summary>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 3878, "completion_tokens": 867, "total_tokens": 4745}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/comments.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/comments.json
new file mode 100644
index 00000000..26261207
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/comments.json
@@ -0,0 +1,16 @@
+[
+  {
+    "category": "Error Handling and Logging",
+    "description": "Potential division by zero error.",
+    "impact": "critical",
+    "rationale": "If 'total_tokens' is zero, it will cause a division by zero error when calculating the summary statistics.",
+    "recommendation": "Add a check to ensure 'total_tokens' is not zero before performing calculations.",
+    "current_code": "print(f'Total tokens used:{total_tokens:,}')",
+    "suggested_code": "if total_tokens > 0:\n    print(f'Total tokens used:{total_tokens:,}')\nelse:\n    print('No tokens used.')",
+    "file_path": "main.py",
+    "start_line": 159,
+    "end_line": 161,
+    "sentiment": "negative",
+    "severity": 9
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/issues.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/issues.json
new file mode 100644
index 00000000..f8fd1693
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/issues.json
@@ -0,0 +1,72 @@
+[
+  {
+    "category": "Code Structure and Design",
+    "description": "Unused import statement.",
+    "impact": "low",
+    "rationale": "The import of 'random' is unnecessary and can clutter the codebase, potentially leading to confusion.",
+    "recommendation": "Remove the unused import statement to improve code clarity.",
+    "current_code": "import random  # Unused import",
+    "suggested_code": "",
+    "file_path": "main.py",
+    "start_line": 8,
+    "end_line": 8,
+    "sentiment": "negative",
+    "severity": 3
+  },
+  {
+    "category": "Error Handling and Logging",
+    "description": "Silent failure during JSON parsing.",
+    "impact": "high",
+    "rationale": "The code silently fails when JSON parsing fails, which can lead to undetected issues in the application. Proper logging or exception handling should be implemented.",
+    "recommendation": "Log the error message or raise an exception to ensure issues are visible during execution.",
+    "current_code": "except json.JSONDecodeError:\n        # Critical: Silent failure without logging",
+    "suggested_code": "except json.JSONDecodeError as e:\n        print(f'Failed to parse content for applicant:{e}')",
+    "file_path": "main.py",
+    "start_line": 82,
+    "end_line": 84,
+    "sentiment": "negative",
+    "severity": 7
+  },
+  {
+    "category": "Error Handling and Logging",
+    "description": "Potential division by zero error.",
+    "impact": "critical",
+    "rationale": "If 'total_tokens' is zero, it will cause a division by zero error when calculating the summary statistics.",
+    "recommendation": "Add a check to ensure 'total_tokens' is not zero before performing calculations.",
+    "current_code": "print(f'Total tokens used:{total_tokens:,}')",
+    "suggested_code": "if total_tokens > 0:\n    print(f'Total tokens used:{total_tokens:,}')\nelse:\n    print('No tokens used.')",
+    "file_path": "main.py",
+    "start_line": 159,
+    "end_line": 161,
+    "sentiment": "negative",
+    "severity": 9
+  },
+  {
+    "category": "Error Handling and Logging",
+    "description": "Lack of error handling for file not found.",
+    "impact": "high",
+    "rationale": "If the user inputs a file name that does not exist, the program will crash without providing a user-friendly message.",
+    "recommendation": "Implement a try-except block around the file reading operation to handle file not found errors gracefully.",
+    "current_code": "main(input_file)",
+    "suggested_code": "try:\n    main(input_file)\nexcept FileNotFoundError:\n    print('Error: The specified file was not found.')",
+    "file_path": "main.py",
+    "start_line": 175,
+    "end_line": 175,
+    "sentiment": "negative",
+    "severity": 8
+  },
+  {
+    "category": "Performance Issues",
+    "description": "Inefficient way to print progress.",
+    "impact": "medium",
+    "rationale": "The current method of printing progress can be inefficient, especially with larger datasets, as it updates the console frequently.",
+    "recommendation": "Consider using a logging library or a more efficient method to report progress.",
+    "current_code": "print(f'\\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}', end='', flush=True)",
+    "suggested_code": "if index % 10 == 0:  # Update every 10 applicants\n    print(f'Progress:[{('=' * int(50 * progress)):<50}]{progress:.0%}')",
+    "file_path": "main.py",
+    "start_line": 121,
+    "end_line": 122,
+    "sentiment": "neutral",
+    "severity": 5
+  }
+]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/review.md b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/review.md
new file mode 100644
index 00000000..a1e7c56a
--- /dev/null
+++ b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/review.md
@@ -0,0 +1,64 @@
+PR URL: https://github.com/sauravpanda/applicant-screening/pull/5
+
+# 🔍 Code Review Summary
+
+❗ **Attention Required:** This push has potential issues. 🚨
+
+## 📊 Stats
+- Total Issues: 5
+- Critical: 1
+- Important: 0
+- Minor: 0
+- Files Affected: 1
+## 🏆 Code Quality
+[███████████████░░░░░] 75% (Fair)
+
+## 🚨 Critical Issues
+
+<details>
+<summary><strong>Error Handling and Logging (1 issues)</strong></summary>
+
+### 1. Potential division by zero error.
+📁 **File:** `main.py:159`
+⚖️ **Severity:** 9/10
+🔍 **Description:** Potential division by zero error.
+💡 **Solution:** 
+
+**Current Code:**
+```python
+print(f'Total tokens used:{total_tokens:,}')
+```
+
+**Suggested Code:**
+```python
+
+```
+
+</details>
+
+## 📝 Minor Notes
+Additional small points that you might want to consider:
+
+<details>
+<summary><strong>Click to expand (1 issues)</strong></summary>
+
+</details>
+
+---
+
+> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
+
+<details>
+<summary>Useful Commands</summary>
+
+- **Feedback:** Reply with `!feedback [your message]`
+- **Ask PR:** Reply with `!ask-pr [your question]`
+- **Review:** Reply with `!review`
+- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
+- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
+- **Update Tests:** Reply with `!unittest` to create a PR with test changes
+</details>
+
+
+----- Cost Usage (gpt-4o-mini)
+{"prompt_tokens": 6025, "completion_tokens": 900, "total_tokens": 6925}
\ No newline at end of file
diff --git a/.experiments/code_review/dataset/pr_222/issues.json b/.experiments/code_review/dataset/pr_222/issues.json
index c50260a4..61607c58 100644
--- a/.experiments/code_review/dataset/pr_222/issues.json
+++ b/.experiments/code_review/dataset/pr_222/issues.json
@@ -1,152 +1,152 @@
 [
   {
-    "topic": "Security",
-    "comment": "Hardcoding API keys in `config.json` can lead to security vulnerabilities.",
-    "confidence": "critical",
-    "reason": "Exposing API keys in the codebase can lead to unauthorized access. This was highlighted by multiple models and is a critical security issue.",
-    "solution": "Use environment variables to store API keys instead of hardcoding them.",
-    "actual_code": "\"api_key\": \"os.environ/AZURE_API_KEY\"",
+    "category": "Security",
+    "description": "Hardcoding API keys in `config.json` can lead to security vulnerabilities.",
+    "impact": "critical",
+    "rationale": "Exposing API keys in the codebase can lead to unauthorized access. This was highlighted by multiple models and is a critical security issue.",
+    "recommendation": "Use environment variables to store API keys instead of hardcoding them.",
+    "suggested_code": "\"api_key\": \"os.environ/AZURE_API_KEY\"",
     "fixed_code": "\"api_key\": \"${AZURE_API_KEY}\"",
-    "file_name": "config.json",
+    "file_path": "config.json",
     "start_line": 13,
     "end_line": 13,
     "side": "RIGHT",
     "sentiment": "negative",
-    "severity_level": 9
+    "severity": 9
   },
   {
-    "topic": "SQL Injection",
-    "comment": "Potential SQL injection vulnerability in the query construction.",
-    "confidence": "critical",
-    "reason": "Using string interpolation for SQL queries can lead to SQL injection attacks. This was identified by multiple models as a critical issue.",
-    "solution": "Use parameterized queries to avoid SQL injection vulnerabilities.",
-    "actual_code": "query = f\"\"\"\nSELECT \n    e.node_id,\n    e.text,\n    e.metadata,\n    1 - (e.embedding <=> %s::vector) as similarity\nFROM \n{self.table_name}e\nJOIN \n    function_abstractions fa ON e.node_id = fa.function_id::text\nJOIN \n    files f ON fa.file_id = f.file_id\nWHERE \n    f.repo_id = %s\nORDER BY \n    similarity DESC\nLIMIT \n    %s\n\"\"\"",
+    "category": "SQL Injection",
+    "description": "Potential SQL injection vulnerability in the query construction.",
+    "impact": "critical",
+    "rationale": "Using string interpolation for SQL queries can lead to SQL injection attacks. This was identified by multiple models as a critical issue.",
+    "recommendation": "Use parameterized queries to avoid SQL injection vulnerabilities.",
+    "suggested_code": "query = f\"\"\"\nSELECT \n    e.node_id,\n    e.text,\n    e.metadata,\n    1 - (e.embedding <=> %s::vector) as similarity\nFROM \n{self.table_name}e\nJOIN \n    function_abstractions fa ON e.node_id = fa.function_id::text\nJOIN \n    files f ON fa.file_id = f.file_id\nWHERE \n    f.repo_id = %s\nORDER BY \n    similarity DESC\nLIMIT \n    %s\n\"\"\"",
     "fixed_code": "query = \"\"\"\nSELECT \n    e.node_id,\n    e.text,\n    e.metadata,\n    1 - (e.embedding <=> %s::vector) as similarity\nFROM \n    %s e\nJOIN \n    function_abstractions fa ON e.node_id = fa.function_id::text\nJOIN \n    files f ON fa.file_id = f.file_id\nWHERE \n    f.repo_id = %s\nORDER BY \n    similarity DESC\nLIMIT \n    %s\n\"\"\"",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "file_path": "kaizen/retriever/custom_vector_store.py",
     "start_line": 19,
     "end_line": 37,
     "side": "RIGHT",
     "sentiment": "negative",
-    "severity_level": 9
+    "severity": 9
   },
   {
-    "topic": "Error Handling",
-    "comment": "Lack of error handling in database operations.",
-    "confidence": "important",
-    "reason": "Multiple models identified the need for better error handling in database operations to prevent crashes and improve debugging.",
-    "solution": "Add try-except blocks to handle potential database errors.",
-    "actual_code": "",
+    "category": "Error Handling",
+    "description": "Lack of error handling in database operations.",
+    "impact": "high",
+    "rationale": "Multiple models identified the need for better error handling in database operations to prevent crashes and improve debugging.",
+    "recommendation": "Add try-except blocks to handle potential database errors.",
+    "suggested_code": "",
     "fixed_code": "try:\n    with self.get_client() as client:\n        with client.cursor() as cur:\n            cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))\n            results = cur.fetchall()\nexcept Exception as e:\n    # Handle exception (e.g., log the error, re-raise, etc.)\n    raise e",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "file_path": "kaizen/retriever/custom_vector_store.py",
     "start_line": 39,
     "end_line": 42,
     "side": "RIGHT",
     "sentiment": "negative",
-    "severity_level": 7
+    "severity": 7
   },
   {
-    "topic": "Code Readability",
-    "comment": "The `chunk_code` function in `code_chunker.py` has nested functions and complex logic that can be refactored for better readability.",
-    "confidence": "important",
-    "reason": "Complex functions with nested logic can be hard to maintain and understand. This was noted by multiple models.",
-    "solution": "Refactor the `chunk_code` function to extract nested functions into separate helper functions.",
-    "actual_code": "",
+    "category": "Code Readability",
+    "description": "The `chunk_code` function in `code_chunker.py` has nested functions and complex logic that can be refactored for better readability.",
+    "impact": "high",
+    "rationale": "Complex functions with nested logic can be hard to maintain and understand. This was noted by multiple models.",
+    "recommendation": "Refactor the `chunk_code` function to extract nested functions into separate helper functions.",
+    "suggested_code": "",
     "fixed_code": "",
-    "file_name": "kaizen/retriever/code_chunker.py",
+    "file_path": "kaizen/retriever/code_chunker.py",
     "start_line": 7,
     "end_line": 62,
     "side": "RIGHT",
     "sentiment": "neutral",
-    "severity_level": 6
+    "severity": 6
   },
   {
-    "topic": "Type Annotations",
-    "comment": "Missing or incomplete type annotations for method parameters and return types.",
-    "confidence": "important",
-    "reason": "Type annotations improve code readability and help with static analysis. This was mentioned by several models.",
-    "solution": "Add or improve type annotations to method parameters and return types.",
-    "actual_code": "def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:",
+    "category": "Type Annotations",
+    "description": "Missing or incomplete type annotations for method parameters and return types.",
+    "impact": "high",
+    "rationale": "Type annotations improve code readability and help with static analysis. This was mentioned by several models.",
+    "recommendation": "Add or improve type annotations to method parameters and return types.",
+    "suggested_code": "def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:",
     "fixed_code": "def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[Dict[str, Any]]:",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
+    "file_path": "kaizen/retriever/custom_vector_store.py",
     "start_line": 13,
     "end_line": 13,
     "side": "RIGHT",
     "sentiment": "neutral",
-    "severity_level": 5
+    "severity": 5
   },
   {
-    "topic": "Documentation",
-    "comment": "Lack of comments and docstrings in complex functions and methods.",
-    "confidence": "important",
-    "reason": "Several models noted the need for better documentation to improve code understanding and maintainability.",
-    "solution": "Add comments explaining complex logic and docstrings to functions and methods.",
-    "actual_code": "",
+    "category": "Documentation",
+    "description": "Lack of comments and docstrings in complex functions and methods.",
+    "impact": "high",
+    "rationale": "Several models noted the need for better documentation to improve code understanding and maintainability.",
+    "recommendation": "Add comments explaining complex logic and docstrings to functions and methods.",
+    "suggested_code": "",
     "fixed_code": "",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
     "start_line": 1,
     "end_line": 1,
     "side": "RIGHT",
     "sentiment": "neutral",
-    "severity_level": 6
+    "severity": 6
   },
   {
-    "topic": "Code Duplication",
-    "comment": "Duplicate code found in test cases and database connection string creation.",
-    "confidence": "important",
-    "reason": "Code duplication was identified by multiple models as an issue that can lead to maintenance problems.",
-    "solution": "Refactor duplicate code into reusable functions or constants.",
-    "actual_code": "",
+    "category": "Code Duplication",
+    "description": "Duplicate code found in test cases and database connection string creation.",
+    "impact": "high",
+    "rationale": "Code duplication was identified by multiple models as an issue that can lead to maintenance problems.",
+    "recommendation": "Refactor duplicate code into reusable functions or constants.",
+    "suggested_code": "",
     "fixed_code": "",
-    "file_name": "tests/retriever/test_chunker.py",
+    "file_path": "tests/retriever/test_chunker.py",
     "start_line": 98,
     "end_line": 101,
     "side": "RIGHT",
     "sentiment": "neutral",
-    "severity_level": 6
+    "severity": 6
   },
   {
-    "topic": "Dependency Management",
-    "comment": "New dependencies added without justification.",
-    "confidence": "important",
-    "reason": "Adding dependencies increases the attack surface and maintenance burden. This was noted as an important consideration by multiple models.",
-    "solution": "Review and justify new dependencies, ensuring they are necessary and compatible.",
-    "actual_code": "",
+    "category": "Dependency Management",
+    "description": "New dependencies added without justification.",
+    "impact": "high",
+    "rationale": "Adding dependencies increases the attack surface and maintenance burden. This was noted as an important consideration by multiple models.",
+    "recommendation": "Review and justify new dependencies, ensuring they are necessary and compatible.",
+    "suggested_code": "",
     "fixed_code": "",
-    "file_name": "pyproject.toml",
+    "file_path": "pyproject.toml",
     "start_line": 27,
     "end_line": 49,
     "side": "RIGHT",
     "sentiment": "neutral",
-    "severity_level": 6
+    "severity": 6
   },
   {
-    "topic": "Performance",
-    "comment": "Potential performance issues in database operations and code parsing.",
-    "confidence": "moderate",
-    "reason": "Several models identified areas where performance could be improved, particularly in database operations and file parsing.",
-    "solution": "Optimize database queries, consider batching operations, and review file parsing logic for potential improvements.",
-    "actual_code": "",
+    "category": "Performance",
+    "description": "Potential performance issues in database operations and code parsing.",
+    "impact": "medium",
+    "rationale": "Several models identified areas where performance could be improved, particularly in database operations and file parsing.",
+    "recommendation": "Optimize database queries, consider batching operations, and review file parsing logic for potential improvements.",
+    "suggested_code": "",
     "fixed_code": "",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
     "start_line": 1,
     "end_line": 1,
     "side": "RIGHT",
     "sentiment": "neutral",
-    "severity_level": 5
+    "severity": 5
   },
   {
-    "topic": "Error Handling",
-    "comment": "Improve error handling in the parse_file method and LanguageLoader class.",
-    "confidence": "important",
-    "reason": "Better error handling was suggested by multiple models to improve debugging and prevent unexpected behavior.",
-    "solution": "Implement more specific exception handling and provide detailed error messages.",
-    "actual_code": "except Exception as e:\n    logger.error(f\"Error processing file{file_path}:{str(e)}\")\n    logger.error(traceback.format_exc())",
+    "category": "Error Handling",
+    "description": "Improve error handling in the parse_file method and LanguageLoader class.",
+    "impact": "high",
+    "rationale": "Better error handling was suggested by multiple models to improve debugging and prevent unexpected behavior.",
+    "recommendation": "Implement more specific exception handling and provide detailed error messages.",
+    "suggested_code": "except Exception as e:\n    logger.error(f\"Error processing file{file_path}:{str(e)}\")\n    logger.error(traceback.format_exc())",
     "fixed_code": "except Exception as e:\n    logger.error(f\"Error processing file{file_path}:{str(e)}\")\n    logger.error(traceback.format_exc())\n    raise",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
+    "file_path": "kaizen/retriever/llama_index_retriever.py",
     "start_line": 108,
     "end_line": 110,
     "side": "RIGHT",
     "sentiment": "negative",
-    "severity_level": 7
+    "severity": 7
   }
 ]
\ No newline at end of file
diff --git a/.experiments/code_review/dataset/pr_232/issues.json b/.experiments/code_review/dataset/pr_232/issues.json
index 8ddf3398..d1b2af64 100644
--- a/.experiments/code_review/dataset/pr_232/issues.json
+++ b/.experiments/code_review/dataset/pr_232/issues.json
@@ -1,107 +1,107 @@
 [
   {
-    "topic": "Unused Imports",
-    "comment": "There are several unused imports across multiple files that should be removed.",
-    "confidence": "important",
-    "reason": "Removing unused imports improves code cleanliness, readability, and potentially reduces bundle size. This issue was identified by both models across multiple files.",
-    "solution": "Remove all unused imports from the affected files.",
-    "actual_code": "",
+    "category": "Unused Imports",
+    "description": "There are several unused imports across multiple files that should be removed.",
+    "impact": "high",
+    "rationale": "Removing unused imports improves code cleanliness, readability, and potentially reduces bundle size. This issue was identified by both models across multiple files.",
+    "recommendation": "Remove all unused imports from the affected files.",
+    "suggested_code": "",
     "fixed_code": "",
-    "file_name": "page.tsx, queryinput.tsx, apps/web/app/(dash)/home/page.tsx, apps/web/app/(dash)/home/queryinput.tsx, packages/ui/shadcn/combobox.tsx",
+    "file_path": "page.tsx, queryinput.tsx, apps/web/app/(dash)/home/page.tsx, apps/web/app/(dash)/home/queryinput.tsx, packages/ui/shadcn/combobox.tsx",
     "start_line": 0,
     "end_line": 0,
     "side": "RIGHT",
     "sentiment": "neutral",
-    "severity_level": 5
+    "severity": 5
   },
   {
-    "topic": "Type Annotations and Definitions",
-    "comment": "Some variables, functions, and components are missing proper type annotations or definitions.",
-    "confidence": "important",
-    "reason": "Proper type annotations improve code readability, maintainability, and help catch type-related errors at compile-time. This issue was noted by both models.",
-    "solution": "Add or improve type annotations for variables, functions, and components where they are missing or inadequate.",
-    "actual_code": "const ComboboxWithCreate = ({",
+    "category": "Type Annotations and Definitions",
+    "description": "Some variables, functions, and components are missing proper type annotations or definitions.",
+    "impact": "high",
+    "rationale": "Proper type annotations improve code readability, maintainability, and help catch type-related errors at compile-time. This issue was noted by both models.",
+    "recommendation": "Add or improve type annotations for variables, functions, and components where they are missing or inadequate.",
+    "suggested_code": "const ComboboxWithCreate = ({",
     "fixed_code": "const ComboboxWithCreate: React.FC<ComboboxWithCreateProps> = ({",
-    "file_name": "queryinput.tsx, packages/ui/shadcn/combobox.tsx, apps/web/app/(dash)/(memories)/content.tsx",
+    "file_path": "queryinput.tsx, packages/ui/shadcn/combobox.tsx, apps/web/app/(dash)/(memories)/content.tsx",
     "start_line": 32,
     "end_line": 32,
     "side": "RIGHT",
     "sentiment": "negative",
-    "severity_level": 6
+    "severity": 6
   },
   {
-    "topic": "Code Organization and Structure",
-    "comment": "Some files contain multiple unrelated components or have poor code organization.",
-    "confidence": "important",
-    "reason": "Proper code organization improves readability, maintainability, and reusability. This issue was identified by both models.",
-    "solution": "Separate unrelated components into their own files and improve overall code structure.",
-    "actual_code": "",
+    "category": "Code Organization and Structure",
+    "description": "Some files contain multiple unrelated components or have poor code organization.",
+    "impact": "high",
+    "rationale": "Proper code organization improves readability, maintainability, and reusability. This issue was identified by both models.",
+    "recommendation": "Separate unrelated components into their own files and improve overall code structure.",
+    "suggested_code": "",
     "fixed_code": "",
-    "file_name": "page.tsx, apps/web/app/(dash)/menu.tsx",
+    "file_path": "page.tsx, apps/web/app/(dash)/menu.tsx",
     "start_line": 0,
     "end_line": 0,
     "side": "RIGHT",
     "sentiment": "neutral",
-    "severity_level": 6
+    "severity": 6
   },
   {
-    "topic": "Error Handling",
-    "comment": "Improve error handling in various parts of the code, particularly in the handleSubmit function.",
-    "confidence": "important",
-    "reason": "Proper error handling is crucial for preventing crashes and providing useful feedback. This issue was highlighted by both models.",
-    "solution": "Implement robust error handling, especially in critical functions like handleSubmit.",
-    "actual_code": "throw new Error(`Memory creation failed: ${cont.error}`);\nreturn cont;",
+    "category": "Error Handling",
+    "description": "Improve error handling in various parts of the code, particularly in the handleSubmit function.",
+    "impact": "high",
+    "rationale": "Proper error handling is crucial for preventing crashes and providing useful feedback. This issue was highlighted by both models.",
+    "recommendation": "Implement robust error handling, especially in critical functions like handleSubmit.",
+    "suggested_code": "throw new Error(`Memory creation failed: ${cont.error}`);\nreturn cont;",
     "fixed_code": "throw new Error(`Memory creation failed: ${cont.error}`);",
-    "file_name": "apps/web/app/(dash)/menu.tsx",
+    "file_path": "apps/web/app/(dash)/menu.tsx",
     "start_line": 230,
     "end_line": 231,
     "side": "RIGHT",
     "sentiment": "negative",
-    "severity_level": 7
+    "severity": 7
   },
   {
-    "topic": "State Management",
-    "comment": "Consider improving state management to avoid prop drilling and improve component encapsulation.",
-    "confidence": "moderate",
-    "reason": "Better state management can improve code maintainability and reduce complexity. This was suggested by the Sonnet model.",
-    "solution": "Consider using React Context or a state management library for managing global state.",
-    "actual_code": "",
+    "category": "State Management",
+    "description": "Consider improving state management to avoid prop drilling and improve component encapsulation.",
+    "impact": "medium",
+    "rationale": "Better state management can improve code maintainability and reduce complexity. This was suggested by the Sonnet model.",
+    "recommendation": "Consider using React Context or a state management library for managing global state.",
+    "suggested_code": "",
     "fixed_code": "",
-    "file_name": "apps/web/app/(dash)/menu.tsx",
+    "file_path": "apps/web/app/(dash)/menu.tsx",
     "start_line": 163,
     "end_line": 167,
     "side": "RIGHT",
     "sentiment": "neutral",
-    "severity_level": 5
+    "severity": 5
   },
   {
-    "topic": "Performance Optimization",
-    "comment": "Some computations, like filtering options, could be optimized to improve performance.",
-    "confidence": "moderate",
-    "reason": "Optimizing expensive computations can lead to better performance, especially for larger datasets.",
-    "solution": "Use memoization techniques like useMemo for expensive computations that don't need to be recalculated on every render.",
-    "actual_code": "const filteredOptions = options.filter(\n\t\t(option) => !selectedSpaces.includes(parseInt(option.value)),\n\t);",
+    "category": "Performance Optimization",
+    "description": "Some computations, like filtering options, could be optimized to improve performance.",
+    "impact": "medium",
+    "rationale": "Optimizing expensive computations can lead to better performance, especially for larger datasets.",
+    "recommendation": "Use memoization techniques like useMemo for expensive computations that don't need to be recalculated on every render.",
+    "suggested_code": "const filteredOptions = options.filter(\n\t\t(option) => !selectedSpaces.includes(parseInt(option.value)),\n\t);",
     "fixed_code": "const filteredOptions = useMemo(() => options.filter(\n\t\t(option) => !selectedSpaces.includes(parseInt(option.value)),\n\t),[options, selectedSpaces]);",
-    "file_name": "packages/ui/shadcn/combobox.tsx",
+    "file_path": "packages/ui/shadcn/combobox.tsx",
     "start_line": 55,
     "end_line": 57,
     "side": "RIGHT",
     "sentiment": "neutral",
-    "severity_level": 4
+    "severity": 4
   },
   {
-    "topic": "Accessibility",
-    "comment": "Some UI elements lack proper accessibility attributes.",
-    "confidence": "moderate",
-    "reason": "Improving accessibility ensures the application is usable by all users, including those with disabilities.",
-    "solution": "Add appropriate aria-labels and other accessibility attributes to interactive elements.",
-    "actual_code": "",
+    "category": "Accessibility",
+    "description": "Some UI elements lack proper accessibility attributes.",
+    "impact": "medium",
+    "rationale": "Improving accessibility ensures the application is usable by all users, including those with disabilities.",
+    "recommendation": "Add appropriate aria-labels and other accessibility attributes to interactive elements.",
+    "suggested_code": "",
     "fixed_code": "",
-    "file_name": "packages/ui/shadcn/combobox.tsx",
+    "file_path": "packages/ui/shadcn/combobox.tsx",
     "start_line": 65,
     "end_line": 72,
     "side": "RIGHT",
     "sentiment": "neutral",
-    "severity_level": 4
+    "severity": 4
   }
 ]
\ No newline at end of file
diff --git a/.experiments/code_review/dataset/pr_252/issues.json b/.experiments/code_review/dataset/pr_252/issues.json
index 568ce90d..1f0c3e8e 100644
--- a/.experiments/code_review/dataset/pr_252/issues.json
+++ b/.experiments/code_review/dataset/pr_252/issues.json
@@ -1,92 +1,92 @@
 [
   {
-    "topic": "Code Structure and Consistency",
-    "comment": "There are inconsistencies in code formatting and structure across different function calls.",
-    "confidence": "moderate",
-    "reason": "Consistent code structure and formatting improves readability and maintainability. This issue was noted by multiple models.",
-    "solution": "Standardize the formatting of function calls, particularly for `generate_twitter_post` and `generate_linkedin_post`. Consider using multi-line formatting for both for consistency.",
-    "actual_code": "twitter_post = work_summary_generator.generate_twitter_post(summary, user=\"oss_example\")\n\nlinkedin_post = work_summary_generator.generate_linkedin_post(\n    summary, user=\"oss_example\"\n)",
+    "category": "Code Structure and Consistency",
+    "description": "There are inconsistencies in code formatting and structure across different function calls.",
+    "impact": "medium",
+    "rationale": "Consistent code structure and formatting improves readability and maintainability. This issue was noted by multiple models.",
+    "recommendation": "Standardize the formatting of function calls, particularly for `generate_twitter_post` and `generate_linkedin_post`. Consider using multi-line formatting for both for consistency.",
+    "suggested_code": "twitter_post = work_summary_generator.generate_twitter_post(summary, user=\"oss_example\")\n\nlinkedin_post = work_summary_generator.generate_linkedin_post(\n    summary, user=\"oss_example\"\n)",
     "fixed_code": "twitter_post = work_summary_generator.generate_twitter_post(\n    summary, user=\"oss_example\"\n)\n\nlinkedin_post = work_summary_generator.generate_linkedin_post(\n    summary, user=\"oss_example\"\n)",
-    "file_name": "examples/work_summarizer/main.py",
+    "file_path": "examples/work_summarizer/main.py",
     "start_line": 59,
     "end_line": 62,
     "side": "RIGHT",
     "sentiment": "neutral",
-    "severity_level": 4
+    "severity": 4
   },
   {
-    "topic": "Code Organization",
-    "comment": "The `WorkSummaryGenerator` class has multiple responsibilities and could be refactored for better organization.",
-    "confidence": "important",
-    "reason": "Separation of Concerns (SoC) principle improves code maintainability and readability.",
-    "solution": "Refactor the `WorkSummaryGenerator` class into separate classes or functions for each responsibility (e.g., summary generation, Twitter post generation, LinkedIn post generation).",
-    "actual_code": "",
+    "category": "Code Organization",
+    "description": "The `WorkSummaryGenerator` class has multiple responsibilities and could be refactored for better organization.",
+    "impact": "high",
+    "rationale": "Separation of Concerns (SoC) principle improves code maintainability and readability.",
+    "recommendation": "Refactor the `WorkSummaryGenerator` class into separate classes or functions for each responsibility (e.g., summary generation, Twitter post generation, LinkedIn post generation).",
+    "suggested_code": "",
     "fixed_code": "",
-    "file_name": "kaizen/reviewer/work_summarizer.py",
+    "file_path": "kaizen/reviewer/work_summarizer.py",
     "start_line": 0,
     "end_line": 0,
     "side": "RIGHT",
     "sentiment": "neutral",
-    "severity_level": 6
+    "severity": 6
   },
   {
-    "topic": "Error Handling",
-    "comment": "The `generate_twitter_post` and `generate_linkedin_post` methods lack error handling.",
-    "confidence": "important",
-    "reason": "Proper error handling improves code robustness and helps with debugging.",
-    "solution": "Add try-except blocks to handle and log any exceptions during the post generation process.",
-    "actual_code": "",
+    "category": "Error Handling",
+    "description": "The `generate_twitter_post` and `generate_linkedin_post` methods lack error handling.",
+    "impact": "high",
+    "rationale": "Proper error handling improves code robustness and helps with debugging.",
+    "recommendation": "Add try-except blocks to handle and log any exceptions during the post generation process.",
+    "suggested_code": "",
     "fixed_code": "",
-    "file_name": "kaizen/reviewer/work_summarizer.py",
+    "file_path": "kaizen/reviewer/work_summarizer.py",
     "start_line": 58,
     "end_line": 74,
     "side": "RIGHT",
     "sentiment": "neutral",
-    "severity_level": 7
+    "severity": 7
   },
   {
-    "topic": "Code Duplication",
-    "comment": "There is code duplication in the `generate_twitter_post` and `generate_linkedin_post` methods, and a duplicated print statement for LinkedIn post.",
-    "confidence": "important",
-    "reason": "Code duplication violates the DRY principle and can lead to maintenance issues.",
-    "solution": "Extract common code from `generate_twitter_post` and `generate_linkedin_post` into a shared method. Remove the duplicated print statement for the LinkedIn post.",
-    "actual_code": "print(f\" LinkedIn Post: \\n{linkedin_post}\\n\")",
+    "category": "Code Duplication",
+    "description": "There is code duplication in the `generate_twitter_post` and `generate_linkedin_post` methods, and a duplicated print statement for LinkedIn post.",
+    "impact": "high",
+    "rationale": "Code duplication violates the DRY principle and can lead to maintenance issues.",
+    "recommendation": "Extract common code from `generate_twitter_post` and `generate_linkedin_post` into a shared method. Remove the duplicated print statement for the LinkedIn post.",
+    "suggested_code": "print(f\" LinkedIn Post: \\n{linkedin_post}\\n\")",
     "fixed_code": "",
-    "file_name": "kaizen/reviewer/work_summarizer.py, examples/work_summarizer/main.py",
+    "file_path": "kaizen/reviewer/work_summarizer.py, examples/work_summarizer/main.py",
     "start_line": 58,
     "end_line": 74,
     "side": "RIGHT",
     "sentiment": "negative",
-    "severity_level": 6
+    "severity": 6
   },
   {
-    "topic": "Code Documentation",
-    "comment": "The `severity_level` field in the code review prompt lacks detailed explanation.",
-    "confidence": "moderate",
-    "reason": "Clear documentation helps users understand how to use features correctly.",
-    "solution": "Add a more detailed explanation of what each severity level represents in the code review prompt.",
-    "actual_code": "For \"severity_level\" score in range of 1 to 10, 1 being not severe and 10 being critical.",
+    "category": "Code Documentation",
+    "description": "The `severity_level` field in the code review prompt lacks detailed explanation.",
+    "impact": "medium",
+    "rationale": "Clear documentation helps users understand how to use features correctly.",
+    "recommendation": "Add a more detailed explanation of what each severity level represents in the code review prompt.",
+    "suggested_code": "For \"severity_level\" score in range of 1 to 10, 1 being not severe and 10 being critical.",
     "fixed_code": "For \"severity_level\" score in range of 1 to 10:\n1-3: Minor issues (style, small optimizations)\n4-6: Moderate issues (potential bugs, performance concerns)\n7-8: Major issues (definite bugs, security vulnerabilities)\n9-10: Critical issues (severe security risks, system-breaking bugs)",
-    "file_name": "kaizen/llms/prompts/code_review_prompts.py",
+    "file_path": "kaizen/llms/prompts/code_review_prompts.py",
     "start_line": 100,
     "end_line": 100,
     "side": "RIGHT",
     "sentiment": "neutral",
-    "severity_level": 4
+    "severity": 4
   },
   {
-    "topic": "Prompt Formatting",
-    "comment": "The TWITTER_POST_PROMPT and LINKEDIN_POST_PROMPT could be improved for better readability.",
-    "confidence": "moderate",
-    "reason": "Well-formatted prompts are easier to read and maintain.",
-    "solution": "Break the prompts into multiple lines, use string formatting, and add comments to explain different sections.",
-    "actual_code": "",
+    "category": "Prompt Formatting",
+    "description": "The TWITTER_POST_PROMPT and LINKEDIN_POST_PROMPT could be improved for better readability.",
+    "impact": "medium",
+    "rationale": "Well-formatted prompts are easier to read and maintain.",
+    "recommendation": "Break the prompts into multiple lines, use string formatting, and add comments to explain different sections.",
+    "suggested_code": "",
     "fixed_code": "",
-    "file_name": "kaizen/llms/prompts/work_summary_prompts.py",
+    "file_path": "kaizen/llms/prompts/work_summary_prompts.py",
     "start_line": 44,
     "end_line": 65,
     "side": "RIGHT",
     "sentiment": "positive",
-    "severity_level": 4
+    "severity": 4
   }
 ]
\ No newline at end of file
diff --git a/.experiments/code_review/dataset/pr_335/issues.json b/.experiments/code_review/dataset/pr_335/issues.json
index cb9843ab..5ec8ba95 100644
--- a/.experiments/code_review/dataset/pr_335/issues.json
+++ b/.experiments/code_review/dataset/pr_335/issues.json
@@ -1,92 +1,92 @@
 [
   {
-    "topic": "Import Changes",
-    "comment": "Import statements have been changed and some may be unused.",
-    "confidence": "important",
-    "reason": "Changing import paths can lead to runtime errors and unused imports clutter the code. This issue was identified by multiple models.",
-    "solution": "Verify that all new import paths are correct, remove any unused imports, and ensure consistency across the codebase.",
-    "actual_code": "from kaizen.llms.prompts.pr_desc_prompts import (",
+    "category": "Import Changes",
+    "description": "Import statements have been changed and some may be unused.",
+    "impact": "high",
+    "rationale": "Changing import paths can lead to runtime errors and unused imports clutter the code. This issue was identified by multiple models.",
+    "recommendation": "Verify that all new import paths are correct, remove any unused imports, and ensure consistency across the codebase.",
+    "suggested_code": "from kaizen.llms.prompts.pr_desc_prompts import (",
     "fixed_code": "",
-    "file_name": "kaizen/generator/pr_description.py",
+    "file_path": "kaizen/generator/pr_description.py",
     "start_line": 8,
     "end_line": 8,
     "side": "RIGHT",
     "sentiment": "neutral",
-    "severity_level": 6
+    "severity": 6
   },
   {
-    "topic": "Removal of Reevaluation Logic",
-    "comment": "The 'reeval_response' parameter and associated logic have been removed from multiple functions.",
-    "confidence": "critical",
-    "reason": "Removing this parameter and logic could significantly change the behavior of the PR description generation. This was noted as a critical issue by multiple models.",
-    "solution": "Carefully review the impact of removing the reevaluation logic. Ensure that the quality of PR descriptions is maintained without this feature. Consider adding unit tests to verify the new behavior.",
-    "actual_code": "",
+    "category": "Removal of Reevaluation Logic",
+    "description": "The 'reeval_response' parameter and associated logic have been removed from multiple functions.",
+    "impact": "critical",
+    "rationale": "Removing this parameter and logic could significantly change the behavior of the PR description generation. This was noted as a critical issue by multiple models.",
+    "recommendation": "Carefully review the impact of removing the reevaluation logic. Ensure that the quality of PR descriptions is maintained without this feature. Consider adding unit tests to verify the new behavior.",
+    "suggested_code": "",
     "fixed_code": "",
-    "file_name": "kaizen/generator/pr_description.py",
+    "file_path": "kaizen/generator/pr_description.py",
     "start_line": 43,
     "end_line": 96,
     "side": "LEFT",
     "sentiment": "negative",
-    "severity_level": 8
+    "severity": 8
   },
   {
-    "topic": "API Change",
-    "comment": "Changed from 'chat_completion_with_json' to 'chat_completion'",
-    "confidence": "important",
-    "reason": "This API change could affect the format of the response and how it's processed. Multiple models highlighted this as an important change.",
-    "solution": "Ensure that the new chat_completion method returns the expected format. Update any dependent code that might be affected by this change. Verify that the response parsing is adjusted accordingly.",
-    "actual_code": "resp, usage = self.provider.chat_completion(prompt, user=user)",
+    "category": "API Change",
+    "description": "Changed from 'chat_completion_with_json' to 'chat_completion'",
+    "impact": "high",
+    "rationale": "This API change could affect the format of the response and how it's processed. Multiple models highlighted this as an important change.",
+    "recommendation": "Ensure that the new chat_completion method returns the expected format. Update any dependent code that might be affected by this change. Verify that the response parsing is adjusted accordingly.",
+    "suggested_code": "resp, usage = self.provider.chat_completion(prompt, user=user)",
     "fixed_code": "resp, usage = self.provider.chat_completion(prompt, user=user)\ndesc = parser.extract_code_from_markdown(resp)",
-    "file_name": "kaizen/generator/pr_description.py",
+    "file_path": "kaizen/generator/pr_description.py",
     "start_line": 79,
     "end_line": 80,
     "side": "RIGHT",
     "sentiment": "neutral",
-    "severity_level": 7
+    "severity": 7
   },
   {
-    "topic": "Prompt Changes",
-    "comment": "Significant changes to PR description prompts and system prompts.",
-    "confidence": "important",
-    "reason": "The prompts have been restructured and moved to a new file. This could impact the quality and structure of generated PR descriptions.",
-    "solution": "Review the new prompt structure to ensure it meets all requirements. Test thoroughly to verify that the generated PR descriptions maintain or improve quality. Update any related documentation.",
-    "actual_code": "",
+    "category": "Prompt Changes",
+    "description": "Significant changes to PR description prompts and system prompts.",
+    "impact": "high",
+    "rationale": "The prompts have been restructured and moved to a new file. This could impact the quality and structure of generated PR descriptions.",
+    "recommendation": "Review the new prompt structure to ensure it meets all requirements. Test thoroughly to verify that the generated PR descriptions maintain or improve quality. Update any related documentation.",
+    "suggested_code": "",
     "fixed_code": "",
-    "file_name": "kaizen/llms/prompts/pr_desc_prompts.py",
+    "file_path": "kaizen/llms/prompts/pr_desc_prompts.py",
     "start_line": 1,
     "end_line": 92,
     "side": "RIGHT",
     "sentiment": "positive",
-    "severity_level": 7
+    "severity": 7
   },
   {
-    "topic": "Error Handling",
-    "comment": "Potential lack of error handling for exceptions in PR description generation.",
-    "confidence": "important",
-    "reason": "Proper error handling is crucial for preventing unexpected crashes and providing useful feedback.",
-    "solution": "Implement try-except blocks where appropriate to handle potential exceptions gracefully. Consider using more specific exception types.",
-    "actual_code": "raise Exception(\"Both diff_text and pull_request_files are empty!\")",
+    "category": "Error Handling",
+    "description": "Potential lack of error handling for exceptions in PR description generation.",
+    "impact": "high",
+    "rationale": "Proper error handling is crucial for preventing unexpected crashes and providing useful feedback.",
+    "recommendation": "Implement try-except blocks where appropriate to handle potential exceptions gracefully. Consider using more specific exception types.",
+    "suggested_code": "raise Exception(\"Both diff_text and pull_request_files are empty!\")",
     "fixed_code": "raise ValueError(\"Both diff_text and pull_request_files are empty!\")",
-    "file_name": "kaizen/generator/pr_description.py",
+    "file_path": "kaizen/generator/pr_description.py",
     "start_line": 51,
     "end_line": 51,
     "side": "LEFT",
     "sentiment": "negative",
-    "severity_level": 7
+    "severity": 7
   },
   {
-    "topic": "Code Style and Documentation",
-    "comment": "Various minor issues with code style, variable naming, and documentation.",
-    "confidence": "moderate",
-    "reason": "Consistent code style and proper documentation improve readability and maintainability.",
-    "solution": "Review and update variable names to follow PEP 8 conventions. Add docstrings or comments explaining the purpose of new prompts and significant changes.",
-    "actual_code": "",
+    "category": "Code Style and Documentation",
+    "description": "Various minor issues with code style, variable naming, and documentation.",
+    "impact": "medium",
+    "rationale": "Consistent code style and proper documentation improve readability and maintainability.",
+    "recommendation": "Review and update variable names to follow PEP 8 conventions. Add docstrings or comments explaining the purpose of new prompts and significant changes.",
+    "suggested_code": "",
     "fixed_code": "",
-    "file_name": "kaizen/llms/prompts/pr_desc_prompts.py, kaizen/generator/pr_description.py",
+    "file_path": "kaizen/llms/prompts/pr_desc_prompts.py, kaizen/generator/pr_description.py",
     "start_line": 1,
     "end_line": 1,
     "side": "RIGHT",
     "sentiment": "neutral",
-    "severity_level": 4
+    "severity": 4
   }
 ]
\ No newline at end of file
diff --git a/.experiments/code_review/dataset/pr_400/issues.json b/.experiments/code_review/dataset/pr_400/issues.json
index 1bc6c452..037ac7a3 100644
--- a/.experiments/code_review/dataset/pr_400/issues.json
+++ b/.experiments/code_review/dataset/pr_400/issues.json
@@ -1,52 +1,52 @@
 [
   {
-    "topic": "Test Coverage",
-    "comment": "Improved test coverage with more comprehensive test cases, including edge cases and error handling scenarios.",
-    "confidence": "important",
-    "reason": "Comprehensive test coverage is crucial for maintaining code quality, catching potential bugs, and ensuring the reliability of the code.",
-    "solution": "Continue to add test cases for edge cases and ensure all new functionality is covered by tests.",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py, .kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py, .kaizen/unit_test/kaizen/helpers/test_create_test_files.py, .kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
+    "category": "Test Coverage",
+    "description": "Improved test coverage with more comprehensive test cases, including edge cases and error handling scenarios.",
+    "impact": "high",
+    "rationale": "Comprehensive test coverage is crucial for maintaining code quality, catching potential bugs, and ensuring the reliability of the code.",
+    "recommendation": "Continue to add test cases for edge cases and ensure all new functionality is covered by tests.",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py, .kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py, .kaizen/unit_test/kaizen/helpers/test_create_test_files.py, .kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
     "sentiment": "positive",
-    "severity_level": 7
+    "severity": 7
   },
   {
-    "topic": "Code Structure and Organization",
-    "comment": "Improved code structure with better modularity, organization, and use of pytest fixtures.",
-    "confidence": "important",
-    "reason": "Good organization improves readability, maintainability, and allows for better separation of concerns.",
-    "solution": "Continue to monitor for potential further improvements in code structure and organization.",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py, kaizen/generator/unit_test.py",
+    "category": "Code Structure and Organization",
+    "description": "Improved code structure with better modularity, organization, and use of pytest fixtures.",
+    "impact": "high",
+    "rationale": "Good organization improves readability, maintainability, and allows for better separation of concerns.",
+    "recommendation": "Continue to monitor for potential further improvements in code structure and organization.",
+    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py, kaizen/generator/unit_test.py",
     "sentiment": "positive",
-    "severity_level": 3
+    "severity": 3
   },
   {
-    "topic": "Error Handling",
-    "comment": "Insufficient error handling in some methods, particularly for file operations.",
-    "confidence": "important",
-    "reason": "Proper error handling ensures the program remains stable and provides useful error messages.",
-    "solution": "Implement try-except blocks to handle potential errors, especially in file operations and network requests.",
-    "file_name": "kaizen/generator/unit_test.py",
+    "category": "Error Handling",
+    "description": "Insufficient error handling in some methods, particularly for file operations.",
+    "impact": "high",
+    "rationale": "Proper error handling ensures the program remains stable and provides useful error messages.",
+    "recommendation": "Implement try-except blocks to handle potential errors, especially in file operations and network requests.",
+    "file_path": "kaizen/generator/unit_test.py",
     "sentiment": "negative",
-    "severity_level": 7
+    "severity": 7
   },
   {
-    "topic": "Logging Configuration",
-    "comment": "The logging configuration might override existing setups and is mixed with import statements.",
-    "confidence": "important",
-    "reason": "Inconsistent logging configuration can lead to loss of important log information and poor code organization.",
-    "solution": "Adjust the logging configuration to respect the LOGLEVEL environment variable and move it to a separate section after all imports.",
-    "file_name": "kaizen/llms/provider.py",
+    "category": "Logging Configuration",
+    "description": "The logging configuration might override existing setups and is mixed with import statements.",
+    "impact": "high",
+    "rationale": "Inconsistent logging configuration can lead to loss of important log information and poor code organization.",
+    "recommendation": "Adjust the logging configuration to respect the LOGLEVEL environment variable and move it to a separate section after all imports.",
+    "file_path": "kaizen/llms/provider.py",
     "sentiment": "negative",
-    "severity_level": 7
+    "severity": 7
   },
   {
-    "topic": "Configuration Update",
-    "comment": "Changes made to sensitive configuration file",
-    "confidence": "critical",
-    "reason": "Changes to config.json may affect system functionality and security",
-    "solution": "Review the changes in config.json carefully, especially the new 'base_model' fields, and ensure that the code using this configuration is updated accordingly.",
-    "file_name": "config.json",
+    "category": "Configuration Update",
+    "description": "Changes made to sensitive configuration file",
+    "impact": "critical",
+    "rationale": "Changes to config.json may affect system functionality and security",
+    "recommendation": "Review the changes in config.json carefully, especially the new 'base_model' fields, and ensure that the code using this configuration is updated accordingly.",
+    "file_path": "config.json",
     "sentiment": "negative",
-    "severity_level": 10
+    "severity": 10
   }
 ]
\ No newline at end of file
diff --git a/.experiments/code_review/dataset/pr_476/issues.json b/.experiments/code_review/dataset/pr_476/issues.json
index c426f116..035e3712 100644
--- a/.experiments/code_review/dataset/pr_476/issues.json
+++ b/.experiments/code_review/dataset/pr_476/issues.json
@@ -1,68 +1,68 @@
 [
   {
-    "topic": "Error Handling",
-    "comment": "Broad exception handling with generic error message",
-    "confidence": "critical",
-    "reason": "Using a generic 'except Exception' block with a non-specific error message can mask important errors and make debugging difficult.",
-    "solution": "Catch specific exceptions where possible and provide more informative error messages. Consider using proper logging instead of print statements.",
-    "actual_code": "except Exception:\n    print(\"Error\")",
+    "category": "Error Handling",
+    "description": "Broad exception handling with generic error message",
+    "impact": "high",
+    "rationale": "Using a generic 'except Exception' block with a non-specific error message can mask important errors and make debugging difficult.",
+    "recommendation": "Catch specific exceptions where possible and provide more informative error messages. Consider using proper logging instead of print statements.",
+    "suggested_code": "except Exception:\n    print(\"Error\")",
     "fixed_code": "except KeyError as e:\n    logger.error(f\"Invalid confidence level: {e}\")\nexcept Exception as e:\n    logger.error(f\"Unexpected error: {e}\")",
-    "file_name": "github_app/github_helper/pull_requests.py",
+    "file_path": "github_app/github_helper/pull_requests.py",
     "start_line": 140,
     "end_line": 141,
     "sentiment": "negative",
-    "severity_level": 9
+    "severity": 9
   },
   {
-    "topic": "Code Efficiency",
-    "comment": "Inefficient sorting implementation",
-    "confidence": "important",
-    "reason": "The custom sorting logic in 'sort_files' function is unnecessarily complex and inefficient for large lists.",
-    "solution": "Use Python's built-in sorted() function with a key function for better performance and readability.",
-    "actual_code": "def sort_files(files):\n    sorted_files = []\n    for file in files:\n        min_index = len(sorted_files)\n        file_name = file[\"filename\"]\n        for i, sorted_file in enumerate(sorted_files):\n            if file_name < sorted_file[\"filename\"]:\n                min_index = i\n                break\n        sorted_files.insert(min_index, file)\n    return sorted_files",
+    "category": "Code Efficiency",
+    "description": "Inefficient sorting implementation",
+    "impact": "high",
+    "rationale": "The custom sorting logic in 'sort_files' function is unnecessarily complex and inefficient for large lists.",
+    "recommendation": "Use Python's built-in sorted() function with a key function for better performance and readability.",
+    "suggested_code": "def sort_files(files):\n    sorted_files = []\n    for file in files:\n        min_index = len(sorted_files)\n        file_name = file[\"filename\"]\n        for i, sorted_file in enumerate(sorted_files):\n            if file_name < sorted_file[\"filename\"]:\n                min_index = i\n                break\n        sorted_files.insert(min_index, file)\n    return sorted_files",
     "fixed_code": "def sort_files(files):\n    return sorted(files, key=lambda x: x[\"filename\"])",
-    "file_name": "github_app/github_helper/pull_requests.py",
+    "file_path": "github_app/github_helper/pull_requests.py",
     "start_line": 184,
     "end_line": 194,
     "sentiment": "negative",
-    "severity_level": 6
+    "severity": 6
   },
   {
-    "topic": "Code Simplification",
-    "comment": "Overly verbose implementation of generate_tests function",
-    "confidence": "moderate",
-    "reason": "The current implementation of generate_tests function can be simplified using a list comprehension.",
-    "solution": "Use a list comprehension to create the list of filenames.",
-    "actual_code": "def generate_tests(pr_files):\n    return [f[\"filename\"] for f in pr_files]",
+    "category": "Code Simplification",
+    "description": "Overly verbose implementation of generate_tests function",
+    "impact": "medium",
+    "rationale": "The current implementation of generate_tests function can be simplified using a list comprehension.",
+    "recommendation": "Use a list comprehension to create the list of filenames.",
+    "suggested_code": "def generate_tests(pr_files):\n    return [f[\"filename\"] for f in pr_files]",
     "fixed_code": "def generate_tests(pr_files):\n    return [f[\"filename\"] for f in pr_files]",
-    "file_name": "github_app/github_helper/pull_requests.py",
+    "file_path": "github_app/github_helper/pull_requests.py",
     "start_line": 199,
     "end_line": 200,
     "sentiment": "positive",
-    "severity_level": 3
+    "severity": 3
   },
   {
-    "topic": "Logging and Debugging",
-    "comment": "Inconsistent use of print statements for debugging",
-    "confidence": "important",
-    "reason": "Using print statements for debugging can clutter the code and make it difficult to control log levels in different environments.",
-    "solution": "Replace print statements with proper logging calls using Python's logging module.",
-    "actual_code": "print(\"diff: \", diff_text)\nprint(\"pr_files\", pr_files)",
+    "category": "Logging and Debugging",
+    "description": "Inconsistent use of print statements for debugging",
+    "impact": "high",
+    "rationale": "Using print statements for debugging can clutter the code and make it difficult to control log levels in different environments.",
+    "recommendation": "Replace print statements with proper logging calls using Python's logging module.",
+    "suggested_code": "print(\"diff: \", diff_text)\nprint(\"pr_files\", pr_files)",
     "fixed_code": "import logging\n\nlogger = logging.getLogger(__name__)\nlogger.debug(f\"diff: {diff_text}\")\nlogger.debug(f\"pr_files: {pr_files}\")",
-    "file_name": "examples/code_review/main.py",
+    "file_path": "examples/code_review/main.py",
     "start_line": 21,
     "end_line": 22,
     "sentiment": "negative",
-    "severity_level": 6
+    "severity": 6
   },
   {
-    "topic": "Configuration Changes",
-    "comment": "Changes made to sensitive configuration file",
-    "confidence": "critical",
-    "reason": "Changes to config.json may affect system functionality and security. The removal of 'enable_observability_logging' option needs to be properly documented.",
-    "solution": "Review all changes in config.json carefully. If removing features, provide a migration guide or deprecation notice for existing users.",
-    "file_name": "config.json",
+    "category": "Configuration Changes",
+    "description": "Changes made to sensitive configuration file",
+    "impact": "critical",
+    "rationale": "Changes to config.json may affect system functionality and security. The removal of 'enable_observability_logging' option needs to be properly documented.",
+    "recommendation": "Review all changes in config.json carefully. If removing features, provide a migration guide or deprecation notice for existing users.",
+    "file_path": "config.json",
     "sentiment": "negative",
-    "severity_level": 10
+    "severity": 10
   }
 ]
\ No newline at end of file
diff --git a/.experiments/code_review/dataset/pr_5/issues.json b/.experiments/code_review/dataset/pr_5/issues.json
index 693b3000..2320f1f7 100644
--- a/.experiments/code_review/dataset/pr_5/issues.json
+++ b/.experiments/code_review/dataset/pr_5/issues.json
@@ -1,93 +1,93 @@
 [
   {
-    "topic": "Unused Import",
-    "comment": "The 'random' module is imported but never used in the code.",
-    "confidence": "trivial",
-    "reason": "The 'random' module is imported but not utilized in the code.",
-    "solution": "Remove the unused import statement for 'random'.",
-    "actual_code": "import random  # Unused import",
+    "category": "Unused Import",
+    "description": "The 'random' module is imported but never used in the code.",
+    "impact": "trivial",
+    "rationale": "The 'random' module is imported but not utilized in the code.",
+    "recommendation": "Remove the unused import statement for 'random'.",
+    "suggested_code": "import random  # Unused import",
     "fixed_code": "",
-    "file_name": "main.py",
+    "file_path": "main.py",
     "start_line": 8,
     "end_line": 8,
-    "severity_level": 1
+    "severity": 1
   },
   {
-    "topic": "API Call Error Handling",
-    "comment": "The API call to 'completion' lacks a retry mechanism.",
-    "confidence": "critical",
-    "reason": "API calls can fail due to network issues or server errors, and without a retry mechanism, the function may fail unexpectedly.",
-    "solution": "Implement a retry mechanism with exponential backoff for the API call.",
-    "actual_code": "response = completion(\n    model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n)",
+    "category": "API Call Error Handling",
+    "description": "The API call to 'completion' lacks a retry mechanism.",
+    "impact": "critical",
+    "rationale": "API calls can fail due to network issues or server errors, and without a retry mechanism, the function may fail unexpectedly.",
+    "recommendation": "Implement a retry mechanism with exponential backoff for the API call.",
+    "suggested_code": "response = completion(\n    model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n)",
     "fixed_code": "import time\n\nfor attempt in range(3):\n    try:\n        response = completion(\n            model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n        )\n        break\n    except Exception as e:\n        if attempt < 2:\n            time.sleep(2 ** attempt)\n        else:\n            raise e",
-    "file_name": "main.py",
+    "file_path": "main.py",
     "start_line": 66,
     "end_line": 68,
-    "severity_level": 9
+    "severity": 9
   },
   {
-    "topic": "Silent Failure in JSON Parsing",
-    "comment": "The exception handling for JSON decoding fails silently without logging.",
-    "confidence": "critical",
-    "reason": "Silent failures make it difficult to diagnose issues when they occur.",
-    "solution": "Add logging to capture the exception details.",
-    "actual_code": "except json.JSONDecodeError:\n    result = {",
+    "category": "Silent Failure in JSON Parsing",
+    "description": "The exception handling for JSON decoding fails silently without logging.",
+    "impact": "critical",
+    "rationale": "Silent failures make it difficult to diagnose issues when they occur.",
+    "recommendation": "Add logging to capture the exception details.",
+    "suggested_code": "except json.JSONDecodeError:\n    result = {",
     "fixed_code": "except json.JSONDecodeError as e:\n    print(f\"Failed to parse content for applicant: {e}\")\n    result = {",
-    "file_name": "main.py",
+    "file_path": "main.py",
     "start_line": 82,
     "end_line": 84,
-    "severity_level": 8
+    "severity": 8
   },
   {
-    "topic": "Inefficient Progress Printing",
-    "comment": "The progress printing method is inefficient.",
-    "confidence": "important",
-    "reason": "Printing progress in this manner can be slow and resource-intensive.",
-    "solution": "Use a more efficient method for printing progress, such as updating the progress less frequently or using a dedicated progress reporting library like tqdm.",
-    "actual_code": "print(f\"\\rProgress:[{'=' * int(50 * progress):<50}]{progress:.0%}\", end=\"\", flush=True)",
+    "category": "Inefficient Progress Printing",
+    "description": "The progress printing method is inefficient.",
+    "impact": "high",
+    "rationale": "Printing progress in this manner can be slow and resource-intensive.",
+    "recommendation": "Use a more efficient method for printing progress, such as updating the progress less frequently or using a dedicated progress reporting library like tqdm.",
+    "suggested_code": "print(f\"\\rProgress:[{'=' * int(50 * progress):<50}]{progress:.0%}\", end=\"\", flush=True)",
     "fixed_code": "if index % max(1, len(df) // 100) == 0:  # Update every 1%\n    print(f\"\\rProgress:[{'=' * int(50 * progress):<50}]{progress:.0%}\", end=\"\", flush=True)",
-    "file_name": "main.py",
+    "file_path": "main.py",
     "start_line": 121,
     "end_line": 122,
-    "severity_level": 5
+    "severity": 5
   },
   {
-    "topic": "Redundant Code",
-    "comment": "The check for an empty DataFrame is redundant.",
-    "confidence": "moderate",
-    "reason": "The code already handles an empty DataFrame gracefully, so this check is unnecessary.",
-    "solution": "Remove the redundant check for an empty DataFrame.",
-    "actual_code": "if len(df) == 0:\n    return",
+    "category": "Redundant Code",
+    "description": "The check for an empty DataFrame is redundant.",
+    "impact": "medium",
+    "rationale": "The code already handles an empty DataFrame gracefully, so this check is unnecessary.",
+    "recommendation": "Remove the redundant check for an empty DataFrame.",
+    "suggested_code": "if len(df) == 0:\n    return",
     "fixed_code": "",
-    "file_name": "main.py",
+    "file_path": "main.py",
     "start_line": 142,
     "end_line": 143,
-    "severity_level": 3
+    "severity": 3
   },
   {
-    "topic": "Division by Zero",
-    "comment": "Potential division by zero when calculating total tokens.",
-    "confidence": "critical",
-    "reason": "If 'total_tokens' is zero, it will cause a division by zero error.",
-    "solution": "Add a check to ensure 'total_tokens' is not zero before performing the division.",
-    "actual_code": "print(f\"Total tokens used: {total_tokens:,}\")\nprint(f\"  - Input tokens: {total_input_tokens:,}\")\nprint(f\"  - Output tokens: {total_output_tokens:,}\")",
+    "category": "Division by Zero",
+    "description": "Potential division by zero when calculating total tokens.",
+    "impact": "critical",
+    "rationale": "If 'total_tokens' is zero, it will cause a division by zero error.",
+    "recommendation": "Add a check to ensure 'total_tokens' is not zero before performing the division.",
+    "suggested_code": "print(f\"Total tokens used: {total_tokens:,}\")\nprint(f\"  - Input tokens: {total_input_tokens:,}\")\nprint(f\"  - Output tokens: {total_output_tokens:,}\")",
     "fixed_code": "print(f\"Total tokens used: {total_tokens:,}\")\nif total_tokens > 0:\n    print(f\"  - Input tokens: {total_input_tokens:,} ({total_input_tokens/total_tokens:.2%})\")\n    print(f\"  - Output tokens: {total_output_tokens:,} ({total_output_tokens/total_tokens:.2%})\")\nelse:\n    print(\"  - No tokens used.\")",
-    "file_name": "main.py",
+    "file_path": "main.py",
     "start_line": 158,
     "end_line": 163,
-    "severity_level": 7
+    "severity": 7
   },
   {
-    "topic": "File Not Found Handling",
-    "comment": "No error handling for file not found.",
-    "confidence": "important",
-    "reason": "If the specified file does not exist, the program will crash.",
-    "solution": "Add error handling to check if the file exists before processing.",
-    "actual_code": "main(input_file)",
+    "category": "File Not Found Handling",
+    "description": "No error handling for file not found.",
+    "impact": "high",
+    "rationale": "If the specified file does not exist, the program will crash.",
+    "recommendation": "Add error handling to check if the file exists before processing.",
+    "suggested_code": "main(input_file)",
     "fixed_code": "try:\n    main(input_file)\nexcept FileNotFoundError:\n    print(f\"Error: The file '{input_file}' does not exist. Please check the file path and try again.\")\nexcept Exception as e:\n    print(f\"An error occurred: {e}\")",
-    "file_name": "main.py",
+    "file_path": "main.py",
     "start_line": 174,
     "end_line": 175,
-    "severity_level": 6
+    "severity": 6
   }
 ]
\ No newline at end of file
diff --git a/.experiments/code_review/evaluate.py b/.experiments/code_review/evaluate.py
index d36faecb..3f13c579 100644
--- a/.experiments/code_review/evaluate.py
+++ b/.experiments/code_review/evaluate.py
@@ -1,6 +1,7 @@
 import json
 import os
-from fuzzywuzzy import fuzz
+from sklearn.feature_extraction.text import TfidfVectorizer
+from sklearn.metrics.pairwise import cosine_similarity
 
 
 def load_json(file_path):
@@ -8,8 +9,10 @@ def load_json(file_path):
         return json.load(f)
 
 
-def fuzzy_match(str1, str2, threshold=50):
-    return fuzz.partial_ratio(str1.lower(), str2.lower()) >= threshold
+def calculate_similarity(str1, str2):
+    vectorizer = TfidfVectorizer()
+    tfidf_matrix = vectorizer.fit_transform([str1, str2])
+    return cosine_similarity(tfidf_matrix[0:1], tfidf_matrix[1:2])[0][0]
 
 
 def compare_issues(ground_truth, model_issues):
@@ -20,27 +23,32 @@ def compare_issues(ground_truth, model_issues):
     for gt_issue in ground_truth:
         found_match = False
         for model_issue in model_issues:
+            category_similarity = calculate_similarity(
+                gt_issue["category"], model_issue["category"]
+            )
+            description_similarity = calculate_similarity(
+                gt_issue["description"], model_issue["description"]
+            )
+
             if (
-                fuzzy_match(gt_issue["topic"], model_issue["topic"])
-                and fuzzy_match(gt_issue["comment"], model_issue["comment"])
-                and gt_issue["file_name"] == model_issue["file_name"]
+                category_similarity > 0.5
+                and description_similarity > 0.5
+                and gt_issue["file_path"] == model_issue["file_path"]
                 and abs(
                     int(gt_issue.get("start_line", 0))
                     - int(model_issue.get("start_line", -10))
                 )
-                <= 1
+                <= 2
                 and abs(
                     int(gt_issue.get("end_line", 0))
                     - int(model_issue.get("end_line", -10))
                 )
-                <= 1
+                <= 2
                 and abs(
-                    int(gt_issue.get("severity_level", 0))
-                    - int(model_issue.get("severity_level", -10))
+                    int(gt_issue.get("severity", 0))
+                    - int(model_issue.get("severity", -10))
                 )
-                <= 1
-                and gt_issue.get("sentiment", "bad")
-                == model_issue.get("sentiment", "hmm")
+                <= 2
             ):
                 matched.append((gt_issue, model_issue))
                 found_match = True
@@ -62,15 +70,11 @@ def evaluate_model(ground_truth, model_issues):
     total_issues = len(ground_truth)
     issues_found = len(model_issues)
     correct_issues = len(matched)
-    false_positives = len(unmatched_model)
     false_negatives = len(unmatched_gt)
+    false_positives = len(unmatched_model)
 
-    precision = (
-        correct_issues / (correct_issues + false_positives)
-        if (correct_issues + false_positives) > 0
-        else 0
-    )
-    recall = correct_issues / total_issues
+    recall = correct_issues / total_issues if total_issues > 0 else 0
+    precision = correct_issues / issues_found if issues_found > 0 else 0
     f1_score = (
         2 * (precision * recall) / (precision + recall)
         if (precision + recall) > 0
@@ -81,10 +85,10 @@ def evaluate_model(ground_truth, model_issues):
         "total_issues": total_issues,
         "issues_found": issues_found,
         "correct_issues": correct_issues,
-        "false_positives": false_positives,
         "false_negatives": false_negatives,
-        "precision": precision,
+        "false_positives": false_positives,
         "recall": recall,
+        "precision": precision,
         "f1_score": f1_score,
     }
 
@@ -98,8 +102,8 @@ def main(folder_name):
     overall_results = {
         "total_issues": 0,
         "correct_issues": 0,
-        "false_positives": 0,
         "false_negatives": 0,
+        "false_positives": 0,
     }
 
     pr_count = 0
@@ -110,13 +114,8 @@ def main(folder_name):
             ground_truth_path = os.path.join(dataset_path, pr_folder, "issues.json")
             model_path = os.path.join(model_base_path, f"pr_{pr_number}", "issues.json")
 
-            if not os.path.exists(ground_truth_path):
-                print(f"Ground truth file not found for PR {pr_number}")
-                continue
-            if not os.path.exists(model_path):
-                print(
-                    f"Model output file not found for {folder_name} on PR {pr_number}"
-                )
+            if not os.path.exists(ground_truth_path) or not os.path.exists(model_path):
+                print(f"Skipping PR {pr_number} due to missing files")
                 continue
 
             ground_truth = load_json(ground_truth_path)
@@ -129,29 +128,29 @@ def main(folder_name):
             print(
                 f"  Correct issues: {results['correct_issues']}/{results['total_issues']}"
             )
-            print(f"  False positives: {results['false_positives']}")
             print(f"  False negatives: {results['false_negatives']}")
-            print(f"  Precision: {results['precision']:.2f}")
+            print(f"  False positives: {results['false_positives']}")
             print(f"  Recall: {results['recall']:.2f}")
+            print(f"  Precision: {results['precision']:.2f}")
             print(f"  F1 Score: {results['f1_score']:.2f}")
 
             for key in [
                 "total_issues",
                 "correct_issues",
-                "false_positives",
                 "false_negatives",
+                "false_positives",
             ]:
                 overall_results[key] += results[key]
 
             pr_count += 1
 
     if pr_count > 0:
-        overall_precision = overall_results["correct_issues"] / (
-            overall_results["correct_issues"] + overall_results["false_positives"]
-        )
         overall_recall = (
             overall_results["correct_issues"] / overall_results["total_issues"]
         )
+        overall_precision = overall_results["correct_issues"] / (
+            overall_results["correct_issues"] + overall_results["false_positives"]
+        )
         overall_f1 = (
             2
             * (overall_precision * overall_recall)
@@ -165,10 +164,10 @@ def main(folder_name):
         print(
             f"  Correct issues: {overall_results['correct_issues']}/{overall_results['total_issues']}"
         )
-        print(f"  False positives: {overall_results['false_positives']}")
         print(f"  False negatives: {overall_results['false_negatives']}")
-        print(f"  Precision: {overall_precision:.2f}")
+        print(f"  False positives: {overall_results['false_positives']}")
         print(f"  Recall: {overall_recall:.2f}")
+        print(f"  Precision: {overall_precision:.2f}")
         print(f"  F1 Score: {overall_f1:.2f}")
     else:
         print(f"No valid PRs found for evaluation of {folder_name}")
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_222/comments.json b/.experiments/code_review/gpt-4o-mini/no_eval/pr_222/comments.json
deleted file mode 100644
index e09425aa..00000000
--- a/.experiments/code_review/gpt-4o-mini/no_eval/pr_222/comments.json
+++ /dev/null
@@ -1,102 +0,0 @@
-[
-  {
-    "topic": "Environment Variables",
-    "comment": "Using 'os.environ' directly in JSON is not valid.",
-    "confidence": "critical",
-    "reason": "Environment variables should be accessed in the code, not hardcoded in configuration files.",
-    "solution": "Use a placeholder or variable in the code to fetch environment variables.",
-    "actual_code": "                    \"api_key\": \"os.environ/AZURE_API_KEY\",",
-    "fixed_code": "                    \"api_key\": \"AZURE_API_KEY\",  // Access this in the code",
-    "file_name": "config.json",
-    "start_line": 13,
-    "end_line": 13,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "Lack of error handling in database operations.",
-    "confidence": "critical",
-    "reason": "Database operations can fail due to various reasons (e.g., connection issues, SQL errors). Without error handling, the application may crash or behave unexpectedly.",
-    "solution": "Wrap database operations in try-except blocks to handle potential exceptions gracefully.",
-    "actual_code": "41   +1:[+]                 cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))",
-    "fixed_code": "41   +1:[+]                 try:\n41.1   +1:[+]                     cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))\n41.2   +1:[+]                 except Exception as e:\n41.3   +1:[+]                     # Handle the exception (e.g., log it, raise a custom error, etc.)\n41.4   +1:[+]                     raise RuntimeError('Database query failed') from e",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 39,
-    "end_line": 41,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 9
-  },
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Docker",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to Dockerfile, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_name": "Dockerfile",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Docker",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to docker-compose.yml, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "15",
-    "end_line": "15",
-    "side": "RIGHT",
-    "file_name": "docker-compose.yml",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Version Control",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to .gitignore, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "164",
-    "end_line": "164",
-    "side": "RIGHT",
-    "file_name": ".gitignore",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Database",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to db_setup/init.sql, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_name": "db_setup/init.sql",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_222/issues.json b/.experiments/code_review/gpt-4o-mini/no_eval/pr_222/issues.json
deleted file mode 100644
index 061bf872..00000000
--- a/.experiments/code_review/gpt-4o-mini/no_eval/pr_222/issues.json
+++ /dev/null
@@ -1,342 +0,0 @@
-[
-  {
-    "topic": "Environment Variables",
-    "comment": "Using 'os.environ' directly in JSON is not valid.",
-    "confidence": "critical",
-    "reason": "Environment variables should be accessed in the code, not hardcoded in configuration files.",
-    "solution": "Use a placeholder or variable in the code to fetch environment variables.",
-    "actual_code": "                    \"api_key\": \"os.environ/AZURE_API_KEY\",",
-    "fixed_code": "                    \"api_key\": \"AZURE_API_KEY\",  // Access this in the code",
-    "file_name": "config.json",
-    "start_line": 13,
-    "end_line": 13,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Dockerfile",
-    "comment": "Consider using multi-stage builds to reduce image size.",
-    "confidence": "important",
-    "reason": "Multi-stage builds can help keep the final image smaller by excluding build dependencies.",
-    "solution": "Use a multi-stage build pattern to install dependencies and copy only necessary files.",
-    "actual_code": "RUN apt-get update && apt-get install -y \\",
-    "fixed_code": "FROM python:3.9 AS builder\nRUN apt-get update && apt-get install -y build-essential git\n\nFROM python:3.9\nCOPY --from=builder /app /app",
-    "file_name": "Dockerfile",
-    "start_line": 8,
-    "end_line": 12,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 6
-  },
-  {
-    "topic": "Docker Compose",
-    "comment": "Ensure proper indentation for YAML files.",
-    "confidence": "important",
-    "reason": "Improper indentation can lead to YAML parsing errors.",
-    "solution": "Review the indentation levels for all entries in the YAML file.",
-    "actual_code": "     networks:\n       - app-network",
-    "fixed_code": "     networks:\n       - app-network",
-    "file_name": "docker-compose-dev.yml",
-    "start_line": 15,
-    "end_line": 16,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  },
-  {
-    "topic": "SQL Initialization",
-    "comment": "Add comments to SQL statements for clarity.",
-    "confidence": "moderate",
-    "reason": "Comments can help future developers understand the purpose of each SQL command.",
-    "solution": "Add comments above each CREATE TABLE statement.",
-    "actual_code": "CREATE TABLE repositories (",
-    "fixed_code": "-- Table to store repository information\nCREATE TABLE repositories (",
-    "file_name": "db_setup/init.sql",
-    "start_line": 4,
-    "end_line": 4,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 4
-  },
-  {
-    "topic": "Shell Script",
-    "comment": "Add error handling to the shell script.",
-    "confidence": "important",
-    "reason": "Error handling can prevent the script from failing silently.",
-    "solution": "Use 'set -e' at the beginning of the script to exit on errors.",
-    "actual_code": "#!/bin/bash",
-    "fixed_code": "#!/bin/bash\nset -e",
-    "file_name": "install_tree_sitter_languages.sh",
-    "start_line": 1,
-    "end_line": 1,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 7
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "Lack of error handling in database operations.",
-    "confidence": "critical",
-    "reason": "Database operations can fail due to various reasons (e.g., connection issues, SQL errors). Without error handling, the application may crash or behave unexpectedly.",
-    "solution": "Wrap database operations in try-except blocks to handle potential exceptions gracefully.",
-    "actual_code": "41   +1:[+]                 cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))",
-    "fixed_code": "41   +1:[+]                 try:\n41.1   +1:[+]                     cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))\n41.2   +1:[+]                 except Exception as e:\n41.3   +1:[+]                     # Handle the exception (e.g., log it, raise a custom error, etc.)\n41.4   +1:[+]                     raise RuntimeError('Database query failed') from e",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 39,
-    "end_line": 41,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 9
-  },
-  {
-    "topic": "Code Readability",
-    "comment": "Consider adding docstrings to methods for better understanding.",
-    "confidence": "important",
-    "reason": "Docstrings help other developers understand the purpose and usage of methods, especially in larger codebases.",
-    "solution": "Add docstrings to the `__init__` and `custom_query` methods to describe their parameters and return values.",
-    "actual_code": "8    +1:[+]     def __init__(self, *args, **kwargs):",
-    "fixed_code": "8    +1:[+]     def __init__(self, *args, **kwargs):\n8.1   +1:[+]         \"\"\"Initialize the CustomPGVectorStore with table name and other parameters.\"\"\"",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 8,
-    "end_line": 8,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  },
-  {
-    "topic": "Code Readability",
-    "comment": "Consider adding type hints for method parameters and return types.",
-    "confidence": "important",
-    "reason": "Type hints improve code clarity and help with static type checking, making it easier for developers to understand expected types.",
-    "solution": "Add type hints to the `custom_query` method parameters and return type.",
-    "actual_code": "13   +1:[+]     def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:",
-    "fixed_code": "13   +1:[+]     def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[Dict[str, Any]]:",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 13,
-    "end_line": 13,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Code Readability",
-    "comment": "Consider using f-strings for consistency in SQL query construction.",
-    "confidence": "moderate",
-    "reason": "Using f-strings consistently improves readability and reduces the risk of SQL injection if not handled properly.",
-    "solution": "Use f-strings for constructing the SQL query instead of concatenation.",
-    "actual_code": "19   +1:[+]         query = f\"\"\"\n26   +1:[+]{self.table_name}e\n32   +1:[+]             f.repo_id = %s\n36   +1:[+]             %s\n\"\"\"",
-    "fixed_code": "19   +1:[+]         query = f\"\"\"\n19.1   +1:[+]         SELECT \n19.2   +1:[+]             e.node_id,\n19.3   +1:[+]             e.text,\n19.4   +1:[+]             e.metadata,\n19.5   +1:[+]             1 - (e.embedding <=> %s::vector) as similarity\n19.6   +1:[+]         FROM \n19.7   +1:[+]{self.table_name}e\n19.8   +1:[+]         JOIN \n19.9   +1:[+]             function_abstractions fa ON e.node_id = fa.function_id::text\n19.10   +1:[+]         JOIN \n19.11   +1:[+]             files f ON fa.file_id = f.file_id\n19.12   +1:[+]         WHERE \n19.13   +1:[+]             f.repo_id = %s\n19.14   +1:[+]         ORDER BY \n19.15   +1:[+]             similarity DESC\n19.16   +1:[+]         LIMIT \n19.17   +1:[+]             %s\n\"\"\"",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 19,
-    "end_line": 37,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "Potential unhandled exceptions in `generate_abstraction` method.",
-    "confidence": "important",
-    "reason": "Raising the exception without handling it can lead to application crashes.",
-    "solution": "Wrap the call to `self.llm_provider.chat_completion` in a try-except block to handle specific exceptions gracefully.",
-    "actual_code": "            raise e",
-    "fixed_code": "            logger.error(f'Error in generating abstraction:{str(e)}')\n            return None, None",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 218,
-    "end_line": 219,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Logging",
-    "comment": "Inconsistent logging levels for errors and debugging.",
-    "confidence": "important",
-    "reason": "Using `logger.debug` for important errors can lead to missed critical information in production logs.",
-    "solution": "Use `logger.error` for logging errors and `logger.debug` for detailed debugging information.",
-    "actual_code": "             logger.debug(f\"Successfully parsed file:{file_path}\")",
-    "fixed_code": "             logger.info(f\"Successfully parsed file:{file_path}\")",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 107,
-    "end_line": 107,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 6
-  },
-  {
-    "topic": "Code Readability",
-    "comment": "Lack of comments explaining complex logic in methods.",
-    "confidence": "moderate",
-    "reason": "While the code is mostly clear, some complex sections could benefit from additional comments for future maintainability.",
-    "solution": "Add comments to explain the purpose and logic of complex code blocks, especially in `store_code_in_db` and `query` methods.",
-    "actual_code": "",
-    "fixed_code": "    # This method stores the code and its abstraction in the database.\n    # Ensure to handle potential conflicts and return the function ID.",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 246,
-    "end_line": 246,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Environment Variables",
-    "comment": "Direct use of environment variables without validation.",
-    "confidence": "important",
-    "reason": "Using environment variables directly can lead to runtime errors if they are not set or misspelled.",
-    "solution": "Implement checks to ensure that required environment variables are set before using them.",
-    "actual_code": "         self.engine = create_engine(\n             f\"postgresql://{os.environ['POSTGRES_USER']}:{os.environ['POSTGRES_PASSWORD']}@{os.environ['POSTGRES_HOST']}:{os.environ['POSTGRES_PORT']}/{os.environ['POSTGRES_DB']}\",",
-    "fixed_code": "         required_env_vars =['POSTGRES_USER', 'POSTGRES_PASSWORD', 'POSTGRES_HOST', 'POSTGRES_PORT', 'POSTGRES_DB']\n         for var in required_env_vars:\n             if var not in os.environ:\n                 raise EnvironmentError(f'Missing required environment variable:{var}')",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 35,
-    "end_line": 39,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "Broad exception handling in load_language and get_parser methods.",
-    "confidence": "important",
-    "reason": "Using a generic Exception can obscure the root cause of issues and make debugging difficult.",
-    "solution": "Catch specific exceptions where possible, and log the relevant error messages.",
-    "actual_code": "except Exception as e:",
-    "fixed_code": "except (ImportError, ValueError) as e:",
-    "file_name": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 28,
-    "end_line": 28,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 6
-  },
-  {
-    "topic": "Logging",
-    "comment": "Consider adding more context to log messages.",
-    "confidence": "moderate",
-    "reason": "Current log messages may not provide enough context for troubleshooting.",
-    "solution": "Include the function name or additional context in the log messages.",
-    "actual_code": "logger.error(f\"Failed to load language{language}:{str(e)}\")",
-    "fixed_code": "logger.error(f\"{__name__}.load_language failed for{language}:{str(e)}\")",
-    "file_name": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 29,
-    "end_line": 29,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  },
-  {
-    "topic": "Code Duplication",
-    "comment": "Duplicated code in traverse_tree for handling different node types.",
-    "confidence": "important",
-    "reason": "Code duplication can lead to maintenance challenges and potential inconsistencies.",
-    "solution": "Consider refactoring to a helper function that handles common logic.",
-    "actual_code": "return{\"type\": \"function\", \"name\": (node.child_by_field_name(\"name\").text.decode(\"utf8\") if node.child_by_field_name(\"name\") else \"anonymous\"), \"code\": code_bytes[node.start_byte : node.end_byte].decode(\"utf8\")}",
-    "fixed_code": "def extract_node_info(node, code_bytes): ...",
-    "file_name": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 55,
-    "end_line": 68,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Function Naming",
-    "comment": "Function names could be more descriptive.",
-    "confidence": "low",
-    "reason": "Descriptive names improve code readability and maintainability.",
-    "solution": "Consider renaming functions to reflect their purpose more clearly.",
-    "actual_code": "def parse_code(code: str, language: str) -> Dict[str, Any]:",
-    "fixed_code": "def parse_source_code(source_code: str, language: str) -> Dict[str, Any]:",
-    "file_name": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 91,
-    "end_line": 91,
-    "side": "LEFT",
-    "sentiment": "positive",
-    "severity_level": 4
-  },
-  {
-    "topic": "Dependency Management",
-    "comment": "Ensure that dependencies in pyproject.toml are up-to-date.",
-    "confidence": "important",
-    "reason": "Using outdated dependencies can lead to security vulnerabilities and compatibility issues.",
-    "solution": "Regularly review and update dependencies to the latest stable versions.",
-    "actual_code": "python = \"^3.8.1\"",
-    "fixed_code": "python = \"^3.9.0\"",
-    "file_name": "pyproject.toml",
-    "start_line": 13,
-    "end_line": 13,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  },
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Docker",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to Dockerfile, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_name": "Dockerfile",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Docker",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to docker-compose.yml, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "15",
-    "end_line": "15",
-    "side": "RIGHT",
-    "file_name": "docker-compose.yml",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Version Control",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to .gitignore, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "164",
-    "end_line": "164",
-    "side": "RIGHT",
-    "file_name": ".gitignore",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Database",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to db_setup/init.sql, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_name": "db_setup/init.sql",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_222/review.md b/.experiments/code_review/gpt-4o-mini/no_eval/pr_222/review.md
deleted file mode 100644
index a5f856a0..00000000
--- a/.experiments/code_review/gpt-4o-mini/no_eval/pr_222/review.md
+++ /dev/null
@@ -1,402 +0,0 @@
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 23
-- Critical: 7
-- Important: 11
-- Minor: 4
-- Files Affected: 11
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Environment Variables (7 issues)</strong></summary>
-
-### 1. Using 'os.environ' directly in JSON is not valid.
-📁 **File:** `config.json:13`
-⚖️ **Severity:** 8/10
-🔍 **Description:** Environment variables should be accessed in the code, not hardcoded in configuration files.
-💡 **Solution:** Use a placeholder or variable in the code to fetch environment variables.
-
-**Current Code:**
-```python
-                    "api_key": "os.environ/AZURE_API_KEY",
-```
-
-**Suggested Code:**
-```python
-                    "api_key": "AZURE_API_KEY",  // Access this in the code
-```
-
-### 2. Lack of error handling in database operations.
-📁 **File:** `kaizen/retriever/custom_vector_store.py:39`
-⚖️ **Severity:** 9/10
-🔍 **Description:** Database operations can fail due to various reasons (e.g., connection issues, SQL errors). Without error handling, the application may crash or behave unexpectedly.
-💡 **Solution:** Wrap database operations in try-except blocks to handle potential exceptions gracefully.
-
-**Current Code:**
-```python
-41   +1:[+]                 cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))
-```
-
-**Suggested Code:**
-```python
-41   +1:[+]                 try:
-41.1   +1:[+]                     cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))
-41.2   +1:[+]                 except Exception as e:
-41.3   +1:[+]                     # Handle the exception (e.g., log it, raise a custom error, etc.)
-41.4   +1:[+]                     raise RuntimeError('Database query failed') from e
-```
-
-### 3. Changes made to sensitive file
-📁 **File:** `config.json:4`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to config.json, which needs review
-💡 **Solution:** NA
-
-### 4. Changes made to sensitive file
-📁 **File:** `Dockerfile:4`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to Dockerfile, which needs review
-💡 **Solution:** NA
-
-### 5. Changes made to sensitive file
-📁 **File:** `docker-compose.yml:15`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to docker-compose.yml, which needs review
-💡 **Solution:** NA
-
-### 6. Changes made to sensitive file
-📁 **File:** `.gitignore:164`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to .gitignore, which needs review
-💡 **Solution:** NA
-
-### 7. Changes made to sensitive file
-📁 **File:** `db_setup/init.sql:1`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to db_setup/init.sql, which needs review
-💡 **Solution:** NA
-
-</details>
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Dockerfile (11 issues)</strong></summary>
-
-### 1. Consider using multi-stage builds to reduce image size.
-📁 **File:** `Dockerfile:8`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Multi-stage builds can help keep the final image smaller by excluding build dependencies.
-💡 **Solution:** Use a multi-stage build pattern to install dependencies and copy only necessary files.
-
-**Current Code:**
-```python
-RUN apt-get update && apt-get install -y \
-```
-
-**Suggested Code:**
-```python
-FROM python:3.9 AS builder
-RUN apt-get update && apt-get install -y build-essential git
-
-FROM python:3.9
-COPY --from=builder /app /app
-```
-
-### 2. Ensure proper indentation for YAML files.
-📁 **File:** `docker-compose-dev.yml:15`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Improper indentation can lead to YAML parsing errors.
-💡 **Solution:** Review the indentation levels for all entries in the YAML file.
-
-**Current Code:**
-```python
-     networks:
-       - app-network
-```
-
-**Suggested Code:**
-```python
-     networks:
-       - app-network
-```
-
-### 3. Add error handling to the shell script.
-📁 **File:** `install_tree_sitter_languages.sh:1`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Error handling can prevent the script from failing silently.
-💡 **Solution:** Use 'set -e' at the beginning of the script to exit on errors.
-
-**Current Code:**
-```python
-#!/bin/bash
-```
-
-**Suggested Code:**
-```python
-#!/bin/bash
-set -e
-```
-
-### 4. Consider adding docstrings to methods for better understanding.
-📁 **File:** `kaizen/retriever/custom_vector_store.py:8`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Docstrings help other developers understand the purpose and usage of methods, especially in larger codebases.
-💡 **Solution:** Add docstrings to the `__init__` and `custom_query` methods to describe their parameters and return values.
-
-**Current Code:**
-```python
-8    +1:[+]     def __init__(self, *args, **kwargs):
-```
-
-**Suggested Code:**
-```python
-8    +1:[+]     def __init__(self, *args, **kwargs):
-8.1   +1:[+]         """Initialize the CustomPGVectorStore with table name and other parameters."""
-```
-
-### 5. Consider adding type hints for method parameters and return types.
-📁 **File:** `kaizen/retriever/custom_vector_store.py:13`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Type hints improve code clarity and help with static type checking, making it easier for developers to understand expected types.
-💡 **Solution:** Add type hints to the `custom_query` method parameters and return type.
-
-**Current Code:**
-```python
-13   +1:[+]     def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:
-```
-
-**Suggested Code:**
-```python
-13   +1:[+]     def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[Dict[str, Any]]:
-```
-
-### 6. Potential unhandled exceptions in `generate_abstraction` method.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:218`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Raising the exception without handling it can lead to application crashes.
-💡 **Solution:** Wrap the call to `self.llm_provider.chat_completion` in a try-except block to handle specific exceptions gracefully.
-
-**Current Code:**
-```python
-            raise e
-```
-
-**Suggested Code:**
-```python
-            logger.error(f'Error in generating abstraction:{str(e)}')
-            return None, None
-```
-
-### 7. Inconsistent logging levels for errors and debugging.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:107`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Using `logger.debug` for important errors can lead to missed critical information in production logs.
-💡 **Solution:** Use `logger.error` for logging errors and `logger.debug` for detailed debugging information.
-
-**Current Code:**
-```python
-             logger.debug(f"Successfully parsed file:{file_path}")
-```
-
-**Suggested Code:**
-```python
-             logger.info(f"Successfully parsed file:{file_path}")
-```
-
-### 8. Direct use of environment variables without validation.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:35`
-⚖️ **Severity:** 8/10
-🔍 **Description:** Using environment variables directly can lead to runtime errors if they are not set or misspelled.
-💡 **Solution:** Implement checks to ensure that required environment variables are set before using them.
-
-**Current Code:**
-```python
-         self.engine = create_engine(
-             f"postgresql://{os.environ['POSTGRES_USER']}:{os.environ['POSTGRES_PASSWORD']}@{os.environ['POSTGRES_HOST']}:{os.environ['POSTGRES_PORT']}/{os.environ['POSTGRES_DB']}",
-```
-
-**Suggested Code:**
-```python
-         required_env_vars =['POSTGRES_USER', 'POSTGRES_PASSWORD', 'POSTGRES_HOST', 'POSTGRES_PORT', 'POSTGRES_DB']
-         for var in required_env_vars:
-             if var not in os.environ:
-                 raise EnvironmentError(f'Missing required environment variable:{var}')
-```
-
-### 9. Broad exception handling in load_language and get_parser methods.
-📁 **File:** `kaizen/retriever/tree_sitter_utils.py:28`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Using a generic Exception can obscure the root cause of issues and make debugging difficult.
-💡 **Solution:** Catch specific exceptions where possible, and log the relevant error messages.
-
-**Current Code:**
-```python
-except Exception as e:
-```
-
-**Suggested Code:**
-```python
-except (ImportError, ValueError) as e:
-```
-
-### 10. Duplicated code in traverse_tree for handling different node types.
-📁 **File:** `kaizen/retriever/tree_sitter_utils.py:55`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Code duplication can lead to maintenance challenges and potential inconsistencies.
-💡 **Solution:** Consider refactoring to a helper function that handles common logic.
-
-**Current Code:**
-```python
-return{"type": "function", "name": (node.child_by_field_name("name").text.decode("utf8") if node.child_by_field_name("name") else "anonymous"), "code": code_bytes[node.start_byte : node.end_byte].decode("utf8")}
-```
-
-**Suggested Code:**
-```python
-def extract_node_info(node, code_bytes): ...
-```
-
-### 11. Ensure that dependencies in pyproject.toml are up-to-date.
-📁 **File:** `pyproject.toml:13`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Using outdated dependencies can lead to security vulnerabilities and compatibility issues.
-💡 **Solution:** Regularly review and update dependencies to the latest stable versions.
-
-**Current Code:**
-```python
-python = "^3.8.1"
-```
-
-**Suggested Code:**
-```python
-python = "^3.9.0"
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (5 issues)</strong></summary>
-
-<details>
-<summary><strong>SQL Initialization (4 issues)</strong></summary>
-
-### 1. Add comments to SQL statements for clarity.
-📁 **File:** `db_setup/init.sql:4`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Comments can help future developers understand the purpose of each SQL command.
-💡 **Solution:** Add comments above each CREATE TABLE statement.
-
-**Current Code:**
-```python
-CREATE TABLE repositories (
-```
-
-**Suggested Code:**
-```python
--- Table to store repository information
-CREATE TABLE repositories (
-```
-
-### 2. Consider using f-strings for consistency in SQL query construction.
-📁 **File:** `kaizen/retriever/custom_vector_store.py:19`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Using f-strings consistently improves readability and reduces the risk of SQL injection if not handled properly.
-💡 **Solution:** Use f-strings for constructing the SQL query instead of concatenation.
-
-**Current Code:**
-```python
-19   +1:[+]         query = f"""
-26   +1:[+]{self.table_name}e
-32   +1:[+]             f.repo_id = %s
-36   +1:[+]             %s
-"""
-```
-
-**Suggested Code:**
-```python
-19   +1:[+]         query = f"""
-19.1   +1:[+]         SELECT 
-19.2   +1:[+]             e.node_id,
-19.3   +1:[+]             e.text,
-19.4   +1:[+]             e.metadata,
-19.5   +1:[+]             1 - (e.embedding <=> %s::vector) as similarity
-19.6   +1:[+]         FROM 
-19.7   +1:[+]{self.table_name}e
-19.8   +1:[+]         JOIN 
-19.9   +1:[+]             function_abstractions fa ON e.node_id = fa.function_id::text
-19.10   +1:[+]         JOIN 
-19.11   +1:[+]             files f ON fa.file_id = f.file_id
-19.12   +1:[+]         WHERE 
-19.13   +1:[+]             f.repo_id = %s
-19.14   +1:[+]         ORDER BY 
-19.15   +1:[+]             similarity DESC
-19.16   +1:[+]         LIMIT 
-19.17   +1:[+]             %s
-"""
-```
-
-### 3. Lack of comments explaining complex logic in methods.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:246`
-⚖️ **Severity:** 4/10
-🔍 **Description:** While the code is mostly clear, some complex sections could benefit from additional comments for future maintainability.
-💡 **Solution:** Add comments to explain the purpose and logic of complex code blocks, especially in `store_code_in_db` and `query` methods.
-
-**Current Code:**
-```python
-
-```
-
-**Suggested Code:**
-```python
-    # This method stores the code and its abstraction in the database.
-    # Ensure to handle potential conflicts and return the function ID.
-```
-
-### 4. Consider adding more context to log messages.
-📁 **File:** `kaizen/retriever/tree_sitter_utils.py:29`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Current log messages may not provide enough context for troubleshooting.
-💡 **Solution:** Include the function name or additional context in the log messages.
-
-**Current Code:**
-```python
-logger.error(f"Failed to load language{language}:{str(e)}")
-```
-
-**Suggested Code:**
-```python
-logger.error(f"{__name__}.load_language failed for{language}:{str(e)}")
-```
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage 
-{"prompt_tokens": 21545, "completion_tokens": 3763, "total_tokens": 25308}
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_335/comments.json b/.experiments/code_review/gpt-4o-mini/no_eval/pr_335/comments.json
deleted file mode 100644
index dfddeae0..00000000
--- a/.experiments/code_review/gpt-4o-mini/no_eval/pr_335/comments.json
+++ /dev/null
@@ -1,17 +0,0 @@
-[
-  {
-    "topic": "Parameter Removal",
-    "comment": "Removal of 'reeval_response' parameter may affect functionality.",
-    "confidence": "critical",
-    "reason": "This parameter is used in multiple methods, and its removal could lead to unexpected behavior.",
-    "solution": "Evaluate the necessity of this parameter and ensure that its removal does not break existing functionality.",
-    "actual_code": "reeval_response: bool = False,",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/pr_description.py",
-    "start_line": 43,
-    "end_line": 43,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 9
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_335/issues.json b/.experiments/code_review/gpt-4o-mini/no_eval/pr_335/issues.json
deleted file mode 100644
index b93239cf..00000000
--- a/.experiments/code_review/gpt-4o-mini/no_eval/pr_335/issues.json
+++ /dev/null
@@ -1,77 +0,0 @@
-[
-  {
-    "topic": "Imports",
-    "comment": "Inconsistent import statements.",
-    "confidence": "important",
-    "reason": "Changing import paths can lead to confusion and potential import errors.",
-    "solution": "Ensure that all import paths are updated consistently throughout the codebase.",
-    "actual_code": "from kaizen.llms.prompts.code_review_prompts import (",
-    "fixed_code": "from kaizen.llms.prompts.pr_desc_prompts import (",
-    "file_name": "kaizen/generator/pr_description.py",
-    "start_line": 8,
-    "end_line": 8,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 6
-  },
-  {
-    "topic": "Parameter Removal",
-    "comment": "Removal of 'reeval_response' parameter may affect functionality.",
-    "confidence": "critical",
-    "reason": "This parameter is used in multiple methods, and its removal could lead to unexpected behavior.",
-    "solution": "Evaluate the necessity of this parameter and ensure that its removal does not break existing functionality.",
-    "actual_code": "reeval_response: bool = False,",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/pr_description.py",
-    "start_line": 43,
-    "end_line": 43,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 9
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "Lack of error handling for potential exceptions.",
-    "confidence": "important",
-    "reason": "The absence of error handling can cause the application to crash unexpectedly.",
-    "solution": "Implement try-except blocks where appropriate to handle potential exceptions gracefully.",
-    "actual_code": "raise Exception(\"Both diff_text and pull_request_files are empty!\")",
-    "fixed_code": "raise ValueError(\"Both diff_text and pull_request_files are empty!\")",
-    "file_name": "kaizen/generator/pr_description.py",
-    "start_line": 51,
-    "end_line": 51,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Functionality Changes",
-    "comment": "Changes to function calls may alter expected behavior.",
-    "confidence": "important",
-    "reason": "Switching from 'chat_completion_with_json' to 'chat_completion' may change the output format.",
-    "solution": "Review the expected output of the new function and ensure compatibility with existing code.",
-    "actual_code": "resp, usage = self.provider.chat_completion_with_json(prompt, user=user)",
-    "fixed_code": "resp, usage = self.provider.chat_completion(prompt, user=user)",
-    "file_name": "kaizen/generator/pr_description.py",
-    "start_line": 83,
-    "end_line": 83,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 8
-  },
-  {
-    "topic": "Documentation",
-    "comment": "Missing documentation for new prompts.",
-    "confidence": "moderate",
-    "reason": "Lack of documentation can make it harder for other developers to understand the purpose of new prompts.",
-    "solution": "Add docstrings or comments explaining the purpose of each new prompt.",
-    "actual_code": "",
-    "fixed_code": "# This prompt is used to generate a PR description based on the provided details.",
-    "file_name": "kaizen/llms/prompts/pr_desc_prompts.py",
-    "start_line": 1,
-    "end_line": 1,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_335/review.md b/.experiments/code_review/gpt-4o-mini/no_eval/pr_335/review.md
deleted file mode 100644
index 7fdf7ed4..00000000
--- a/.experiments/code_review/gpt-4o-mini/no_eval/pr_335/review.md
+++ /dev/null
@@ -1,139 +0,0 @@
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 5
-- Critical: 1
-- Important: 3
-- Minor: 1
-- Files Affected: 2
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Parameter Removal (1 issues)</strong></summary>
-
-### 1. Removal of 'reeval_response' parameter may affect functionality.
-📁 **File:** `kaizen/generator/pr_description.py:43`
-⚖️ **Severity:** 9/10
-🔍 **Description:** This parameter is used in multiple methods, and its removal could lead to unexpected behavior.
-💡 **Solution:** Evaluate the necessity of this parameter and ensure that its removal does not break existing functionality.
-
-**Current Code:**
-```python
-reeval_response: bool = False,
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Imports (3 issues)</strong></summary>
-
-### 1. Inconsistent import statements.
-📁 **File:** `kaizen/generator/pr_description.py:8`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Changing import paths can lead to confusion and potential import errors.
-💡 **Solution:** Ensure that all import paths are updated consistently throughout the codebase.
-
-**Current Code:**
-```python
-from kaizen.llms.prompts.code_review_prompts import (
-```
-
-**Suggested Code:**
-```python
-from kaizen.llms.prompts.pr_desc_prompts import (
-```
-
-### 2. Lack of error handling for potential exceptions.
-📁 **File:** `kaizen/generator/pr_description.py:51`
-⚖️ **Severity:** 7/10
-🔍 **Description:** The absence of error handling can cause the application to crash unexpectedly.
-💡 **Solution:** Implement try-except blocks where appropriate to handle potential exceptions gracefully.
-
-**Current Code:**
-```python
-raise Exception("Both diff_text and pull_request_files are empty!")
-```
-
-**Suggested Code:**
-```python
-raise ValueError("Both diff_text and pull_request_files are empty!")
-```
-
-### 3. Changes to function calls may alter expected behavior.
-📁 **File:** `kaizen/generator/pr_description.py:83`
-⚖️ **Severity:** 8/10
-🔍 **Description:** Switching from 'chat_completion_with_json' to 'chat_completion' may change the output format.
-💡 **Solution:** Review the expected output of the new function and ensure compatibility with existing code.
-
-**Current Code:**
-```python
-resp, usage = self.provider.chat_completion_with_json(prompt, user=user)
-```
-
-**Suggested Code:**
-```python
-resp, usage = self.provider.chat_completion(prompt, user=user)
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (1 issues)</strong></summary>
-
-<details>
-<summary><strong>Documentation (1 issues)</strong></summary>
-
-### 1. Missing documentation for new prompts.
-📁 **File:** `kaizen/llms/prompts/pr_desc_prompts.py:1`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Lack of documentation can make it harder for other developers to understand the purpose of new prompts.
-💡 **Solution:** Add docstrings or comments explaining the purpose of each new prompt.
-
-**Current Code:**
-```python
-
-```
-
-**Suggested Code:**
-```python
-# This prompt is used to generate a PR description based on the provided details.
-```
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage 
-{"prompt_tokens": 6751, "completion_tokens": 826, "total_tokens": 7577}
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_440/issues.json b/.experiments/code_review/gpt-4o-mini/no_eval/pr_440/issues.json
deleted file mode 100644
index 7e262d38..00000000
--- a/.experiments/code_review/gpt-4o-mini/no_eval/pr_440/issues.json
+++ /dev/null
@@ -1,47 +0,0 @@
-[
-  {
-    "topic": "Repository Setup",
-    "comment": "The setup_repository method is commented out in the previous version, which may lead to confusion about its necessity.",
-    "confidence": "important",
-    "reason": "Commenting out essential setup code can lead to runtime errors if the repository is not properly initialized.",
-    "solution": "Ensure that the setup_repository method is called appropriately to avoid potential issues.",
-    "actual_code": "analyzer.setup_repository(\"./github_app/\")",
-    "fixed_code": "analyzer.setup_repository(\"./github_app/\")",
-    "file_name": "examples/ragify_codebase/main.py",
-    "start_line": 7,
-    "end_line": 7,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Code Comment",
-    "comment": "The TODO comment lacks specificity regarding how to handle duplicates.",
-    "confidence": "moderate",
-    "reason": "Vague comments can lead to misunderstandings and may not provide enough guidance for future developers.",
-    "solution": "Specify the conditions under which duplicates should be checked and how to handle them.",
-    "actual_code": "# TODO: DONT PUSH DUPLICATE",
-    "fixed_code": "# TODO: Implement a check to prevent pushing duplicate embeddings based on a unique identifier.",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 157,
-    "end_line": 157,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  },
-  {
-    "topic": "Dependency Update",
-    "comment": "The dependency version for llama-index-core has been updated without a clear reason.",
-    "confidence": "important",
-    "reason": "Updating dependencies can introduce breaking changes; it's important to ensure compatibility.",
-    "solution": "Review the changelog for llama-index-core to confirm that the new version does not introduce breaking changes.",
-    "actual_code": "llama-index-core = \"^0.10.47\"",
-    "fixed_code": "llama-index-core = \"0.10.65\"",
-    "file_name": "pyproject.toml",
-    "start_line": 27,
-    "end_line": 27,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 6
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_440/review.md b/.experiments/code_review/gpt-4o-mini/no_eval/pr_440/review.md
deleted file mode 100644
index 08c9e2bc..00000000
--- a/.experiments/code_review/gpt-4o-mini/no_eval/pr_440/review.md
+++ /dev/null
@@ -1,100 +0,0 @@
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 3
-- Critical: 0
-- Important: 2
-- Minor: 1
-- Files Affected: 3
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Repository Setup (2 issues)</strong></summary>
-
-### 1. The setup_repository method is commented out in the previous version, which may lead to confusion about its necessity.
-📁 **File:** `examples/ragify_codebase/main.py:7`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Commenting out essential setup code can lead to runtime errors if the repository is not properly initialized.
-💡 **Solution:** Ensure that the setup_repository method is called appropriately to avoid potential issues.
-
-**Current Code:**
-```python
-analyzer.setup_repository("./github_app/")
-```
-
-**Suggested Code:**
-```python
-analyzer.setup_repository("./github_app/")
-```
-
-### 2. The dependency version for llama-index-core has been updated without a clear reason.
-📁 **File:** `pyproject.toml:27`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Updating dependencies can introduce breaking changes; it's important to ensure compatibility.
-💡 **Solution:** Review the changelog for llama-index-core to confirm that the new version does not introduce breaking changes.
-
-**Current Code:**
-```python
-llama-index-core = "^0.10.47"
-```
-
-**Suggested Code:**
-```python
-llama-index-core = "0.10.65"
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (1 issues)</strong></summary>
-
-<details>
-<summary><strong>Code Comment (1 issues)</strong></summary>
-
-### 1. The TODO comment lacks specificity regarding how to handle duplicates.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:157`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Vague comments can lead to misunderstandings and may not provide enough guidance for future developers.
-💡 **Solution:** Specify the conditions under which duplicates should be checked and how to handle them.
-
-**Current Code:**
-```python
-# TODO: DONT PUSH DUPLICATE
-```
-
-**Suggested Code:**
-```python
-# TODO: Implement a check to prevent pushing duplicate embeddings based on a unique identifier.
-```
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage 
-{"prompt_tokens": 1364, "completion_tokens": 544, "total_tokens": 1908}
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_476/comments.json b/.experiments/code_review/gpt-4o-mini/no_eval/pr_476/comments.json
deleted file mode 100644
index a9d40eac..00000000
--- a/.experiments/code_review/gpt-4o-mini/no_eval/pr_476/comments.json
+++ /dev/null
@@ -1,16 +0,0 @@
-[
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_476/issues.json b/.experiments/code_review/gpt-4o-mini/no_eval/pr_476/issues.json
deleted file mode 100644
index 498888fb..00000000
--- a/.experiments/code_review/gpt-4o-mini/no_eval/pr_476/issues.json
+++ /dev/null
@@ -1,91 +0,0 @@
-[
-  {
-    "topic": "Error Handling",
-    "comment": "Broad exception handling can obscure specific errors.",
-    "confidence": "important",
-    "reason": "Using a generic Exception can make debugging difficult and hide underlying issues.",
-    "solution": "Catch specific exceptions where possible.",
-    "actual_code": "except Exception:",
-    "fixed_code": "except KeyError:",
-    "file_name": "github_app/github_helper/pull_requests.py",
-    "start_line": 140,
-    "end_line": 141,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 6
-  },
-  {
-    "topic": "Function Signature",
-    "comment": "The function 'post_pull_request' has an additional parameter that should be documented.",
-    "confidence": "important",
-    "reason": "New parameters should be documented to ensure clarity for future maintainers.",
-    "solution": "Update the function docstring to include the 'tests' parameter.",
-    "actual_code": "def post_pull_request(url, data, installation_id, tests=None):",
-    "fixed_code": "def post_pull_request(url, data, installation_id, tests=None):  # tests: List of test files",
-    "file_name": "github_app/github_helper/pull_requests.py",
-    "start_line": 106,
-    "end_line": 106,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  },
-  {
-    "topic": "Code Readability",
-    "comment": "The new function 'sort_files' lacks a docstring.",
-    "confidence": "important",
-    "reason": "Docstrings are essential for understanding the purpose and usage of functions.",
-    "solution": "Add a docstring to describe the function's purpose and parameters.",
-    "actual_code": "def sort_files(files):",
-    "fixed_code": "def sort_files(files):  # Sorts a list of file dictionaries by filename.",
-    "file_name": "github_app/github_helper/pull_requests.py",
-    "start_line": 184,
-    "end_line": 184,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Variable Naming",
-    "comment": "The variable 'tests' could be more descriptive.",
-    "confidence": "moderate",
-    "reason": "Descriptive variable names improve code readability and maintainability.",
-    "solution": "Consider renaming 'tests' to 'generated_tests' for clarity.",
-    "actual_code": "tests = generate_tests(pr_files)",
-    "fixed_code": "generated_tests = generate_tests(pr_files)",
-    "file_name": "github_app/github_helper/pull_requests.py",
-    "start_line": 58,
-    "end_line": 58,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 3
-  },
-  {
-    "topic": "Logging",
-    "comment": "Consider using logging instead of print statements for error reporting.",
-    "confidence": "important",
-    "reason": "Using logging allows for better control over the output and can be configured for different environments.",
-    "solution": "Replace print statements with appropriate logging calls.",
-    "actual_code": "print(\"Error\")",
-    "fixed_code": "logger.error(\"Error occurred\")",
-    "file_name": "github_app/github_helper/pull_requests.py",
-    "start_line": 141,
-    "end_line": 141,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_476/review.md b/.experiments/code_review/gpt-4o-mini/no_eval/pr_476/review.md
deleted file mode 100644
index 5d27ff4f..00000000
--- a/.experiments/code_review/gpt-4o-mini/no_eval/pr_476/review.md
+++ /dev/null
@@ -1,145 +0,0 @@
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 6
-- Critical: 1
-- Important: 4
-- Minor: 1
-- Files Affected: 2
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Configuration (1 issues)</strong></summary>
-
-### 1. Changes made to sensitive file
-📁 **File:** `config.json:1`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to config.json, which needs review
-💡 **Solution:** NA
-
-</details>
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Error Handling (4 issues)</strong></summary>
-
-### 1. Broad exception handling can obscure specific errors.
-📁 **File:** `github_app/github_helper/pull_requests.py:140`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Using a generic Exception can make debugging difficult and hide underlying issues.
-💡 **Solution:** Catch specific exceptions where possible.
-
-**Current Code:**
-```python
-except Exception:
-```
-
-**Suggested Code:**
-```python
-except KeyError:
-```
-
-### 2. The function 'post_pull_request' has an additional parameter that should be documented.
-📁 **File:** `github_app/github_helper/pull_requests.py:106`
-⚖️ **Severity:** 5/10
-🔍 **Description:** New parameters should be documented to ensure clarity for future maintainers.
-💡 **Solution:** Update the function docstring to include the 'tests' parameter.
-
-**Current Code:**
-```python
-def post_pull_request(url, data, installation_id, tests=None):
-```
-
-**Suggested Code:**
-```python
-def post_pull_request(url, data, installation_id, tests=None):  # tests: List of test files
-```
-
-### 3. The new function 'sort_files' lacks a docstring.
-📁 **File:** `github_app/github_helper/pull_requests.py:184`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Docstrings are essential for understanding the purpose and usage of functions.
-💡 **Solution:** Add a docstring to describe the function's purpose and parameters.
-
-**Current Code:**
-```python
-def sort_files(files):
-```
-
-**Suggested Code:**
-```python
-def sort_files(files):  # Sorts a list of file dictionaries by filename.
-```
-
-### 4. Consider using logging instead of print statements for error reporting.
-📁 **File:** `github_app/github_helper/pull_requests.py:141`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Using logging allows for better control over the output and can be configured for different environments.
-💡 **Solution:** Replace print statements with appropriate logging calls.
-
-**Current Code:**
-```python
-print("Error")
-```
-
-**Suggested Code:**
-```python
-logger.error("Error occurred")
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (1 issues)</strong></summary>
-
-<details>
-<summary><strong>Variable Naming (1 issues)</strong></summary>
-
-### 1. The variable 'tests' could be more descriptive.
-📁 **File:** `github_app/github_helper/pull_requests.py:58`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Descriptive variable names improve code readability and maintainability.
-💡 **Solution:** Consider renaming 'tests' to 'generated_tests' for clarity.
-
-**Current Code:**
-```python
-tests = generate_tests(pr_files)
-```
-
-**Suggested Code:**
-```python
-generated_tests = generate_tests(pr_files)
-```
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage 
-{"prompt_tokens": 4007, "completion_tokens": 796, "total_tokens": 4803}
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_5/comments.json b/.experiments/code_review/gpt-4o-mini/no_eval/pr_5/comments.json
deleted file mode 100644
index c38f9fea..00000000
--- a/.experiments/code_review/gpt-4o-mini/no_eval/pr_5/comments.json
+++ /dev/null
@@ -1,47 +0,0 @@
-[
-  {
-    "topic": "Error Handling",
-    "comment": "API call may fail without a retry mechanism.",
-    "confidence": "critical",
-    "reason": "The completion function call does not handle potential API failures.",
-    "solution": "Implement a retry mechanism or error handling for API calls.",
-    "actual_code": "response = completion(",
-    "fixed_code": "# Implement retry logic here",
-    "file_name": "main.py",
-    "start_line": 66,
-    "end_line": 68,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Silent Failure",
-    "comment": "Silent failure without logging in case of JSON parsing error.",
-    "confidence": "critical",
-    "reason": "The code does not log the error, making debugging difficult.",
-    "solution": "Log the error message to provide feedback in case of failure.",
-    "actual_code": "print(f\"Failed to parse content for applicant\")",
-    "fixed_code": "print(f\"Failed to parse content for applicant:{e}\")",
-    "file_name": "main.py",
-    "start_line": 86,
-    "end_line": 86,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Division by Zero",
-    "comment": "Potential division by zero error.",
-    "confidence": "critical",
-    "reason": "If total_tokens is zero, this will raise an error.",
-    "solution": "Add a check to prevent division by zero.",
-    "actual_code": "print(f\"Total tokens used:{total_tokens:,}\")",
-    "fixed_code": "if total_tokens > 0: print(f\"Total tokens used:{total_tokens:,}\")",
-    "file_name": "main.py",
-    "start_line": 159,
-    "end_line": 159,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 9
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_5/issues.json b/.experiments/code_review/gpt-4o-mini/no_eval/pr_5/issues.json
deleted file mode 100644
index 5dd7c900..00000000
--- a/.experiments/code_review/gpt-4o-mini/no_eval/pr_5/issues.json
+++ /dev/null
@@ -1,107 +0,0 @@
-[
-  {
-    "topic": "Imports",
-    "comment": "Unused import detected.",
-    "confidence": "trivial",
-    "reason": "The import of 'random' is not utilized in the code.",
-    "solution": "Remove the unused import to clean up the code.",
-    "actual_code": "import random  # Unused import",
-    "fixed_code": "",
-    "file_name": "main.py",
-    "start_line": 8,
-    "end_line": 8,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 2
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "API call may fail without a retry mechanism.",
-    "confidence": "critical",
-    "reason": "The completion function call does not handle potential API failures.",
-    "solution": "Implement a retry mechanism or error handling for API calls.",
-    "actual_code": "response = completion(",
-    "fixed_code": "# Implement retry logic here",
-    "file_name": "main.py",
-    "start_line": 66,
-    "end_line": 68,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Silent Failure",
-    "comment": "Silent failure without logging in case of JSON parsing error.",
-    "confidence": "critical",
-    "reason": "The code does not log the error, making debugging difficult.",
-    "solution": "Log the error message to provide feedback in case of failure.",
-    "actual_code": "print(f\"Failed to parse content for applicant\")",
-    "fixed_code": "print(f\"Failed to parse content for applicant:{e}\")",
-    "file_name": "main.py",
-    "start_line": 86,
-    "end_line": 86,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Inefficient Progress Reporting",
-    "comment": "Inefficient way to print progress.",
-    "confidence": "important",
-    "reason": "The current method of printing progress can be improved for better performance.",
-    "solution": "Consider using a logging library or a more efficient progress reporting method.",
-    "actual_code": "print(f\"\\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}\", end=\"\", flush=True)",
-    "fixed_code": "# Use logging or a more efficient progress reporting",
-    "file_name": "main.py",
-    "start_line": 121,
-    "end_line": 121,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  },
-  {
-    "topic": "Redundant Code",
-    "comment": "Redundant check for empty DataFrame.",
-    "confidence": "moderate",
-    "reason": "The check for an empty DataFrame is unnecessary as the process will handle it.",
-    "solution": "Remove the redundant check to simplify the code.",
-    "actual_code": "if len(df) == 0:",
-    "fixed_code": "",
-    "file_name": "main.py",
-    "start_line": 142,
-    "end_line": 143,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Division by Zero",
-    "comment": "Potential division by zero error.",
-    "confidence": "critical",
-    "reason": "If total_tokens is zero, this will raise an error.",
-    "solution": "Add a check to prevent division by zero.",
-    "actual_code": "print(f\"Total tokens used:{total_tokens:,}\")",
-    "fixed_code": "if total_tokens > 0: print(f\"Total tokens used:{total_tokens:,}\")",
-    "file_name": "main.py",
-    "start_line": 159,
-    "end_line": 159,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 9
-  },
-  {
-    "topic": "File Handling",
-    "comment": "No error handling for file not found.",
-    "confidence": "important",
-    "reason": "If the file does not exist, it will raise an unhandled exception.",
-    "solution": "Add error handling to manage file not found scenarios.",
-    "actual_code": "main(input_file)",
-    "fixed_code": "# Add error handling for file not found",
-    "file_name": "main.py",
-    "start_line": 175,
-    "end_line": 175,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 6
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o-mini/no_eval/pr_5/review.md b/.experiments/code_review/gpt-4o-mini/no_eval/pr_5/review.md
deleted file mode 100644
index fcc46911..00000000
--- a/.experiments/code_review/gpt-4o-mini/no_eval/pr_5/review.md
+++ /dev/null
@@ -1,155 +0,0 @@
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 7
-- Critical: 3
-- Important: 2
-- Minor: 1
-- Files Affected: 1
-## 🏆 Code Quality
-[███████████████░░░░░] 75% (Fair)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Error Handling (3 issues)</strong></summary>
-
-### 1. API call may fail without a retry mechanism.
-📁 **File:** `main.py:66`
-⚖️ **Severity:** 8/10
-🔍 **Description:** The completion function call does not handle potential API failures.
-💡 **Solution:** Implement a retry mechanism or error handling for API calls.
-
-**Current Code:**
-```python
-response = completion(
-```
-
-**Suggested Code:**
-```python
-# Implement retry logic here
-```
-
-### 2. Silent failure without logging in case of JSON parsing error.
-📁 **File:** `main.py:86`
-⚖️ **Severity:** 7/10
-🔍 **Description:** The code does not log the error, making debugging difficult.
-💡 **Solution:** Log the error message to provide feedback in case of failure.
-
-**Current Code:**
-```python
-print(f"Failed to parse content for applicant")
-```
-
-**Suggested Code:**
-```python
-print(f"Failed to parse content for applicant:{e}")
-```
-
-### 3. Potential division by zero error.
-📁 **File:** `main.py:159`
-⚖️ **Severity:** 9/10
-🔍 **Description:** If total_tokens is zero, this will raise an error.
-💡 **Solution:** Add a check to prevent division by zero.
-
-**Current Code:**
-```python
-print(f"Total tokens used:{total_tokens:,}")
-```
-
-**Suggested Code:**
-```python
-if total_tokens > 0: print(f"Total tokens used:{total_tokens:,}")
-```
-
-</details>
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Inefficient Progress Reporting (2 issues)</strong></summary>
-
-### 1. Inefficient way to print progress.
-📁 **File:** `main.py:121`
-⚖️ **Severity:** 5/10
-🔍 **Description:** The current method of printing progress can be improved for better performance.
-💡 **Solution:** Consider using a logging library or a more efficient progress reporting method.
-
-**Current Code:**
-```python
-print(f"\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}", end="", flush=True)
-```
-
-**Suggested Code:**
-```python
-# Use logging or a more efficient progress reporting
-```
-
-### 2. No error handling for file not found.
-📁 **File:** `main.py:175`
-⚖️ **Severity:** 6/10
-🔍 **Description:** If the file does not exist, it will raise an unhandled exception.
-💡 **Solution:** Add error handling to manage file not found scenarios.
-
-**Current Code:**
-```python
-main(input_file)
-```
-
-**Suggested Code:**
-```python
-# Add error handling for file not found
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (2 issues)</strong></summary>
-
-<details>
-<summary><strong>Redundant Code (1 issues)</strong></summary>
-
-### 1. Redundant check for empty DataFrame.
-📁 **File:** `main.py:142`
-⚖️ **Severity:** 4/10
-🔍 **Description:** The check for an empty DataFrame is unnecessary as the process will handle it.
-💡 **Solution:** Remove the redundant check to simplify the code.
-
-**Current Code:**
-```python
-if len(df) == 0:
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage 
-{"prompt_tokens": 6154, "completion_tokens": 1046, "total_tokens": 7200}
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_222/comments.json b/.experiments/code_review/gpt-4o/no_eval/pr_222/comments.json
deleted file mode 100644
index 5ebff860..00000000
--- a/.experiments/code_review/gpt-4o/no_eval/pr_222/comments.json
+++ /dev/null
@@ -1,132 +0,0 @@
-[
-  {
-    "topic": "Security",
-    "comment": "Hardcoding API keys in `config.json` can lead to security vulnerabilities.",
-    "confidence": "critical",
-    "reason": "Exposing API keys in the codebase can lead to unauthorized access.",
-    "solution": "Use environment variables to store API keys instead of hardcoding them.",
-    "actual_code": "\"api_key\": \"os.environ/AZURE_API_KEY\"",
-    "fixed_code": "\"api_key\": \"${AZURE_API_KEY}\"",
-    "file_name": "config.json",
-    "start_line": 13,
-    "end_line": 13,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 9
-  },
-  {
-    "topic": "Security",
-    "comment": "Database connection strings should not be constructed using string interpolation.",
-    "confidence": "critical",
-    "reason": "Using string interpolation for connection strings can expose the application to SQL injection attacks.",
-    "solution": "Use parameterized queries or a configuration management tool to handle sensitive information.",
-    "actual_code": "self.engine = create_engine(\n    f\"postgresql://{os.environ['POSTGRES_USER']}:{os.environ['POSTGRES_PASSWORD']}@{os.environ['POSTGRES_HOST']}:{os.environ['POSTGRES_PORT']}/{os.environ['POSTGRES_DB']}\",\n    pool_size=10,\n    max_overflow=20,\n)",
-    "fixed_code": "self.engine = create_engine(\n    'postgresql://{user}:{password}@{host}:{port}/{db}'.format(\n        user=os.environ['POSTGRES_USER'],\n        password=os.environ['POSTGRES_PASSWORD'],\n        host=os.environ['POSTGRES_HOST'],\n        port=os.environ['POSTGRES_PORT'],\n        db=os.environ['POSTGRES_DB']\n    ),\n    pool_size=10,\n    max_overflow=20,\n)",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 35,
-    "end_line": 39,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 9
-  },
-  {
-    "topic": "SQL Injection",
-    "comment": "Potential SQL injection vulnerability in the query construction.",
-    "confidence": "critical",
-    "reason": "Using f-strings to construct SQL queries can lead to SQL injection attacks if user input is not properly sanitized.",
-    "solution": "Use parameterized queries to avoid SQL injection vulnerabilities.",
-    "actual_code": "query = f\"\"\"\nSELECT \n    e.node_id,\n    e.text,\n    e.metadata,\n    1 - (e.embedding <=> %s::vector) as similarity\nFROM \n{self.table_name}e\nJOIN \n    function_abstractions fa ON e.node_id = fa.function_id::text\nJOIN \n    files f ON fa.file_id = f.file_id\nWHERE \n    f.repo_id = %s\nORDER BY \n    similarity DESC\nLIMIT \n    %s\n\"\"\"",
-    "fixed_code": "query = \"\"\"\nSELECT \n    e.node_id,\n    e.text,\n    e.metadata,\n    1 - (e.embedding <=> %s::vector) as similarity\nFROM \n    %s e\nJOIN \n    function_abstractions fa ON e.node_id = fa.function_id::text\nJOIN \n    files f ON fa.file_id = f.file_id\nWHERE \n    f.repo_id = %s\nORDER BY \n    similarity DESC\nLIMIT \n    %s\n\"\"\"",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 19,
-    "end_line": 37,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 9
-  },
-  {
-    "topic": "Exception Handling",
-    "comment": "Broad exception handling should be avoided.",
-    "confidence": "critical",
-    "reason": "Catching all exceptions can hide bugs and make debugging difficult.",
-    "solution": "Catch specific exceptions instead of using a broad except clause.",
-    "actual_code": "except Exception as e:",
-    "fixed_code": "except ImportError as e:\n    logger.error(f\"Failed to import module:{str(e)}\")\n    raise\nexcept ValueError as e:\n    logger.error(f\"Invalid value:{str(e)}\")\n    raise",
-    "file_name": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 28,
-    "end_line": 30,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Docker",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to Dockerfile, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_name": "Dockerfile",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Docker",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to docker-compose.yml, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "15",
-    "end_line": "15",
-    "side": "RIGHT",
-    "file_name": "docker-compose.yml",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Version Control",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to .gitignore, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "164",
-    "end_line": "164",
-    "side": "RIGHT",
-    "file_name": ".gitignore",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Database",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to db_setup/init.sql, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_name": "db_setup/init.sql",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_222/issues.json b/.experiments/code_review/gpt-4o/no_eval/pr_222/issues.json
deleted file mode 100644
index a46082c3..00000000
--- a/.experiments/code_review/gpt-4o/no_eval/pr_222/issues.json
+++ /dev/null
@@ -1,387 +0,0 @@
-[
-  {
-    "topic": "Documentation",
-    "comment": "The `install_tree_sitter_languages.sh` script lacks comments explaining the purpose of each step.",
-    "confidence": "moderate",
-    "reason": "Comments improve readability and maintainability, especially for complex scripts.",
-    "solution": "Add comments explaining the purpose of each step in the script.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "install_tree_sitter_languages.sh",
-    "start_line": 1,
-    "end_line": 47,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "The `install_tree_sitter_languages.sh` script does not handle errors during the installation process.",
-    "confidence": "important",
-    "reason": "Error handling ensures that the script fails gracefully and provides useful error messages.",
-    "solution": "Add error handling for each critical step in the script.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "install_tree_sitter_languages.sh",
-    "start_line": 1,
-    "end_line": 47,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Security",
-    "comment": "Hardcoding API keys in `config.json` can lead to security vulnerabilities.",
-    "confidence": "critical",
-    "reason": "Exposing API keys in the codebase can lead to unauthorized access.",
-    "solution": "Use environment variables to store API keys instead of hardcoding them.",
-    "actual_code": "\"api_key\": \"os.environ/AZURE_API_KEY\"",
-    "fixed_code": "\"api_key\": \"${AZURE_API_KEY}\"",
-    "file_name": "config.json",
-    "start_line": 13,
-    "end_line": 13,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 9
-  },
-  {
-    "topic": "Performance",
-    "comment": "The Dockerfile installs build dependencies and then removes them, which is good practice but can be optimized.",
-    "confidence": "moderate",
-    "reason": "Optimizing Dockerfile layers can reduce image size and build time.",
-    "solution": "Combine RUN commands to reduce the number of layers.",
-    "actual_code": "RUN apt-get update && apt-get install -y \\\n    build-essential \\\n    git \\\n    postgresql-server-dev-16\nRUN apt-get remove -y build-essential git postgresql-server-dev-16 \\\n    && apt-get autoremove -y \\\n    && rm -rf /var/lib/apt/lists/* /pgvector",
-    "fixed_code": "RUN apt-get update && apt-get install -y \\\n    build-essential \\\n    git \\\n    postgresql-server-dev-16 \\\n    && git clone https://github.com/pgvector/pgvector.git \\\n    && cd pgvector \\\n    && make \\\n    && make install \\\n    && apt-get remove -y build-essential git postgresql-server-dev-16 \\\n    && apt-get autoremove -y \\\n    && rm -rf /var/lib/apt/lists/* /pgvector",
-    "file_name": "Dockerfile-postgres",
-    "start_line": 4,
-    "end_line": 18,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 5
-  },
-  {
-    "topic": "Code Readability",
-    "comment": "The `chunk_code` function in `code_chunker.py` has nested functions and complex logic that can be refactored for better readability.",
-    "confidence": "important",
-    "reason": "Refactoring complex functions into smaller, well-named functions improves readability and maintainability.",
-    "solution": "Refactor the `chunk_code` function to extract nested functions into separate helper functions.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/retriever/code_chunker.py",
-    "start_line": 7,
-    "end_line": 62,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 6
-  },
-  {
-    "topic": "SQL Injection",
-    "comment": "Potential SQL injection vulnerability in the query construction.",
-    "confidence": "critical",
-    "reason": "Using f-strings to construct SQL queries can lead to SQL injection attacks if user input is not properly sanitized.",
-    "solution": "Use parameterized queries to avoid SQL injection vulnerabilities.",
-    "actual_code": "query = f\"\"\"\nSELECT \n    e.node_id,\n    e.text,\n    e.metadata,\n    1 - (e.embedding <=> %s::vector) as similarity\nFROM \n{self.table_name}e\nJOIN \n    function_abstractions fa ON e.node_id = fa.function_id::text\nJOIN \n    files f ON fa.file_id = f.file_id\nWHERE \n    f.repo_id = %s\nORDER BY \n    similarity DESC\nLIMIT \n    %s\n\"\"\"",
-    "fixed_code": "query = \"\"\"\nSELECT \n    e.node_id,\n    e.text,\n    e.metadata,\n    1 - (e.embedding <=> %s::vector) as similarity\nFROM \n    %s e\nJOIN \n    function_abstractions fa ON e.node_id = fa.function_id::text\nJOIN \n    files f ON fa.file_id = f.file_id\nWHERE \n    f.repo_id = %s\nORDER BY \n    similarity DESC\nLIMIT \n    %s\n\"\"\"",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 19,
-    "end_line": 37,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 9
-  },
-  {
-    "topic": "Code Readability",
-    "comment": "The normalization of the query embedding can be simplified for better readability.",
-    "confidence": "moderate",
-    "reason": "Simplifying code improves readability and maintainability.",
-    "solution": "Combine the normalization steps into a single line.",
-    "actual_code": "query_embedding_np = np.array(query_embedding)\nquery_embedding_normalized = query_embedding_np / np.linalg.norm(query_embedding_np)",
-    "fixed_code": "query_embedding_normalized = np.array(query_embedding) / np.linalg.norm(query_embedding)",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 15,
-    "end_line": 16,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 3
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "Lack of error handling in database operations.",
-    "confidence": "important",
-    "reason": "Database operations can fail; it's important to handle exceptions to avoid crashes.",
-    "solution": "Add try-except blocks to handle potential database errors.",
-    "actual_code": "",
-    "fixed_code": "try:\n    with self.get_client() as client:\n        with client.cursor() as cur:\n            cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))\n            results = cur.fetchall()\nexcept Exception as e:\n    # Handle exception (e.g., log the error, re-raise, etc.)\n    raise e",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 39,
-    "end_line": 42,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Type Annotations",
-    "comment": "Missing type annotations for method parameters and return types.",
-    "confidence": "moderate",
-    "reason": "Type annotations improve code readability and help with static analysis.",
-    "solution": "Add type annotations to the method parameters and return types.",
-    "actual_code": "def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:",
-    "fixed_code": "def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[Dict[str, Any]]:",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 13,
-    "end_line": 13,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Dictionary Default Values",
-    "comment": "Using `None` as a default return value for `get_feedback` method.",
-    "confidence": "low",
-    "reason": "Returning `None` can lead to potential `NoneType` errors if not handled properly.",
-    "solution": "Return an empty dictionary instead of `None`.",
-    "actual_code": "return self.feedback_store.get(code_id, None)",
-    "fixed_code": "return self.feedback_store.get(code_id,{})",
-    "file_name": "kaizen/retriever/feedback_system.py",
-    "start_line": 18,
-    "end_line": 18,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 2
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "Exception handling in `generate_abstraction` method is too generic.",
-    "confidence": "important",
-    "reason": "Catching all exceptions without specific handling can obscure the root cause of errors and make debugging difficult.",
-    "solution": "Catch specific exceptions and handle them appropriately.",
-    "actual_code": "except Exception as e:\n    raise e",
-    "fixed_code": "except SomeSpecificException as e:\n    logger.error(f\"Specific error occurred:{str(e)}\")\n    raise e",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 218,
-    "end_line": 219,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 7
-  },
-  {
-    "topic": "Security",
-    "comment": "Database connection strings should not be constructed using string interpolation.",
-    "confidence": "critical",
-    "reason": "Using string interpolation for connection strings can expose the application to SQL injection attacks.",
-    "solution": "Use parameterized queries or a configuration management tool to handle sensitive information.",
-    "actual_code": "self.engine = create_engine(\n    f\"postgresql://{os.environ['POSTGRES_USER']}:{os.environ['POSTGRES_PASSWORD']}@{os.environ['POSTGRES_HOST']}:{os.environ['POSTGRES_PORT']}/{os.environ['POSTGRES_DB']}\",\n    pool_size=10,\n    max_overflow=20,\n)",
-    "fixed_code": "self.engine = create_engine(\n    'postgresql://{user}:{password}@{host}:{port}/{db}'.format(\n        user=os.environ['POSTGRES_USER'],\n        password=os.environ['POSTGRES_PASSWORD'],\n        host=os.environ['POSTGRES_HOST'],\n        port=os.environ['POSTGRES_PORT'],\n        db=os.environ['POSTGRES_DB']\n    ),\n    pool_size=10,\n    max_overflow=20,\n)",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 35,
-    "end_line": 39,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 9
-  },
-  {
-    "topic": "Performance",
-    "comment": "Using `os.walk` and `ThreadPoolExecutor` for file parsing can be optimized.",
-    "confidence": "important",
-    "reason": "The current implementation may not efficiently utilize available CPU cores and can be improved for better performance.",
-    "solution": "Consider using asynchronous I/O operations or more efficient file traversal methods.",
-    "actual_code": "with ThreadPoolExecutor(max_workers=os.cpu_count()) as executor:\n    futures =[]\n    for root, _, files in os.walk(repo_path):\n        for file in files:\n            if file.endswith((\".py\", \".js\", \".ts\", \".rs\")):\n                file_path = os.path.join(root, file)\n                futures.append(executor.submit(self.parse_file, file_path))",
-    "fixed_code": "import asyncio\nfrom aiofiles import open as aio_open\n\nasync def parse_repository_async(self, repo_path: str):\n    self.total_usage = self.llm_provider.DEFAULT_USAGE\n    logger.info(f\"Starting repository setup for:{repo_path}\")\n    await self.parse_repository(repo_path)\n    self.store_function_relationships()\n    logger.info(\"Repository setup completed successfully\")\n\nasync def parse_file_async(self, file_path: str):\n    logger.debug(f\"Parsing file:{file_path}\")\n    try:\n        async with aio_open(file_path, \"r\", encoding=\"utf-8\") as file:\n            content = await file.read()\n        language = self.get_language_from_extension(file_path)\n        chunked_code = chunk_code(content, language)\n        for section, items in chunked_code.items():\n            if isinstance(items, dict):\n                for name, code_info in items.items():\n                    await self.process_code_block_async(code_info, file_path, section, name)\n            elif isinstance(items, list):\n                for i, code_info in enumerate(items):\n                    await self.process_code_block_async(code_info, file_path, section, f\"{section}_{i}\")\n        logger.debug(f\"Successfully parsed file:{file_path}\")\n    except Exception as e:\n        logger.error(f\"Error processing file{file_path}:{str(e)}\")\n        logger.error(traceback.format_exc())",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 71,
-    "end_line": 79,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 6
-  },
-  {
-    "topic": "Logging",
-    "comment": "Logging level for parsing files should be more granular.",
-    "confidence": "moderate",
-    "reason": "Using `logger.debug` for file parsing can help in better debugging without cluttering the log files.",
-    "solution": "Change logging level to `debug` for detailed logs during file parsing.",
-    "actual_code": "logger.info(f\"Parsing repository:{repo_path}\")",
-    "fixed_code": "logger.debug(f\"Parsing repository:{repo_path}\")",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 70,
-    "end_line": 70,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Code Readability",
-    "comment": "The `generate_abstraction` method is too long and complex.",
-    "confidence": "important",
-    "reason": "Long methods can be difficult to read and maintain. They should be broken down into smaller, more manageable functions.",
-    "solution": "Refactor the `generate_abstraction` method into smaller helper methods.",
-    "actual_code": "def generate_abstraction(\n    self, code_block: str, language: str, max_tokens: int = 300\n) -> str:\n    prompt = f\"\"\"Generate a concise yet comprehensive abstract description of the following{language}code block. \n    Include information about:\n    1. The purpose or functionality of the code\n    2. Input parameters and return values (if applicable)\n    3. Any important algorithms or data structures used\n    4. Key dependencies or external libraries used\n    5. Any notable design patterns or architectural choices\n    6. Potential edge cases or error handling\n\n    Code:\n    ```{language}\n{code_block}\n    ```\n    \"\"\"\n\n    estimated_prompt_tokens = len(tokenizer.encode(prompt))\n    adjusted_max_tokens = min(max(150, estimated_prompt_tokens), 1000)\n\n    try:\n        abstraction, usage = self.llm_provider.chat_completion(\n            prompt=\"\",\n            messages=[\n{\n                    \"role\": \"system\",\n                    \"content\": \"You are an expert programmer tasked with generating comprehensive and accurate abstractions of code snippets.\",\n},\n{\"role\": \"user\", \"content\": prompt},\n            ],\n            custom_model={\"max_tokens\": adjusted_max_tokens, \"model\": \"small\"},\n        )\n        return abstraction, usage\n\n    except Exception as e:\n        raise e",
-    "fixed_code": "def generate_abstraction(self, code_block: str, language: str, max_tokens: int = 300) -> str:\n    prompt = self._create_prompt(code_block, language)\n    estimated_prompt_tokens = len(tokenizer.encode(prompt))\n    adjusted_max_tokens = min(max(150, estimated_prompt_tokens), 1000)\n\n    try:\n        abstraction, usage = self._get_abstraction_from_llm(prompt, adjusted_max_tokens)\n        return abstraction, usage\n    except Exception as e:\n        raise e\n\n def _create_prompt(self, code_block: str, language: str) -> str:\n    return f\"\"\"Generate a concise yet comprehensive abstract description of the following{language}code block. \n    Include information about:\n    1. The purpose or functionality of the code\n    2. Input parameters and return values (if applicable)\n    3. Any important algorithms or data structures used\n    4. Key dependencies or external libraries used\n    5. Any notable design patterns or architectural choices\n    6. Potential edge cases or error handling\n\n    Code:\n    ```{language}\n{code_block}\n    ```\n    \"\"\"\n\n def _get_abstraction_from_llm(self, prompt: str, max_tokens: int) -> str:\n    return self.llm_provider.chat_completion(\n        prompt=\"\",\n        messages=[\n{\n                \"role\": \"system\",\n                \"content\": \"You are an expert programmer tasked with generating comprehensive and accurate abstractions of code snippets.\",\n},\n{\"role\": \"user\", \"content\": prompt},\n        ],\n        custom_model={\"max_tokens\": max_tokens, \"model\": \"small\"},\n    )",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 184,
-    "end_line": 219,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  },
-  {
-    "topic": "Logging Configuration",
-    "comment": "Logging configuration should be done in the main entry point of the application, not in the module.",
-    "confidence": "important",
-    "reason": "Configuring logging in a module can lead to unexpected behavior if the module is imported multiple times.",
-    "solution": "Move the logging configuration to the main entry point of the application.",
-    "actual_code": "logging.basicConfig(level=logging.INFO)",
-    "fixed_code": "",
-    "file_name": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 8,
-    "end_line": 8,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 6
-  },
-  {
-    "topic": "Exception Handling",
-    "comment": "Broad exception handling should be avoided.",
-    "confidence": "critical",
-    "reason": "Catching all exceptions can hide bugs and make debugging difficult.",
-    "solution": "Catch specific exceptions instead of using a broad except clause.",
-    "actual_code": "except Exception as e:",
-    "fixed_code": "except ImportError as e:\n    logger.error(f\"Failed to import module:{str(e)}\")\n    raise\nexcept ValueError as e:\n    logger.error(f\"Invalid value:{str(e)}\")\n    raise",
-    "file_name": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 28,
-    "end_line": 30,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Function Documentation",
-    "comment": "Public functions should have docstrings.",
-    "confidence": "moderate",
-    "reason": "Docstrings provide a convenient way of associating documentation with functions.",
-    "solution": "Add docstrings to public functions.",
-    "actual_code": "",
-    "fixed_code": "def load_language(language: str) -> Language:\n    \"\"\"\n    Load the specified language.\n    :param language: The name of the language to load.\n    :return: The loaded Language object.\n    \"\"\"",
-    "file_name": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 15,
-    "end_line": 15,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Code Duplication",
-    "comment": "Duplicate code found in test cases.",
-    "confidence": "important",
-    "reason": "Duplicate code can lead to maintenance issues and bugs.",
-    "solution": "Refactor the duplicate code into a helper function.",
-    "actual_code": "print_chunks(\"JavaScript\", chunk_code(javascript_code, \"javascript\"))",
-    "fixed_code": "def test_chunk(language, code):\n    print_chunks(language, chunk_code(code, language))\n\ntest_chunk(\"Python\", python_code)\ntest_chunk(\"JavaScript\", javascript_code)\ntest_chunk(\"React\", react_nextjs_code)",
-    "file_name": "tests/retriever/test_chunker.py",
-    "start_line": 98,
-    "end_line": 101,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  },
-  {
-    "topic": "Versioning",
-    "comment": "Version bump should be accompanied by a changelog update.",
-    "confidence": "moderate",
-    "reason": "A changelog helps track changes and improvements in the project.",
-    "solution": "Update the changelog to reflect the changes made in this version.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "pyproject.toml",
-    "start_line": 3,
-    "end_line": 3,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 3
-  },
-  {
-    "topic": "Dependency Management",
-    "comment": "New dependencies added without justification.",
-    "confidence": "important",
-    "reason": "Adding dependencies increases the attack surface and maintenance burden.",
-    "solution": "Provide justification for new dependencies in the pull request description.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "pyproject.toml",
-    "start_line": 27,
-    "end_line": 49,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 6
-  },
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Docker",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to Dockerfile, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_name": "Dockerfile",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Docker",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to docker-compose.yml, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "15",
-    "end_line": "15",
-    "side": "RIGHT",
-    "file_name": "docker-compose.yml",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Version Control",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to .gitignore, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "164",
-    "end_line": "164",
-    "side": "RIGHT",
-    "file_name": ".gitignore",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Database",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to db_setup/init.sql, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_name": "db_setup/init.sql",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_222/review.md b/.experiments/code_review/gpt-4o/no_eval/pr_222/review.md
deleted file mode 100644
index 5a796ee4..00000000
--- a/.experiments/code_review/gpt-4o/no_eval/pr_222/review.md
+++ /dev/null
@@ -1,560 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/222
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 26
-- Critical: 9
-- Important: 9
-- Minor: 7
-- Files Affected: 14
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Security (9 issues)</strong></summary>
-
-### 1. Hardcoding API keys in `config.json` can lead to security vulnerabilities.
-📁 **File:** `config.json:13`
-⚖️ **Severity:** 9/10
-🔍 **Description:** Exposing API keys in the codebase can lead to unauthorized access.
-💡 **Solution:** Use environment variables to store API keys instead of hardcoding them.
-
-**Current Code:**
-```python
-"api_key": "os.environ/AZURE_API_KEY"
-```
-
-**Suggested Code:**
-```python
-"api_key": "${AZURE_API_KEY}"
-```
-
-### 2. Potential SQL injection vulnerability in the query construction.
-📁 **File:** `kaizen/retriever/custom_vector_store.py:19`
-⚖️ **Severity:** 9/10
-🔍 **Description:** Using f-strings to construct SQL queries can lead to SQL injection attacks if user input is not properly sanitized.
-💡 **Solution:** Use parameterized queries to avoid SQL injection vulnerabilities.
-
-**Current Code:**
-```python
-query = f"""
-SELECT 
-    e.node_id,
-    e.text,
-    e.metadata,
-    1 - (e.embedding <=> %s::vector) as similarity
-FROM 
-{self.table_name}e
-JOIN 
-    function_abstractions fa ON e.node_id = fa.function_id::text
-JOIN 
-    files f ON fa.file_id = f.file_id
-WHERE 
-    f.repo_id = %s
-ORDER BY 
-    similarity DESC
-LIMIT 
-    %s
-"""
-```
-
-**Suggested Code:**
-```python
-query = """
-SELECT 
-    e.node_id,
-    e.text,
-    e.metadata,
-    1 - (e.embedding <=> %s::vector) as similarity
-FROM 
-    %s e
-JOIN 
-    function_abstractions fa ON e.node_id = fa.function_id::text
-JOIN 
-    files f ON fa.file_id = f.file_id
-WHERE 
-    f.repo_id = %s
-ORDER BY 
-    similarity DESC
-LIMIT 
-    %s
-"""
-```
-
-### 3. Database connection strings should not be constructed using string interpolation.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:35`
-⚖️ **Severity:** 9/10
-🔍 **Description:** Using string interpolation for connection strings can expose the application to SQL injection attacks.
-💡 **Solution:** Use parameterized queries or a configuration management tool to handle sensitive information.
-
-**Current Code:**
-```python
-self.engine = create_engine(
-    f"postgresql://{os.environ['POSTGRES_USER']}:{os.environ['POSTGRES_PASSWORD']}@{os.environ['POSTGRES_HOST']}:{os.environ['POSTGRES_PORT']}/{os.environ['POSTGRES_DB']}",
-    pool_size=10,
-    max_overflow=20,
-)
-```
-
-**Suggested Code:**
-```python
-self.engine = create_engine(
-    'postgresql://{user}:{password}@{host}:{port}/{db}'.format(
-        user=os.environ['POSTGRES_USER'],
-        password=os.environ['POSTGRES_PASSWORD'],
-        host=os.environ['POSTGRES_HOST'],
-        port=os.environ['POSTGRES_PORT'],
-        db=os.environ['POSTGRES_DB']
-    ),
-    pool_size=10,
-    max_overflow=20,
-)
-```
-
-### 4. Broad exception handling should be avoided.
-📁 **File:** `kaizen/retriever/tree_sitter_utils.py:28`
-⚖️ **Severity:** 8/10
-🔍 **Description:** Catching all exceptions can hide bugs and make debugging difficult.
-💡 **Solution:** Catch specific exceptions instead of using a broad except clause.
-
-**Current Code:**
-```python
-except Exception as e:
-```
-
-**Suggested Code:**
-```python
-except ImportError as e:
-    logger.error(f"Failed to import module:{str(e)}")
-    raise
-except ValueError as e:
-    logger.error(f"Invalid value:{str(e)}")
-    raise
-```
-
-### 5. Changes made to sensitive file
-📁 **File:** `config.json:4`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to config.json, which needs review
-💡 **Solution:** NA
-
-### 6. Changes made to sensitive file
-📁 **File:** `Dockerfile:4`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to Dockerfile, which needs review
-💡 **Solution:** NA
-
-### 7. Changes made to sensitive file
-📁 **File:** `docker-compose.yml:15`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to docker-compose.yml, which needs review
-💡 **Solution:** NA
-
-### 8. Changes made to sensitive file
-📁 **File:** `.gitignore:164`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to .gitignore, which needs review
-💡 **Solution:** NA
-
-### 9. Changes made to sensitive file
-📁 **File:** `db_setup/init.sql:1`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to db_setup/init.sql, which needs review
-💡 **Solution:** NA
-
-</details>
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Error Handling (9 issues)</strong></summary>
-
-### 1. The `install_tree_sitter_languages.sh` script does not handle errors during the installation process.
-📁 **File:** `install_tree_sitter_languages.sh:1`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Error handling ensures that the script fails gracefully and provides useful error messages.
-💡 **Solution:** Add error handling for each critical step in the script.
-
-### 2. The `chunk_code` function in `code_chunker.py` has nested functions and complex logic that can be refactored for better readability.
-📁 **File:** `kaizen/retriever/code_chunker.py:7`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Refactoring complex functions into smaller, well-named functions improves readability and maintainability.
-💡 **Solution:** Refactor the `chunk_code` function to extract nested functions into separate helper functions.
-
-### 3. Lack of error handling in database operations.
-📁 **File:** `kaizen/retriever/custom_vector_store.py:39`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Database operations can fail; it's important to handle exceptions to avoid crashes.
-💡 **Solution:** Add try-except blocks to handle potential database errors.
-
-**Current Code:**
-```python
-
-```
-
-**Suggested Code:**
-```python
-try:
-    with self.get_client() as client:
-        with client.cursor() as cur:
-            cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))
-            results = cur.fetchall()
-except Exception as e:
-    # Handle exception (e.g., log the error, re-raise, etc.)
-    raise e
-```
-
-### 4. Exception handling in `generate_abstraction` method is too generic.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:218`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Catching all exceptions without specific handling can obscure the root cause of errors and make debugging difficult.
-💡 **Solution:** Catch specific exceptions and handle them appropriately.
-
-**Current Code:**
-```python
-except Exception as e:
-    raise e
-```
-
-**Suggested Code:**
-```python
-except SomeSpecificException as e:
-    logger.error(f"Specific error occurred:{str(e)}")
-    raise e
-```
-
-### 5. Using `os.walk` and `ThreadPoolExecutor` for file parsing can be optimized.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:71`
-⚖️ **Severity:** 6/10
-🔍 **Description:** The current implementation may not efficiently utilize available CPU cores and can be improved for better performance.
-💡 **Solution:** Consider using asynchronous I/O operations or more efficient file traversal methods.
-
-**Current Code:**
-```python
-with ThreadPoolExecutor(max_workers=os.cpu_count()) as executor:
-    futures =[]
-    for root, _, files in os.walk(repo_path):
-        for file in files:
-            if file.endswith((".py", ".js", ".ts", ".rs")):
-                file_path = os.path.join(root, file)
-                futures.append(executor.submit(self.parse_file, file_path))
-```
-
-**Suggested Code:**
-```python
-import asyncio
-from aiofiles import open as aio_open
-
-async def parse_repository_async(self, repo_path: str):
-    self.total_usage = self.llm_provider.DEFAULT_USAGE
-    logger.info(f"Starting repository setup for:{repo_path}")
-    await self.parse_repository(repo_path)
-    self.store_function_relationships()
-    logger.info("Repository setup completed successfully")
-
-async def parse_file_async(self, file_path: str):
-    logger.debug(f"Parsing file:{file_path}")
-    try:
-        async with aio_open(file_path, "r", encoding="utf-8") as file:
-            content = await file.read()
-        language = self.get_language_from_extension(file_path)
-        chunked_code = chunk_code(content, language)
-        for section, items in chunked_code.items():
-            if isinstance(items, dict):
-                for name, code_info in items.items():
-                    await self.process_code_block_async(code_info, file_path, section, name)
-            elif isinstance(items, list):
-                for i, code_info in enumerate(items):
-                    await self.process_code_block_async(code_info, file_path, section, f"{section}_{i}")
-        logger.debug(f"Successfully parsed file:{file_path}")
-    except Exception as e:
-        logger.error(f"Error processing file{file_path}:{str(e)}")
-        logger.error(traceback.format_exc())
-```
-
-### 6. The `generate_abstraction` method is too long and complex.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:184`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Long methods can be difficult to read and maintain. They should be broken down into smaller, more manageable functions.
-💡 **Solution:** Refactor the `generate_abstraction` method into smaller helper methods.
-
-**Current Code:**
-```python
-def generate_abstraction(
-    self, code_block: str, language: str, max_tokens: int = 300
-) -> str:
-    prompt = f"""Generate a concise yet comprehensive abstract description of the following{language}code block. 
-    Include information about:
-    1. The purpose or functionality of the code
-    2. Input parameters and return values (if applicable)
-    3. Any important algorithms or data structures used
-    4. Key dependencies or external libraries used
-    5. Any notable design patterns or architectural choices
-    6. Potential edge cases or error handling
-
-    Code:
-    ```{language}
-{code_block}
-    ```
-    """
-
-    estimated_prompt_tokens = len(tokenizer.encode(prompt))
-    adjusted_max_tokens = min(max(150, estimated_prompt_tokens), 1000)
-
-    try:
-        abstraction, usage = self.llm_provider.chat_completion(
-            prompt="",
-            messages=[
-{
-                    "role": "system",
-                    "content": "You are an expert programmer tasked with generating comprehensive and accurate abstractions of code snippets.",
-},
-{"role": "user", "content": prompt},
-            ],
-            custom_model={"max_tokens": adjusted_max_tokens, "model": "small"},
-        )
-        return abstraction, usage
-
-    except Exception as e:
-        raise e
-```
-
-**Suggested Code:**
-```python
-def generate_abstraction(self, code_block: str, language: str, max_tokens: int = 300) -> str:
-    prompt = self._create_prompt(code_block, language)
-    estimated_prompt_tokens = len(tokenizer.encode(prompt))
-    adjusted_max_tokens = min(max(150, estimated_prompt_tokens), 1000)
-
-    try:
-        abstraction, usage = self._get_abstraction_from_llm(prompt, adjusted_max_tokens)
-        return abstraction, usage
-    except Exception as e:
-        raise e
-
- def _create_prompt(self, code_block: str, language: str) -> str:
-    return f"""Generate a concise yet comprehensive abstract description of the following{language}code block. 
-    Include information about:
-    1. The purpose or functionality of the code
-    2. Input parameters and return values (if applicable)
-    3. Any important algorithms or data structures used
-    4. Key dependencies or external libraries used
-    5. Any notable design patterns or architectural choices
-    6. Potential edge cases or error handling
-
-    Code:
-    ```{language}
-{code_block}
-    ```
-    """
-
- def _get_abstraction_from_llm(self, prompt: str, max_tokens: int) -> str:
-    return self.llm_provider.chat_completion(
-        prompt="",
-        messages=[
-{
-                "role": "system",
-                "content": "You are an expert programmer tasked with generating comprehensive and accurate abstractions of code snippets.",
-},
-{"role": "user", "content": prompt},
-        ],
-        custom_model={"max_tokens": max_tokens, "model": "small"},
-    )
-```
-
-### 7. Logging configuration should be done in the main entry point of the application, not in the module.
-📁 **File:** `kaizen/retriever/tree_sitter_utils.py:8`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Configuring logging in a module can lead to unexpected behavior if the module is imported multiple times.
-💡 **Solution:** Move the logging configuration to the main entry point of the application.
-
-**Current Code:**
-```python
-logging.basicConfig(level=logging.INFO)
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 8. Duplicate code found in test cases.
-📁 **File:** `tests/retriever/test_chunker.py:98`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Duplicate code can lead to maintenance issues and bugs.
-💡 **Solution:** Refactor the duplicate code into a helper function.
-
-**Current Code:**
-```python
-print_chunks("JavaScript", chunk_code(javascript_code, "javascript"))
-```
-
-**Suggested Code:**
-```python
-def test_chunk(language, code):
-    print_chunks(language, chunk_code(code, language))
-
-test_chunk("Python", python_code)
-test_chunk("JavaScript", javascript_code)
-test_chunk("React", react_nextjs_code)
-```
-
-### 9. New dependencies added without justification.
-📁 **File:** `pyproject.toml:27`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Adding dependencies increases the attack surface and maintenance burden.
-💡 **Solution:** Provide justification for new dependencies in the pull request description.
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (8 issues)</strong></summary>
-
-<details>
-<summary><strong>Documentation (7 issues)</strong></summary>
-
-### 1. The `install_tree_sitter_languages.sh` script lacks comments explaining the purpose of each step.
-📁 **File:** `install_tree_sitter_languages.sh:1`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Comments improve readability and maintainability, especially for complex scripts.
-💡 **Solution:** Add comments explaining the purpose of each step in the script.
-
-### 2. The Dockerfile installs build dependencies and then removes them, which is good practice but can be optimized.
-📁 **File:** `Dockerfile-postgres:4`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Optimizing Dockerfile layers can reduce image size and build time.
-💡 **Solution:** Combine RUN commands to reduce the number of layers.
-
-**Current Code:**
-```python
-RUN apt-get update && apt-get install -y \
-    build-essential \
-    git \
-    postgresql-server-dev-16
-RUN apt-get remove -y build-essential git postgresql-server-dev-16 \
-    && apt-get autoremove -y \
-    && rm -rf /var/lib/apt/lists/* /pgvector
-```
-
-**Suggested Code:**
-```python
-RUN apt-get update && apt-get install -y \
-    build-essential \
-    git \
-    postgresql-server-dev-16 \
-    && git clone https://github.com/pgvector/pgvector.git \
-    && cd pgvector \
-    && make \
-    && make install \
-    && apt-get remove -y build-essential git postgresql-server-dev-16 \
-    && apt-get autoremove -y \
-    && rm -rf /var/lib/apt/lists/* /pgvector
-```
-
-### 3. The normalization of the query embedding can be simplified for better readability.
-📁 **File:** `kaizen/retriever/custom_vector_store.py:15`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Simplifying code improves readability and maintainability.
-💡 **Solution:** Combine the normalization steps into a single line.
-
-**Current Code:**
-```python
-query_embedding_np = np.array(query_embedding)
-query_embedding_normalized = query_embedding_np / np.linalg.norm(query_embedding_np)
-```
-
-**Suggested Code:**
-```python
-query_embedding_normalized = np.array(query_embedding) / np.linalg.norm(query_embedding)
-```
-
-### 4. Missing type annotations for method parameters and return types.
-📁 **File:** `kaizen/retriever/custom_vector_store.py:13`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Type annotations improve code readability and help with static analysis.
-💡 **Solution:** Add type annotations to the method parameters and return types.
-
-**Current Code:**
-```python
-def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:
-```
-
-**Suggested Code:**
-```python
-def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[Dict[str, Any]]:
-```
-
-### 5. Logging level for parsing files should be more granular.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:70`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Using `logger.debug` for file parsing can help in better debugging without cluttering the log files.
-💡 **Solution:** Change logging level to `debug` for detailed logs during file parsing.
-
-**Current Code:**
-```python
-logger.info(f"Parsing repository:{repo_path}")
-```
-
-**Suggested Code:**
-```python
-logger.debug(f"Parsing repository:{repo_path}")
-```
-
-### 6. Public functions should have docstrings.
-📁 **File:** `kaizen/retriever/tree_sitter_utils.py:15`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Docstrings provide a convenient way of associating documentation with functions.
-💡 **Solution:** Add docstrings to public functions.
-
-**Current Code:**
-```python
-
-```
-
-**Suggested Code:**
-```python
-def load_language(language: str) -> Language:
-    """
-    Load the specified language.
-    :param language: The name of the language to load.
-    :return: The loaded Language object.
-    """
-```
-
-### 7. Version bump should be accompanied by a changelog update.
-📁 **File:** `pyproject.toml:3`
-⚖️ **Severity:** 3/10
-🔍 **Description:** A changelog helps track changes and improvements in the project.
-💡 **Solution:** Update the changelog to reflect the changes made in this version.
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-2024-05-13)
-{"prompt_tokens": 21545, "completion_tokens": 5255, "total_tokens": 26800}
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_335/comments.json b/.experiments/code_review/gpt-4o/no_eval/pr_335/comments.json
deleted file mode 100644
index 7ab211a4..00000000
--- a/.experiments/code_review/gpt-4o/no_eval/pr_335/comments.json
+++ /dev/null
@@ -1,32 +0,0 @@
-[
-  {
-    "topic": "Function Parameters",
-    "comment": "Removed `reeval_response` parameter from multiple functions. Ensure that this parameter is no longer needed and does not affect the functionality.",
-    "confidence": "critical",
-    "reason": "Removing a parameter can lead to missing functionality if the parameter was being used elsewhere in the code.",
-    "solution": "Verify that `reeval_response` is not required for the functions to operate correctly.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/pr_description.py",
-    "start_line": 43,
-    "end_line": 43,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Function Implementation",
-    "comment": "Changed the implementation of `_process_full_diff` to remove `reeval_response` logic. Ensure that the re-evaluation logic is no longer needed.",
-    "confidence": "critical",
-    "reason": "Removing logic can lead to missing functionality if the logic was essential.",
-    "solution": "Verify that the re-evaluation logic is not required for the function to operate correctly.",
-    "actual_code": "resp, usage = self.provider.chat_completion(prompt, user=user)\ndesc = parser.extract_code_from_markdown(resp)\nreturn desc",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/pr_description.py",
-    "start_line": 79,
-    "end_line": 83,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 8
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_335/issues.json b/.experiments/code_review/gpt-4o/no_eval/pr_335/issues.json
deleted file mode 100644
index 49da45e6..00000000
--- a/.experiments/code_review/gpt-4o/no_eval/pr_335/issues.json
+++ /dev/null
@@ -1,92 +0,0 @@
-[
-  {
-    "topic": "Imports",
-    "comment": "The import statement was changed to import from a different module. Ensure that the new module contains the required constants.",
-    "confidence": "important",
-    "reason": "Changing import paths can lead to runtime errors if the new module does not contain the expected constants.",
-    "solution": "Verify that `PR_DESCRIPTION_SYSTEM_PROMPT` exists in `pr_desc_prompts` and is correctly defined.",
-    "actual_code": "from kaizen.llms.prompts.pr_desc_prompts import (",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/pr_description.py",
-    "start_line": 8,
-    "end_line": 8,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 6
-  },
-  {
-    "topic": "Initialization",
-    "comment": "The system prompt was changed from `CODE_REVIEW_SYSTEM_PROMPT` to `PR_DESCRIPTION_SYSTEM_PROMPT`. Ensure this change aligns with the intended functionality.",
-    "confidence": "important",
-    "reason": "Changing the system prompt can alter the behavior of the LLMProvider, potentially affecting the output.",
-    "solution": "Confirm that `PR_DESCRIPTION_SYSTEM_PROMPT` is the correct prompt for the intended functionality.",
-    "actual_code": "self.provider.system_prompt = PR_DESCRIPTION_SYSTEM_PROMPT",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/pr_description.py",
-    "start_line": 28,
-    "end_line": 28,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 6
-  },
-  {
-    "topic": "Function Parameters",
-    "comment": "Removed `reeval_response` parameter from multiple functions. Ensure that this parameter is no longer needed and does not affect the functionality.",
-    "confidence": "critical",
-    "reason": "Removing a parameter can lead to missing functionality if the parameter was being used elsewhere in the code.",
-    "solution": "Verify that `reeval_response` is not required for the functions to operate correctly.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/pr_description.py",
-    "start_line": 43,
-    "end_line": 43,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Function Calls",
-    "comment": "Changed function calls to remove `reeval_response` parameter. Ensure that the new function signatures match the updated calls.",
-    "confidence": "important",
-    "reason": "Mismatch in function signatures can lead to runtime errors.",
-    "solution": "Ensure that `_process_full_diff` and `_process_files` functions do not require `reeval_response` parameter.",
-    "actual_code": "desc = self._process_full_diff(prompt, user)",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/pr_description.py",
-    "start_line": 52,
-    "end_line": 52,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 6
-  },
-  {
-    "topic": "Function Implementation",
-    "comment": "Changed the implementation of `_process_full_diff` to remove `reeval_response` logic. Ensure that the re-evaluation logic is no longer needed.",
-    "confidence": "critical",
-    "reason": "Removing logic can lead to missing functionality if the logic was essential.",
-    "solution": "Verify that the re-evaluation logic is not required for the function to operate correctly.",
-    "actual_code": "resp, usage = self.provider.chat_completion(prompt, user=user)\ndesc = parser.extract_code_from_markdown(resp)\nreturn desc",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/pr_description.py",
-    "start_line": 79,
-    "end_line": 83,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Prompt Definitions",
-    "comment": "Added new prompt definitions in `pr_desc_prompts.py`. Ensure that these new prompts are correctly formatted and used in the code.",
-    "confidence": "moderate",
-    "reason": "New prompt definitions need to be correctly formatted and integrated into the existing codebase.",
-    "solution": "Review the new prompt definitions for correctness and ensure they are used appropriately.",
-    "actual_code": "PR_DESCRIPTION_SYSTEM_PROMPT = \"\"\"\nAs a senior software developer reviewing code submissions, provide thorough, constructive feedback and suggestions for improvements. Consider best practices, error handling, performance, readability, and maintainability. Offer objective and respectful reviews that help developers enhance their skills and code quality. Use your expertise to provide comprehensive feedback without asking clarifying questions.\n\"\"\"",
-    "fixed_code": "",
-    "file_name": "kaizen/llms/prompts/pr_desc_prompts.py",
-    "start_line": 1,
-    "end_line": 3,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 4
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_335/review.md b/.experiments/code_review/gpt-4o/no_eval/pr_335/review.md
deleted file mode 100644
index e13a98d6..00000000
--- a/.experiments/code_review/gpt-4o/no_eval/pr_335/review.md
+++ /dev/null
@@ -1,151 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/335
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 6
-- Critical: 2
-- Important: 3
-- Minor: 1
-- Files Affected: 2
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Function Parameters (2 issues)</strong></summary>
-
-### 1. Removed `reeval_response` parameter from multiple functions. Ensure that this parameter is no longer needed and does not affect the functionality.
-📁 **File:** `kaizen/generator/pr_description.py:43`
-⚖️ **Severity:** 8/10
-🔍 **Description:** Removing a parameter can lead to missing functionality if the parameter was being used elsewhere in the code.
-💡 **Solution:** Verify that `reeval_response` is not required for the functions to operate correctly.
-
-### 2. Changed the implementation of `_process_full_diff` to remove `reeval_response` logic. Ensure that the re-evaluation logic is no longer needed.
-📁 **File:** `kaizen/generator/pr_description.py:79`
-⚖️ **Severity:** 8/10
-🔍 **Description:** Removing logic can lead to missing functionality if the logic was essential.
-💡 **Solution:** Verify that the re-evaluation logic is not required for the function to operate correctly.
-
-**Current Code:**
-```python
-resp, usage = self.provider.chat_completion(prompt, user=user)
-desc = parser.extract_code_from_markdown(resp)
-return desc
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Imports (3 issues)</strong></summary>
-
-### 1. The import statement was changed to import from a different module. Ensure that the new module contains the required constants.
-📁 **File:** `kaizen/generator/pr_description.py:8`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Changing import paths can lead to runtime errors if the new module does not contain the expected constants.
-💡 **Solution:** Verify that `PR_DESCRIPTION_SYSTEM_PROMPT` exists in `pr_desc_prompts` and is correctly defined.
-
-**Current Code:**
-```python
-from kaizen.llms.prompts.pr_desc_prompts import (
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 2. The system prompt was changed from `CODE_REVIEW_SYSTEM_PROMPT` to `PR_DESCRIPTION_SYSTEM_PROMPT`. Ensure this change aligns with the intended functionality.
-📁 **File:** `kaizen/generator/pr_description.py:28`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Changing the system prompt can alter the behavior of the LLMProvider, potentially affecting the output.
-💡 **Solution:** Confirm that `PR_DESCRIPTION_SYSTEM_PROMPT` is the correct prompt for the intended functionality.
-
-**Current Code:**
-```python
-self.provider.system_prompt = PR_DESCRIPTION_SYSTEM_PROMPT
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 3. Changed function calls to remove `reeval_response` parameter. Ensure that the new function signatures match the updated calls.
-📁 **File:** `kaizen/generator/pr_description.py:52`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Mismatch in function signatures can lead to runtime errors.
-💡 **Solution:** Ensure that `_process_full_diff` and `_process_files` functions do not require `reeval_response` parameter.
-
-**Current Code:**
-```python
-desc = self._process_full_diff(prompt, user)
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (1 issues)</strong></summary>
-
-<details>
-<summary><strong>Prompt Definitions (1 issues)</strong></summary>
-
-### 1. Added new prompt definitions in `pr_desc_prompts.py`. Ensure that these new prompts are correctly formatted and used in the code.
-📁 **File:** `kaizen/llms/prompts/pr_desc_prompts.py:1`
-⚖️ **Severity:** 4/10
-🔍 **Description:** New prompt definitions need to be correctly formatted and integrated into the existing codebase.
-💡 **Solution:** Review the new prompt definitions for correctness and ensure they are used appropriately.
-
-**Current Code:**
-```python
-PR_DESCRIPTION_SYSTEM_PROMPT = """
-As a senior software developer reviewing code submissions, provide thorough, constructive feedback and suggestions for improvements. Consider best practices, error handling, performance, readability, and maintainability. Offer objective and respectful reviews that help developers enhance their skills and code quality. Use your expertise to provide comprehensive feedback without asking clarifying questions.
-"""
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-2024-05-13)
-{"prompt_tokens": 6751, "completion_tokens": 1126, "total_tokens": 7877}
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_440/issues.json b/.experiments/code_review/gpt-4o/no_eval/pr_440/issues.json
deleted file mode 100644
index ccaaf465..00000000
--- a/.experiments/code_review/gpt-4o/no_eval/pr_440/issues.json
+++ /dev/null
@@ -1,47 +0,0 @@
-[
-  {
-    "topic": "Comment Clarity",
-    "comment": "The comment 'TODO: DONT PUSH DUPLICATE' is ambiguous and lacks context.",
-    "confidence": "moderate",
-    "reason": "Comments should provide clear guidance or context for future developers.",
-    "solution": "Provide a more descriptive comment explaining what needs to be done to avoid pushing duplicates.",
-    "actual_code": "# TODO: DONT PUSH DUPLICATE",
-    "fixed_code": "# TODO: Ensure that duplicate embeddings are not pushed to the database.",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 157,
-    "end_line": 157,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 3
-  },
-  {
-    "topic": "Code Removal Impact",
-    "comment": "The removal of the llama-index-llms-openai dependency might cause issues if any part of the codebase relies on it.",
-    "confidence": "important",
-    "reason": "Removing dependencies without ensuring they are not used elsewhere can lead to runtime errors.",
-    "solution": "Verify that no part of the codebase uses llama-index-llms-openai before removing it.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "pyproject.toml",
-    "start_line": 27,
-    "end_line": 28,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Hardcoded Paths",
-    "comment": "The path './github_app/' is hardcoded, which can cause issues in different environments.",
-    "confidence": "important",
-    "reason": "Hardcoded paths can lead to errors when the code is run in different environments or directories.",
-    "solution": "Use a configuration file or environment variables to manage paths.",
-    "actual_code": "analyzer.setup_repository(\"./github_app/\")",
-    "fixed_code": "analyzer.setup_repository(os.getenv('GITHUB_APP_PATH', './github_app/'))",
-    "file_name": "examples/ragify_codebase/main.py",
-    "start_line": 7,
-    "end_line": 7,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 6
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_440/review.md b/.experiments/code_review/gpt-4o/no_eval/pr_440/review.md
deleted file mode 100644
index ab3069eb..00000000
--- a/.experiments/code_review/gpt-4o/no_eval/pr_440/review.md
+++ /dev/null
@@ -1,92 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/440
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 3
-- Critical: 0
-- Important: 2
-- Minor: 1
-- Files Affected: 3
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Code Removal Impact (2 issues)</strong></summary>
-
-### 1. The removal of the llama-index-llms-openai dependency might cause issues if any part of the codebase relies on it.
-📁 **File:** `pyproject.toml:27`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Removing dependencies without ensuring they are not used elsewhere can lead to runtime errors.
-💡 **Solution:** Verify that no part of the codebase uses llama-index-llms-openai before removing it.
-
-### 2. The path './github_app/' is hardcoded, which can cause issues in different environments.
-📁 **File:** `examples/ragify_codebase/main.py:7`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Hardcoded paths can lead to errors when the code is run in different environments or directories.
-💡 **Solution:** Use a configuration file or environment variables to manage paths.
-
-**Current Code:**
-```python
-analyzer.setup_repository("./github_app/")
-```
-
-**Suggested Code:**
-```python
-analyzer.setup_repository(os.getenv('GITHUB_APP_PATH', './github_app/'))
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (1 issues)</strong></summary>
-
-<details>
-<summary><strong>Comment Clarity (1 issues)</strong></summary>
-
-### 1. The comment 'TODO: DONT PUSH DUPLICATE' is ambiguous and lacks context.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:157`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Comments should provide clear guidance or context for future developers.
-💡 **Solution:** Provide a more descriptive comment explaining what needs to be done to avoid pushing duplicates.
-
-**Current Code:**
-```python
-# TODO: DONT PUSH DUPLICATE
-```
-
-**Suggested Code:**
-```python
-# TODO: Ensure that duplicate embeddings are not pushed to the database.
-```
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-2024-05-13)
-{"prompt_tokens": 1364, "completion_tokens": 531, "total_tokens": 1895}
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_476/comments.json b/.experiments/code_review/gpt-4o/no_eval/pr_476/comments.json
deleted file mode 100644
index f9fab36c..00000000
--- a/.experiments/code_review/gpt-4o/no_eval/pr_476/comments.json
+++ /dev/null
@@ -1,31 +0,0 @@
-[
-  {
-    "topic": "Error Handling",
-    "comment": "Generic exception handling without logging specific error details.",
-    "confidence": "critical",
-    "reason": "Using a generic `except Exception` block without logging the specific error details can make debugging difficult.",
-    "solution": "Log the specific error message in the exception block.",
-    "actual_code": "except Exception:\n    print(\"Error\")",
-    "fixed_code": "except Exception as e:\n    print(f\"Error:{e}\")",
-    "file_name": "github_app/github_helper/pull_requests.py",
-    "start_line": 140,
-    "end_line": 141,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 9
-  },
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_476/issues.json b/.experiments/code_review/gpt-4o/no_eval/pr_476/issues.json
deleted file mode 100644
index e5c492a5..00000000
--- a/.experiments/code_review/gpt-4o/no_eval/pr_476/issues.json
+++ /dev/null
@@ -1,91 +0,0 @@
-[
-  {
-    "topic": "Error Handling",
-    "comment": "Generic exception handling without logging specific error details.",
-    "confidence": "critical",
-    "reason": "Using a generic `except Exception` block without logging the specific error details can make debugging difficult.",
-    "solution": "Log the specific error message in the exception block.",
-    "actual_code": "except Exception:\n    print(\"Error\")",
-    "fixed_code": "except Exception as e:\n    print(f\"Error:{e}\")",
-    "file_name": "github_app/github_helper/pull_requests.py",
-    "start_line": 140,
-    "end_line": 141,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 9
-  },
-  {
-    "topic": "Code Readability",
-    "comment": "Unnecessary print statements left in the code.",
-    "confidence": "important",
-    "reason": "Leaving print statements in production code can clutter the output and is generally not recommended.",
-    "solution": "Remove or replace print statements with proper logging.",
-    "actual_code": "print(\"diff: \", diff_text)\nprint(\"pr_files\", pr_files)",
-    "fixed_code": "",
-    "file_name": "examples/code_review/main.py",
-    "start_line": 21,
-    "end_line": 22,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 6
-  },
-  {
-    "topic": "Function Signature",
-    "comment": "Modified function signature without updating all references.",
-    "confidence": "important",
-    "reason": "Changing a function signature without updating all references can lead to runtime errors.",
-    "solution": "Ensure all references to `post_pull_request` are updated to include the new `tests` parameter.",
-    "actual_code": "def post_pull_request(url, data, installation_id, tests=None):",
-    "fixed_code": "",
-    "file_name": "github_app/github_helper/pull_requests.py",
-    "start_line": 107,
-    "end_line": 107,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 7
-  },
-  {
-    "topic": "Code Maintainability",
-    "comment": "Redundant code for sorting files.",
-    "confidence": "moderate",
-    "reason": "The custom sorting logic can be replaced with Python's built-in sorting functions for better readability and maintainability.",
-    "solution": "Use Python's `sorted` function with a key parameter.",
-    "actual_code": "sorted_files =[]\nfor file in files:\n    min_index = len(sorted_files)\n    file_name = file[\"filename\"]\n    for i, sorted_file in enumerate(sorted_files):\n        if file_name < sorted_file[\"filename\"]:\n            min_index = i\n            break\n    sorted_files.insert(min_index, file)\nreturn sorted_files",
-    "fixed_code": "sorted_files = sorted(files, key=lambda x: x[\"filename\"])\nreturn sorted_files",
-    "file_name": "github_app/github_helper/pull_requests.py",
-    "start_line": 185,
-    "end_line": 194,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 5
-  },
-  {
-    "topic": "Code Quality",
-    "comment": "Unnecessary variable assignment.",
-    "confidence": "low",
-    "reason": "Assigning `issues` in the loop is unnecessary and can be removed.",
-    "solution": "Remove the assignment of `issues` within the loop.",
-    "actual_code": "issues = review",
-    "fixed_code": "",
-    "file_name": "github_app/github_helper/pull_requests.py",
-    "start_line": 153,
-    "end_line": 153,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 3
-  },
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_476/review.md b/.experiments/code_review/gpt-4o/no_eval/pr_476/review.md
deleted file mode 100644
index 06252bed..00000000
--- a/.experiments/code_review/gpt-4o/no_eval/pr_476/review.md
+++ /dev/null
@@ -1,144 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/476
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 6
-- Critical: 2
-- Important: 2
-- Minor: 1
-- Files Affected: 3
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Error Handling (2 issues)</strong></summary>
-
-### 1. Generic exception handling without logging specific error details.
-📁 **File:** `github_app/github_helper/pull_requests.py:140`
-⚖️ **Severity:** 9/10
-🔍 **Description:** Using a generic `except Exception` block without logging the specific error details can make debugging difficult.
-💡 **Solution:** Log the specific error message in the exception block.
-
-**Current Code:**
-```python
-except Exception:
-    print("Error")
-```
-
-**Suggested Code:**
-```python
-except Exception as e:
-    print(f"Error:{e}")
-```
-
-### 2. Changes made to sensitive file
-📁 **File:** `config.json:1`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to config.json, which needs review
-💡 **Solution:** NA
-
-</details>
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Code Readability (2 issues)</strong></summary>
-
-### 1. Unnecessary print statements left in the code.
-📁 **File:** `examples/code_review/main.py:21`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Leaving print statements in production code can clutter the output and is generally not recommended.
-💡 **Solution:** Remove or replace print statements with proper logging.
-
-**Current Code:**
-```python
-print("diff: ", diff_text)
-print("pr_files", pr_files)
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 2. Modified function signature without updating all references.
-📁 **File:** `github_app/github_helper/pull_requests.py:107`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Changing a function signature without updating all references can lead to runtime errors.
-💡 **Solution:** Ensure all references to `post_pull_request` are updated to include the new `tests` parameter.
-
-**Current Code:**
-```python
-def post_pull_request(url, data, installation_id, tests=None):
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (2 issues)</strong></summary>
-
-<details>
-<summary><strong>Code Maintainability (1 issues)</strong></summary>
-
-### 1. Redundant code for sorting files.
-📁 **File:** `github_app/github_helper/pull_requests.py:185`
-⚖️ **Severity:** 5/10
-🔍 **Description:** The custom sorting logic can be replaced with Python's built-in sorting functions for better readability and maintainability.
-💡 **Solution:** Use Python's `sorted` function with a key parameter.
-
-**Current Code:**
-```python
-sorted_files =[]
-for file in files:
-    min_index = len(sorted_files)
-    file_name = file["filename"]
-    for i, sorted_file in enumerate(sorted_files):
-        if file_name < sorted_file["filename"]:
-            min_index = i
-            break
-    sorted_files.insert(min_index, file)
-return sorted_files
-```
-
-**Suggested Code:**
-```python
-sorted_files = sorted(files, key=lambda x: x["filename"])
-return sorted_files
-```
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-2024-05-13)
-{"prompt_tokens": 4007, "completion_tokens": 873, "total_tokens": 4880}
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_5/comments.json b/.experiments/code_review/gpt-4o/no_eval/pr_5/comments.json
deleted file mode 100644
index b67c7510..00000000
--- a/.experiments/code_review/gpt-4o/no_eval/pr_5/comments.json
+++ /dev/null
@@ -1,47 +0,0 @@
-[
-  {
-    "topic": "API Call Error Handling",
-    "comment": "The API call to 'completion' lacks a retry mechanism.",
-    "confidence": "critical",
-    "reason": "API calls can fail due to network issues or server errors, and without a retry mechanism, the function may fail unexpectedly.",
-    "solution": "Implement a retry mechanism with exponential backoff for the API call.",
-    "actual_code": "response = completion(\n    model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n)",
-    "fixed_code": "import time\n\nfor attempt in range(3):\n    try:\n        response = completion(\n            model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n        )\n        break\n    except Exception as e:\n        if attempt < 2:\n            time.sleep(2 ** attempt)\n        else:\n            raise e",
-    "file_name": "main.py",
-    "start_line": 66,
-    "end_line": 68,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 9
-  },
-  {
-    "topic": "Silent Failure",
-    "comment": "The exception handling for JSON decoding fails silently without logging.",
-    "confidence": "critical",
-    "reason": "Silent failures make it difficult to diagnose issues when they occur.",
-    "solution": "Add logging to capture the exception details.",
-    "actual_code": "except json.JSONDecodeError:\n    result ={",
-    "fixed_code": "except json.JSONDecodeError as e:\n    print(f\"Failed to parse content for applicant:{e}\")\n    result ={",
-    "file_name": "main.py",
-    "start_line": 82,
-    "end_line": 84,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Division by Zero",
-    "comment": "Potential division by zero when calculating total tokens.",
-    "confidence": "critical",
-    "reason": "If 'total_tokens' is zero, it will cause a division by zero error.",
-    "solution": "Add a check to ensure 'total_tokens' is not zero before performing the division.",
-    "actual_code": "total_tokens = total_input_tokens + total_output_tokens",
-    "fixed_code": "total_tokens = total_input_tokens + total_output_tokens\nif total_tokens == 0:\n    print(\"No tokens were used.\")\n    return",
-    "file_name": "main.py",
-    "start_line": 156,
-    "end_line": 158,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_5/issues.json b/.experiments/code_review/gpt-4o/no_eval/pr_5/issues.json
deleted file mode 100644
index 0847dab5..00000000
--- a/.experiments/code_review/gpt-4o/no_eval/pr_5/issues.json
+++ /dev/null
@@ -1,107 +0,0 @@
-[
-  {
-    "topic": "Unused Import",
-    "comment": "The import statement for the 'random' module is unnecessary.",
-    "confidence": "trivial",
-    "reason": "The 'random' module is imported but never used in the code.",
-    "solution": "Remove the import statement for 'random'.",
-    "actual_code": "import random  # Unused import",
-    "fixed_code": "",
-    "file_name": "main.py",
-    "start_line": 8,
-    "end_line": 8,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 1
-  },
-  {
-    "topic": "API Call Error Handling",
-    "comment": "The API call to 'completion' lacks a retry mechanism.",
-    "confidence": "critical",
-    "reason": "API calls can fail due to network issues or server errors, and without a retry mechanism, the function may fail unexpectedly.",
-    "solution": "Implement a retry mechanism with exponential backoff for the API call.",
-    "actual_code": "response = completion(\n    model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n)",
-    "fixed_code": "import time\n\nfor attempt in range(3):\n    try:\n        response = completion(\n            model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n        )\n        break\n    except Exception as e:\n        if attempt < 2:\n            time.sleep(2 ** attempt)\n        else:\n            raise e",
-    "file_name": "main.py",
-    "start_line": 66,
-    "end_line": 68,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 9
-  },
-  {
-    "topic": "Silent Failure",
-    "comment": "The exception handling for JSON decoding fails silently without logging.",
-    "confidence": "critical",
-    "reason": "Silent failures make it difficult to diagnose issues when they occur.",
-    "solution": "Add logging to capture the exception details.",
-    "actual_code": "except json.JSONDecodeError:\n    result ={",
-    "fixed_code": "except json.JSONDecodeError as e:\n    print(f\"Failed to parse content for applicant:{e}\")\n    result ={",
-    "file_name": "main.py",
-    "start_line": 82,
-    "end_line": 84,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Inefficient Progress Printing",
-    "comment": "The progress printing method is inefficient.",
-    "confidence": "important",
-    "reason": "Printing progress in this manner can be slow and resource-intensive.",
-    "solution": "Use a more efficient method for printing progress, such as updating the progress less frequently.",
-    "actual_code": "print(f\"\\rProgress:[{'=' * int(50 * progress):<50}]{progress:.0%}\", end=\"\", flush=True)",
-    "fixed_code": "if index % 10 == 0 or index == total - 1:\n    print(f\"\\rProgress:[{'=' * int(50 * progress):<50}]{progress:.0%}\", end=\"\", flush=True)",
-    "file_name": "main.py",
-    "start_line": 121,
-    "end_line": 122,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  },
-  {
-    "topic": "Redundant Code",
-    "comment": "The check for an empty DataFrame is redundant.",
-    "confidence": "moderate",
-    "reason": "The code already handles an empty DataFrame gracefully, so this check is unnecessary.",
-    "solution": "Remove the redundant check for an empty DataFrame.",
-    "actual_code": "if len(df) == 0:\n    return",
-    "fixed_code": "",
-    "file_name": "main.py",
-    "start_line": 142,
-    "end_line": 143,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 3
-  },
-  {
-    "topic": "Division by Zero",
-    "comment": "Potential division by zero when calculating total tokens.",
-    "confidence": "critical",
-    "reason": "If 'total_tokens' is zero, it will cause a division by zero error.",
-    "solution": "Add a check to ensure 'total_tokens' is not zero before performing the division.",
-    "actual_code": "total_tokens = total_input_tokens + total_output_tokens",
-    "fixed_code": "total_tokens = total_input_tokens + total_output_tokens\nif total_tokens == 0:\n    print(\"No tokens were used.\")\n    return",
-    "file_name": "main.py",
-    "start_line": 156,
-    "end_line": 158,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "File Not Found Handling",
-    "comment": "No error handling for file not found.",
-    "confidence": "important",
-    "reason": "If the specified file does not exist, the program will crash.",
-    "solution": "Add error handling to check if the file exists before processing.",
-    "actual_code": "main(input_file)",
-    "fixed_code": "if not os.path.isfile(input_file):\n    print(f\"File not found:{input_file}\")\n    return\nmain(input_file)",
-    "file_name": "main.py",
-    "start_line": 174,
-    "end_line": 175,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 6
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/gpt-4o/no_eval/pr_5/review.md b/.experiments/code_review/gpt-4o/no_eval/pr_5/review.md
deleted file mode 100644
index f36368e6..00000000
--- a/.experiments/code_review/gpt-4o/no_eval/pr_5/review.md
+++ /dev/null
@@ -1,182 +0,0 @@
-PR URL: https://github.com/sauravpanda/applicant-screening/pull/5
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 7
-- Critical: 3
-- Important: 2
-- Minor: 1
-- Files Affected: 1
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>API Call Error Handling (3 issues)</strong></summary>
-
-### 1. The API call to 'completion' lacks a retry mechanism.
-📁 **File:** `main.py:66`
-⚖️ **Severity:** 9/10
-🔍 **Description:** API calls can fail due to network issues or server errors, and without a retry mechanism, the function may fail unexpectedly.
-💡 **Solution:** Implement a retry mechanism with exponential backoff for the API call.
-
-**Current Code:**
-```python
-response = completion(
-    model=os.environ.get("model", "anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1"), messages=messages
-)
-```
-
-**Suggested Code:**
-```python
-import time
-
-for attempt in range(3):
-    try:
-        response = completion(
-            model=os.environ.get("model", "anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1"), messages=messages
-        )
-        break
-    except Exception as e:
-        if attempt < 2:
-            time.sleep(2 ** attempt)
-        else:
-            raise e
-```
-
-### 2. The exception handling for JSON decoding fails silently without logging.
-📁 **File:** `main.py:82`
-⚖️ **Severity:** 8/10
-🔍 **Description:** Silent failures make it difficult to diagnose issues when they occur.
-💡 **Solution:** Add logging to capture the exception details.
-
-**Current Code:**
-```python
-except json.JSONDecodeError:
-    result ={
-```
-
-**Suggested Code:**
-```python
-except json.JSONDecodeError as e:
-    print(f"Failed to parse content for applicant:{e}")
-    result ={
-```
-
-### 3. Potential division by zero when calculating total tokens.
-📁 **File:** `main.py:156`
-⚖️ **Severity:** 7/10
-🔍 **Description:** If 'total_tokens' is zero, it will cause a division by zero error.
-💡 **Solution:** Add a check to ensure 'total_tokens' is not zero before performing the division.
-
-**Current Code:**
-```python
-total_tokens = total_input_tokens + total_output_tokens
-```
-
-**Suggested Code:**
-```python
-total_tokens = total_input_tokens + total_output_tokens
-if total_tokens == 0:
-    print("No tokens were used.")
-    return
-```
-
-</details>
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Inefficient Progress Printing (2 issues)</strong></summary>
-
-### 1. The progress printing method is inefficient.
-📁 **File:** `main.py:121`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Printing progress in this manner can be slow and resource-intensive.
-💡 **Solution:** Use a more efficient method for printing progress, such as updating the progress less frequently.
-
-**Current Code:**
-```python
-print(f"\rProgress:[{'=' * int(50 * progress):<50}]{progress:.0%}", end="", flush=True)
-```
-
-**Suggested Code:**
-```python
-if index % 10 == 0 or index == total - 1:
-    print(f"\rProgress:[{'=' * int(50 * progress):<50}]{progress:.0%}", end="", flush=True)
-```
-
-### 2. No error handling for file not found.
-📁 **File:** `main.py:174`
-⚖️ **Severity:** 6/10
-🔍 **Description:** If the specified file does not exist, the program will crash.
-💡 **Solution:** Add error handling to check if the file exists before processing.
-
-**Current Code:**
-```python
-main(input_file)
-```
-
-**Suggested Code:**
-```python
-if not os.path.isfile(input_file):
-    print(f"File not found:{input_file}")
-    return
-main(input_file)
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (2 issues)</strong></summary>
-
-<details>
-<summary><strong>Redundant Code (1 issues)</strong></summary>
-
-### 1. The check for an empty DataFrame is redundant.
-📁 **File:** `main.py:142`
-⚖️ **Severity:** 3/10
-🔍 **Description:** The code already handles an empty DataFrame gracefully, so this check is unnecessary.
-💡 **Solution:** Remove the redundant check for an empty DataFrame.
-
-**Current Code:**
-```python
-if len(df) == 0:
-    return
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-2024-05-13)
-{"prompt_tokens": 6154, "completion_tokens": 1315, "total_tokens": 7469}
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_222/comments.json b/.experiments/code_review/haiku/no_eval/pr_222/comments.json
deleted file mode 100644
index 7edfffbe..00000000
--- a/.experiments/code_review/haiku/no_eval/pr_222/comments.json
+++ /dev/null
@@ -1,72 +0,0 @@
-[
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Docker",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to Dockerfile, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_name": "Dockerfile",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Docker",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to docker-compose.yml, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "15",
-    "end_line": "15",
-    "side": "RIGHT",
-    "file_name": "docker-compose.yml",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Version Control",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to .gitignore, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "164",
-    "end_line": "164",
-    "side": "RIGHT",
-    "file_name": ".gitignore",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Database",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to db_setup/init.sql, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_name": "db_setup/init.sql",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_222/issues.json b/.experiments/code_review/haiku/no_eval/pr_222/issues.json
deleted file mode 100644
index b3bb8f78..00000000
--- a/.experiments/code_review/haiku/no_eval/pr_222/issues.json
+++ /dev/null
@@ -1,312 +0,0 @@
-[
-  {
-    "topic": "Dockerfile",
-    "comment": "The Dockerfile should install system dependencies before installing Poetry.",
-    "confidence": "important",
-    "reason": "Installing system dependencies before Poetry ensures that the necessary build tools are available for the Poetry installation process.",
-    "solution": "Move the system dependency installation block before the Poetry installation step.",
-    "actual_code": "",
-    "fixed_code": "# Install system dependencies\nRUN apt-get update && apt-get install -y \\\n    git \\\n    build-essential \\\n    && rm -rf /var/lib/apt/lists/*\n\n# Install Poetry\nRUN pip install --no-cache-dir poetry",
-    "file_name": "Dockerfile",
-    "start_line": 7,
-    "end_line": 11,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Dockerfile",
-    "comment": "The Dockerfile should make the Tree-sitter language installation script executable before running it.",
-    "confidence": "important",
-    "reason": "Making the script executable ensures that it can be properly executed during the build process.",
-    "solution": "Add a step to make the script executable before running it.",
-    "actual_code": "",
-    "fixed_code": "# Make the installation script executable\nRUN chmod +x install_tree_sitter_languages.sh\n\n# Run the Tree-sitter language installation script\nRUN ./install_tree_sitter_languages.sh",
-    "file_name": "Dockerfile",
-    "start_line": 25,
-    "end_line": 29,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "config.json",
-    "comment": "The config.json file should use environment variables for sensitive information like API keys.",
-    "confidence": "important",
-    "reason": "Using environment variables instead of hardcoding sensitive information in the config file improves security and makes the configuration more flexible.",
-    "solution": "Replace the API key and API base values with environment variable references, e.g., `os.environ['AZURE_API_KEY']`.",
-    "actual_code": "\"api_key\": \"azure/text-embedding-small\",\n\"api_base\": \"azure/gpt-4o-mini\"",
-    "fixed_code": "\"api_key\": \"os.environ['AZURE_API_KEY']\",\n\"api_base\": \"os.environ['AZURE_API_BASE']\"",
-    "file_name": "config.json",
-    "start_line": 13,
-    "end_line": 14,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "docker-compose.yml",
-    "comment": "The docker-compose.yml file should use the same Postgres image as the Dockerfile-postgres file.",
-    "confidence": "important",
-    "reason": "Using the same Postgres image ensures consistency and reduces potential issues with different versions or configurations.",
-    "solution": "Replace the Postgres image in the docker-compose.yml file with the one used in the Dockerfile-postgres file.",
-    "actual_code": "image: postgres:16-bullseye",
-    "fixed_code": "build:\n  context: .\n  dockerfile: Dockerfile-postgres",
-    "file_name": "docker-compose.yml",
-    "start_line": 18,
-    "end_line": 26,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "examples/ragify_codebase/main.py",
-    "comment": "The main.py file should provide more context and examples for using the RepositoryAnalyzer.",
-    "confidence": "moderate",
-    "reason": "Adding more context and examples will help users understand how to effectively use the RepositoryAnalyzer in their own projects.",
-    "solution": "Expand the example code to include more comments and explanations, such as how to set up the repository, how to perform different types of queries, and how to handle the results.",
-    "actual_code": "",
-    "fixed_code": "# Initialize the analyzer\nanalyzer = RepositoryAnalyzer()\n\n# Set up the repository (do this when you first analyze a repo or when you want to update it)\nanalyzer.setup_repository(\"./github_app/\")\n\n# Perform queries (you can do this as many times as you want without calling setup_repository again)\nresults = analyzer.query(\"Find functions that handle authentication\")\nfor result in results:\n    print(f\"File:{result['file_path']}\")\n    print(f\"Abstraction:{result['abstraction']}\")\n    print(f\"result:\\n{result}\")\n    print(f\"Relevance Score:{result['relevance_score']}\")\n    print(\"---\")\n\n# If you make changes to the repository and want to update the analysis:\nanalyzer.setup_repository(\"/path/to/your/repo\")\n\n# Then you can query again with the updated data\nresults = analyzer.query(\"authentication\")",
-    "file_name": "examples/ragify_codebase/main.py",
-    "start_line": 1,
-    "end_line": 22,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 5
-  },
-  {
-    "topic": "Normalization of query embedding",
-    "comment": "The query embedding is normalized correctly by dividing it by its L2 norm. This ensures that the query embedding has a unit length, which is important for cosine similarity calculations.",
-    "confidence": "positive",
-    "reason": "Normalizing the query embedding is a common best practice in vector similarity search to ensure that the magnitude of the vector does not affect the similarity calculation.",
-    "solution": "The current implementation of normalizing the query embedding is appropriate and does not require any changes.",
-    "actual_code": "query_embedding_normalized = query_embedding_np / np.linalg.norm(query_embedding_np)",
-    "fixed_code": "",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 15,
-    "end_line": 16,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 5
-  },
-  {
-    "topic": "SQL query construction",
-    "comment": "The SQL query is well-constructed and includes the necessary filters and ordering to retrieve the top-k most similar results based on the cosine similarity.",
-    "confidence": "positive",
-    "reason": "The query includes a join between the `embeddings` table and the `function_abstractions` and `files` tables to filter the results by the repository ID. The `ORDER BY` and `LIMIT` clauses ensure that only the top-k most similar results are returned.",
-    "solution": "The current implementation of the SQL query is appropriate and does not require any changes.",
-    "actual_code": "```sql\nSELECT \n    e.node_id,\n    e.text,\n    e.metadata,\n    1 - (e.embedding <=> %s::vector) as similarity\nFROM \n{self.table_name}e\nJOIN \n    function_abstractions fa ON e.node_id = fa.function_id::text\nJOIN \n    files f ON fa.file_id = f.file_id\nWHERE \n    f.repo_id = %s\nORDER BY \n    similarity DESC\nLIMIT \n    %s\n```",
-    "fixed_code": "",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 19,
-    "end_line": 36,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 5
-  },
-  {
-    "topic": "Metadata handling",
-    "comment": "The code correctly handles the metadata column, converting it to a dictionary if it is not already in that format.",
-    "confidence": "positive",
-    "reason": "The `row[2]` value is checked to see if it is already a dictionary, and if not, it is converted to a dictionary using the `Json` class from `psycopg2.extras`.",
-    "solution": "The current implementation of metadata handling is appropriate and does not require any changes.",
-    "actual_code": "\"metadata\": row[2] if isinstance(row[2], dict) else Json(row[2]),",
-    "fixed_code": "",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 48,
-    "end_line": 48,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 4
-  },
-  {
-    "topic": "Feedback system implementation",
-    "comment": "The `AbstractionFeedback` class provides a simple and effective way to store and retrieve feedback for code abstractions.",
-    "confidence": "positive",
-    "reason": "The class uses a dictionary to store the feedback, with the code ID as the key and the feedback details as the value. The `add_feedback` and `get_feedback` methods provide a clear interface for managing the feedback.",
-    "solution": "The current implementation of the `AbstractionFeedback` class is appropriate and does not require any changes.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/retriever/feedback_system.py",
-    "start_line": 0,
-    "end_line": 0,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 4
-  },
-  {
-    "topic": "Unused Imports",
-    "comment": "The following imports are not used in the code and can be removed: `from llama_index.core.schema import TextNode`, `from concurrent.futures import ThreadPoolExecutor, as_completed`, `from llama_index.embeddings.litellm import LiteLLMEmbedding`, `from llama_index.core import QueryBundle`.",
-    "confidence": "moderate",
-    "reason": "Unused imports can clutter the codebase and make it harder to maintain.",
-    "solution": "Remove the unused imports to improve code readability and maintainability.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 7,
-    "end_line": 19,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 3
-  },
-  {
-    "topic": "Logging Configuration",
-    "comment": "The logging configuration is set up in the global scope, which can lead to issues if the module is imported in multiple places. Consider moving the logging setup to a function or class initialization to ensure it's only configured once.",
-    "confidence": "moderate",
-    "reason": "Global logging configuration can cause conflicts if the module is used in multiple places.",
-    "solution": "Move the logging setup to a function or class initialization to ensure it's only configured once.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 22,
-    "end_line": 26,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 4
-  },
-  {
-    "topic": "Tokenizer Initialization",
-    "comment": "The tokenizer is initialized in the global scope, which can lead to issues if the module is imported in multiple places. Consider moving the tokenizer initialization to a function or class initialization to ensure it's only initialized once.",
-    "confidence": "moderate",
-    "reason": "Global tokenizer initialization can cause conflicts if the module is used in multiple places.",
-    "solution": "Move the tokenizer initialization to a function or class initialization to ensure it's only initialized once.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 28,
-    "end_line": 29,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 4
-  },
-  {
-    "topic": "Unused Code",
-    "comment": "The commented-out code block starting from line 315 appears to be an unused query method. Consider removing this code if it's no longer needed.",
-    "confidence": "important",
-    "reason": "Unused code can make the codebase harder to maintain and understand.",
-    "solution": "Remove the commented-out code block if it's no longer needed.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 315,
-    "end_line": 335,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 5
-  },
-  {
-    "topic": "Potential Performance Optimization",
-    "comment": "The `query` method retrieves the function details from the database for each result node. Consider optimizing this by fetching all the required information in a single database query.",
-    "confidence": "important",
-    "reason": "Fetching data from the database for each result node can be inefficient, especially for larger result sets.",
-    "solution": "Modify the `query` method to fetch all the required information in a single database query to improve performance.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 337,
-    "end_line": 390,
-    "side": "LEFT",
-    "sentiment": "positive",
-    "severity_level": 6
-  },
-  {
-    "topic": "Dependencies",
-    "comment": "The project dependencies have been updated to use newer versions of some libraries, such as Python 3.9 and various tree-sitter language parsers.",
-    "confidence": "important",
-    "reason": "Keeping dependencies up-to-date is important for security, performance, and access to new features.",
-    "solution": "The changes look good and should help improve the project's overall maintainability.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "pyproject.toml",
-    "start_line": 3,
-    "end_line": 13,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 7
-  },
-  {
-    "topic": "Code Chunking",
-    "comment": "The new `chunk_code` function in `test_chunker.py` looks like a useful utility for testing the code chunking functionality.",
-    "confidence": "moderate",
-    "reason": "The function provides a clear way to test the code chunking behavior for different programming languages.",
-    "solution": "Consider adding more test cases to cover edge cases and ensure the chunking works as expected for a variety of code samples.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "tests/retriever/test_chunker.py",
-    "start_line": 1,
-    "end_line": 101,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 6
-  },
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Docker",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to Dockerfile, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_name": "Dockerfile",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Docker",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to docker-compose.yml, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "15",
-    "end_line": "15",
-    "side": "RIGHT",
-    "file_name": "docker-compose.yml",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Version Control",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to .gitignore, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "164",
-    "end_line": "164",
-    "side": "RIGHT",
-    "file_name": ".gitignore",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Database",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to db_setup/init.sql, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_name": "db_setup/init.sql",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_222/review.md b/.experiments/code_review/haiku/no_eval/pr_222/review.md
deleted file mode 100644
index de7eb626..00000000
--- a/.experiments/code_review/haiku/no_eval/pr_222/review.md
+++ /dev/null
@@ -1,247 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/222
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 21
-- Critical: 5
-- Important: 7
-- Minor: 5
-- Files Affected: 11
-## 🏆 Code Quality
-[██████████████████░░] 90% (Excellent)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Configuration (5 issues)</strong></summary>
-
-### 1. Changes made to sensitive file
-📁 **File:** `config.json:4`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to config.json, which needs review
-💡 **Solution:** NA
-
-### 2. Changes made to sensitive file
-📁 **File:** `Dockerfile:4`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to Dockerfile, which needs review
-💡 **Solution:** NA
-
-### 3. Changes made to sensitive file
-📁 **File:** `docker-compose.yml:15`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to docker-compose.yml, which needs review
-💡 **Solution:** NA
-
-### 4. Changes made to sensitive file
-📁 **File:** `.gitignore:164`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to .gitignore, which needs review
-💡 **Solution:** NA
-
-### 5. Changes made to sensitive file
-📁 **File:** `db_setup/init.sql:1`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to db_setup/init.sql, which needs review
-💡 **Solution:** NA
-
-</details>
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Dockerfile (7 issues)</strong></summary>
-
-### 1. The Dockerfile should install system dependencies before installing Poetry.
-📁 **File:** `Dockerfile:7`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Installing system dependencies before Poetry ensures that the necessary build tools are available for the Poetry installation process.
-💡 **Solution:** Move the system dependency installation block before the Poetry installation step.
-
-**Current Code:**
-```python
-
-```
-
-**Suggested Code:**
-```python
-# Install system dependencies
-RUN apt-get update && apt-get install -y \
-    git \
-    build-essential \
-    && rm -rf /var/lib/apt/lists/*
-
-# Install Poetry
-RUN pip install --no-cache-dir poetry
-```
-
-### 2. The Dockerfile should make the Tree-sitter language installation script executable before running it.
-📁 **File:** `Dockerfile:25`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Making the script executable ensures that it can be properly executed during the build process.
-💡 **Solution:** Add a step to make the script executable before running it.
-
-**Current Code:**
-```python
-
-```
-
-**Suggested Code:**
-```python
-# Make the installation script executable
-RUN chmod +x install_tree_sitter_languages.sh
-
-# Run the Tree-sitter language installation script
-RUN ./install_tree_sitter_languages.sh
-```
-
-### 3. The config.json file should use environment variables for sensitive information like API keys.
-📁 **File:** `config.json:13`
-⚖️ **Severity:** 8/10
-🔍 **Description:** Using environment variables instead of hardcoding sensitive information in the config file improves security and makes the configuration more flexible.
-💡 **Solution:** Replace the API key and API base values with environment variable references, e.g., `os.environ['AZURE_API_KEY']`.
-
-**Current Code:**
-```python
-"api_key": "azure/text-embedding-small",
-"api_base": "azure/gpt-4o-mini"
-```
-
-**Suggested Code:**
-```python
-"api_key": "os.environ['AZURE_API_KEY']",
-"api_base": "os.environ['AZURE_API_BASE']"
-```
-
-### 4. The docker-compose.yml file should use the same Postgres image as the Dockerfile-postgres file.
-📁 **File:** `docker-compose.yml:18`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Using the same Postgres image ensures consistency and reduces potential issues with different versions or configurations.
-💡 **Solution:** Replace the Postgres image in the docker-compose.yml file with the one used in the Dockerfile-postgres file.
-
-**Current Code:**
-```python
-image: postgres:16-bullseye
-```
-
-**Suggested Code:**
-```python
-build:
-  context: .
-  dockerfile: Dockerfile-postgres
-```
-
-### 5. The commented-out code block starting from line 315 appears to be an unused query method. Consider removing this code if it's no longer needed.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:315`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Unused code can make the codebase harder to maintain and understand.
-💡 **Solution:** Remove the commented-out code block if it's no longer needed.
-
-### 6. The `query` method retrieves the function details from the database for each result node. Consider optimizing this by fetching all the required information in a single database query.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:337`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Fetching data from the database for each result node can be inefficient, especially for larger result sets.
-💡 **Solution:** Modify the `query` method to fetch all the required information in a single database query to improve performance.
-
-### 7. The project dependencies have been updated to use newer versions of some libraries, such as Python 3.9 and various tree-sitter language parsers.
-📁 **File:** `pyproject.toml:3`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Keeping dependencies up-to-date is important for security, performance, and access to new features.
-💡 **Solution:** The changes look good and should help improve the project's overall maintainability.
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (5 issues)</strong></summary>
-
-<details>
-<summary><strong>examples/ragify_codebase/main.py (5 issues)</strong></summary>
-
-### 1. The main.py file should provide more context and examples for using the RepositoryAnalyzer.
-📁 **File:** `examples/ragify_codebase/main.py:1`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Adding more context and examples will help users understand how to effectively use the RepositoryAnalyzer in their own projects.
-💡 **Solution:** Expand the example code to include more comments and explanations, such as how to set up the repository, how to perform different types of queries, and how to handle the results.
-
-**Current Code:**
-```python
-
-```
-
-**Suggested Code:**
-```python
-# Initialize the analyzer
-analyzer = RepositoryAnalyzer()
-
-# Set up the repository (do this when you first analyze a repo or when you want to update it)
-analyzer.setup_repository("./github_app/")
-
-# Perform queries (you can do this as many times as you want without calling setup_repository again)
-results = analyzer.query("Find functions that handle authentication")
-for result in results:
-    print(f"File:{result['file_path']}")
-    print(f"Abstraction:{result['abstraction']}")
-    print(f"result:\n{result}")
-    print(f"Relevance Score:{result['relevance_score']}")
-    print("---")
-
-# If you make changes to the repository and want to update the analysis:
-analyzer.setup_repository("/path/to/your/repo")
-
-# Then you can query again with the updated data
-results = analyzer.query("authentication")
-```
-
-### 2. The following imports are not used in the code and can be removed: `from llama_index.core.schema import TextNode`, `from concurrent.futures import ThreadPoolExecutor, as_completed`, `from llama_index.embeddings.litellm import LiteLLMEmbedding`, `from llama_index.core import QueryBundle`.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:7`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Unused imports can clutter the codebase and make it harder to maintain.
-💡 **Solution:** Remove the unused imports to improve code readability and maintainability.
-
-### 3. The logging configuration is set up in the global scope, which can lead to issues if the module is imported in multiple places. Consider moving the logging setup to a function or class initialization to ensure it's only configured once.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:22`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Global logging configuration can cause conflicts if the module is used in multiple places.
-💡 **Solution:** Move the logging setup to a function or class initialization to ensure it's only configured once.
-
-### 4. The tokenizer is initialized in the global scope, which can lead to issues if the module is imported in multiple places. Consider moving the tokenizer initialization to a function or class initialization to ensure it's only initialized once.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:28`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Global tokenizer initialization can cause conflicts if the module is used in multiple places.
-💡 **Solution:** Move the tokenizer initialization to a function or class initialization to ensure it's only initialized once.
-
-### 5. The new `chunk_code` function in `test_chunker.py` looks like a useful utility for testing the code chunking functionality.
-📁 **File:** `tests/retriever/test_chunker.py:1`
-⚖️ **Severity:** 6/10
-🔍 **Description:** The function provides a clear way to test the code chunking behavior for different programming languages.
-💡 **Solution:** Consider adding more test cases to cover edge cases and ensure the chunking works as expected for a variety of code samples.
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (anthropic.claude-3-haiku-20240307-v1:0)
-{"prompt_tokens": 24844, "completion_tokens": 4069, "total_tokens": 28913}
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_252/issues.json b/.experiments/code_review/haiku/no_eval/pr_252/issues.json
deleted file mode 100644
index 38344bd0..00000000
--- a/.experiments/code_review/haiku/no_eval/pr_252/issues.json
+++ /dev/null
@@ -1,32 +0,0 @@
-[
-  {
-    "topic": "Linkedin Post Generation",
-    "comment": "The LinkedIn post generation code is not formatted correctly.",
-    "confidence": "moderate",
-    "reason": "The LinkedIn post generation code is spread across multiple lines, making it less readable and maintainable.",
-    "solution": "Condense the LinkedIn post generation code into a single line, similar to the Twitter post generation.",
-    "actual_code": "linkedin_post = work_summary_generator.generate_linkedin_post(\n    summary, user=\"oss_example\"\n)",
-    "fixed_code": "linkedin_post = work_summary_generator.generate_linkedin_post(summary, user=\"oss_example\")",
-    "file_name": "examples/work_summarizer/main.py",
-    "start_line": 60,
-    "end_line": 62,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 3
-  },
-  {
-    "topic": "Prompt Formatting",
-    "comment": "The TWITTER_POST_PROMPT and LINKEDIN_POST_PROMPT could be improved for better readability.",
-    "confidence": "moderate",
-    "reason": "The prompts are currently formatted as a single long string, making them difficult to read and maintain.",
-    "solution": "Consider breaking the prompts into multiple lines, using string formatting, and adding comments to explain the different sections.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/llms/prompts/work_summary_prompts.py",
-    "start_line": 44,
-    "end_line": 65,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 4
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_252/review.md b/.experiments/code_review/haiku/no_eval/pr_252/review.md
deleted file mode 100644
index e28ec0f5..00000000
--- a/.experiments/code_review/haiku/no_eval/pr_252/review.md
+++ /dev/null
@@ -1,70 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/252
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 2
-- Critical: 0
-- Important: 0
-- Minor: 2
-- Files Affected: 2
-## 🏆 Code Quality
-[██████████████████░░] 90% (Excellent)
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (2 issues)</strong></summary>
-
-<details>
-<summary><strong>Linkedin Post Generation (2 issues)</strong></summary>
-
-### 1. The LinkedIn post generation code is not formatted correctly.
-📁 **File:** `examples/work_summarizer/main.py:60`
-⚖️ **Severity:** 3/10
-🔍 **Description:** The LinkedIn post generation code is spread across multiple lines, making it less readable and maintainable.
-💡 **Solution:** Condense the LinkedIn post generation code into a single line, similar to the Twitter post generation.
-
-**Current Code:**
-```python
-linkedin_post = work_summary_generator.generate_linkedin_post(
-    summary, user="oss_example"
-)
-```
-
-**Suggested Code:**
-```python
-linkedin_post = work_summary_generator.generate_linkedin_post(summary, user="oss_example")
-```
-
-### 2. The TWITTER_POST_PROMPT and LINKEDIN_POST_PROMPT could be improved for better readability.
-📁 **File:** `kaizen/llms/prompts/work_summary_prompts.py:44`
-⚖️ **Severity:** 4/10
-🔍 **Description:** The prompts are currently formatted as a single long string, making them difficult to read and maintain.
-💡 **Solution:** Consider breaking the prompts into multiple lines, using string formatting, and adding comments to explain the different sections.
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (anthropic.claude-3-haiku-20240307-v1:0)
-{"prompt_tokens": 4436, "completion_tokens": 465, "total_tokens": 4901}
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_335/issues.json b/.experiments/code_review/haiku/no_eval/pr_335/issues.json
deleted file mode 100644
index d941a81f..00000000
--- a/.experiments/code_review/haiku/no_eval/pr_335/issues.json
+++ /dev/null
@@ -1,62 +0,0 @@
-[
-  {
-    "topic": "Prompt Changes",
-    "comment": "The `PR_DESCRIPTION_PROMPT` and `PR_FILE_DESCRIPTION_PROMPT` have been updated to use a more concise and structured format for the output.",
-    "confidence": "important",
-    "reason": "The new prompts provide a clearer and more organized structure for the generated pull request description, making it easier to read and understand.",
-    "solution": "The changes look good and should improve the overall quality and readability of the generated pull request descriptions.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/llms/prompts/pr_desc_prompts.py",
-    "start_line": 1,
-    "end_line": 92,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 7
-  },
-  {
-    "topic": "Removal of `PR_DESC_EVALUATION_PROMPT`",
-    "comment": "The `PR_DESC_EVALUATION_PROMPT` has been removed from the `code_review_prompts.py` file.",
-    "confidence": "moderate",
-    "reason": "The `PR_DESC_EVALUATION_PROMPT` was previously used for re-evaluating the generated pull request description, but it has been removed in the current changes.",
-    "solution": "Ensure that the functionality for re-evaluating the pull request description is still maintained, either through a different mechanism or by incorporating the evaluation logic directly into the main description generation process.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/llms/prompts/code_review_prompts.py",
-    "start_line": 190,
-    "end_line": 201,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  },
-  {
-    "topic": "Removal of `reeval_response` parameter",
-    "comment": "The `reeval_response` parameter has been removed from several method calls in the `pr_description.py` file.",
-    "confidence": "moderate",
-    "reason": "The `reeval_response` parameter was previously used for re-evaluating the generated pull request description, but it has been removed in the current changes.",
-    "solution": "Ensure that the functionality for re-evaluating the pull request description is still maintained, either through a different mechanism or by incorporating the evaluation logic directly into the main description generation process.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/pr_description.py",
-    "start_line": 43,
-    "end_line": 96,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  },
-  {
-    "topic": "Removal of `CODE_REVIEW_SYSTEM_PROMPT`",
-    "comment": "The `CODE_REVIEW_SYSTEM_PROMPT` has been removed and replaced with `PR_DESCRIPTION_SYSTEM_PROMPT` in the `pr_description.py` file.",
-    "confidence": "important",
-    "reason": "The `CODE_REVIEW_SYSTEM_PROMPT` was previously used as the system prompt for the code review process, but it has been replaced with a new prompt specifically for generating pull request descriptions.",
-    "solution": "The changes look good and should help align the system prompt with the updated pull request description generation process.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/pr_description.py",
-    "start_line": 29,
-    "end_line": 29,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 7
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_335/review.md b/.experiments/code_review/haiku/no_eval/pr_335/review.md
deleted file mode 100644
index c4f147c7..00000000
--- a/.experiments/code_review/haiku/no_eval/pr_335/review.md
+++ /dev/null
@@ -1,78 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/335
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 4
-- Critical: 0
-- Important: 2
-- Minor: 2
-- Files Affected: 3
-## 🏆 Code Quality
-[██████████████████░░] 90% (Excellent)
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Prompt Changes (2 issues)</strong></summary>
-
-### 1. The `PR_DESCRIPTION_PROMPT` and `PR_FILE_DESCRIPTION_PROMPT` have been updated to use a more concise and structured format for the output.
-📁 **File:** `kaizen/llms/prompts/pr_desc_prompts.py:1`
-⚖️ **Severity:** 7/10
-🔍 **Description:** The new prompts provide a clearer and more organized structure for the generated pull request description, making it easier to read and understand.
-💡 **Solution:** The changes look good and should improve the overall quality and readability of the generated pull request descriptions.
-
-### 2. The `CODE_REVIEW_SYSTEM_PROMPT` has been removed and replaced with `PR_DESCRIPTION_SYSTEM_PROMPT` in the `pr_description.py` file.
-📁 **File:** `kaizen/generator/pr_description.py:29`
-⚖️ **Severity:** 7/10
-🔍 **Description:** The `CODE_REVIEW_SYSTEM_PROMPT` was previously used as the system prompt for the code review process, but it has been replaced with a new prompt specifically for generating pull request descriptions.
-💡 **Solution:** The changes look good and should help align the system prompt with the updated pull request description generation process.
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (2 issues)</strong></summary>
-
-<details>
-<summary><strong>Removal of `PR_DESC_EVALUATION_PROMPT` (2 issues)</strong></summary>
-
-### 1. The `PR_DESC_EVALUATION_PROMPT` has been removed from the `code_review_prompts.py` file.
-📁 **File:** `kaizen/llms/prompts/code_review_prompts.py:190`
-⚖️ **Severity:** 5/10
-🔍 **Description:** The `PR_DESC_EVALUATION_PROMPT` was previously used for re-evaluating the generated pull request description, but it has been removed in the current changes.
-💡 **Solution:** Ensure that the functionality for re-evaluating the pull request description is still maintained, either through a different mechanism or by incorporating the evaluation logic directly into the main description generation process.
-
-### 2. The `reeval_response` parameter has been removed from several method calls in the `pr_description.py` file.
-📁 **File:** `kaizen/generator/pr_description.py:43`
-⚖️ **Severity:** 5/10
-🔍 **Description:** The `reeval_response` parameter was previously used for re-evaluating the generated pull request description, but it has been removed in the current changes.
-💡 **Solution:** Ensure that the functionality for re-evaluating the pull request description is still maintained, either through a different mechanism or by incorporating the evaluation logic directly into the main description generation process.
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (anthropic.claude-3-haiku-20240307-v1:0)
-{"prompt_tokens": 7614, "completion_tokens": 945, "total_tokens": 8559}
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_400/comments.json b/.experiments/code_review/haiku/no_eval/pr_400/comments.json
deleted file mode 100644
index 33197e40..00000000
--- a/.experiments/code_review/haiku/no_eval/pr_400/comments.json
+++ /dev/null
@@ -1,16 +0,0 @@
-[
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "11",
-    "end_line": "11",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_400/issues.json b/.experiments/code_review/haiku/no_eval/pr_400/issues.json
deleted file mode 100644
index 03fa0f89..00000000
--- a/.experiments/code_review/haiku/no_eval/pr_400/issues.json
+++ /dev/null
@@ -1,406 +0,0 @@
-[
-  {
-    "topic": "Boundary Conditions",
-    "comment": "The test cases for boundary conditions (very long descriptions) look good. The execution time is also printed, which is a nice addition.",
-    "confidence": "important",
-    "reason": "Handling large inputs is an important aspect of the function's robustness.",
-    "solution": "No changes needed.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
-    "start_line": 45,
-    "end_line": 61,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 8
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "The error handling tests cover various invalid input scenarios, which is good.",
-    "confidence": "important",
-    "reason": "Proper error handling is crucial for the function's reliability.",
-    "solution": "No changes needed.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
-    "start_line": 31,
-    "end_line": 43,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 8
-  },
-  {
-    "topic": "Collapsible Template",
-    "comment": "The collapsible template for the original description has been improved to include newlines for better readability.",
-    "confidence": "moderate",
-    "reason": "The previous template did not have proper newline formatting.",
-    "solution": "No changes needed.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
-    "start_line": 5,
-    "end_line": 6,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 5
-  },
-  {
-    "topic": "Readability",
-    "comment": "The updated PR_COLLAPSIBLE_TEMPLATE is more readable and follows better formatting.",
-    "confidence": "high",
-    "reason": "The previous template used a mix of string concatenation and formatting, which made it less readable and maintainable. The new template uses a multi-line string with proper indentation and formatting.",
-    "solution": "Keep the updated PR_COLLAPSIBLE_TEMPLATE, as it improves the overall readability of the code.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 4,
-    "end_line": 16,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 3
-  },
-  {
-    "topic": "Test Coverage",
-    "comment": "The new test cases cover a wider range of scenarios, including empty topics, single topic with single review, and multiple topics with multiple reviews.",
-    "confidence": "high",
-    "reason": "The additional test cases ensure the `create_pr_review_text` function handles different input scenarios correctly.",
-    "solution": "Keep the new test cases, as they improve the overall test coverage and ensure the function's robustness.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 74,
-    "end_line": 270,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 4
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "The new test cases cover scenarios where the review data is missing required fields, such as 'solution', 'reason', 'confidence', and 'severity_level'.",
-    "confidence": "high",
-    "reason": "Handling missing fields is important to ensure the function can gracefully handle incomplete review data.",
-    "solution": "Keep the new test cases, as they ensure the function can handle missing fields in the review data without crashing or producing unexpected output.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 142,
-    "end_line": 234,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 4
-  },
-  {
-    "topic": "Maintainability",
-    "comment": "The new test case for 'reviews_with_missing_comment' ensures the function can handle missing 'comment' field in the review data.",
-    "confidence": "high",
-    "reason": "Handling missing fields, such as 'comment', is important for the function's robustness and maintainability.",
-    "solution": "Keep the new test case, as it ensures the function can handle missing 'comment' field without crashing or producing unexpected output.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 238,
-    "end_line": 269,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 4
-  },
-  {
-    "topic": "Efficiency",
-    "comment": "The new test case for 'empty_list_in_topics' ensures the function can handle an empty list of reviews for a given topic.",
-    "confidence": "high",
-    "reason": "Handling empty lists of reviews is important for the function's efficiency and edge case handling.",
-    "solution": "Keep the new test case, as it ensures the function can handle empty lists of reviews without producing unexpected output.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 271,
-    "end_line": 276,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 3
-  },
-  {
-    "topic": "Unused imports",
-    "comment": "The imports `mock` and `pytest` are not used in the updated tests.",
-    "confidence": "moderate",
-    "reason": "Unused imports can make the code harder to read and maintain.",
-    "solution": "Remove the unused imports.",
-    "actual_code": "import os\nimport pytest\nfrom unittest import mock\nfrom kaizen.helpers.output import get_parent_folder",
-    "fixed_code": "import os\nfrom kaizen.helpers.output import get_parent_folder",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py",
-    "start_line": 1,
-    "end_line": 6,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 3
-  },
-  {
-    "topic": "Unnecessary mocking",
-    "comment": "The `test_get_parent_folder_normal()` function does not need to mock `os.getcwd()` as the actual implementation can be used.",
-    "confidence": "moderate",
-    "reason": "Mocking should be used only when necessary, as it can make the tests more complex and harder to maintain.",
-    "solution": "Remove the mocking in `test_get_parent_folder_normal()` and use the actual implementation of `get_parent_folder()`.",
-    "actual_code": "    with mock.patch('os.getcwd', return_value='/home/user/project'):\n        expected = '/home/user/project'\n        result = get_parent_folder()\n        assert result == expected, f\"Expected{expected}, but got{result}\"",
-    "fixed_code": "    expected = os.path.dirname(os.getcwd())\n    result = get_parent_folder()\n    assert result == expected, f\"Expected{expected}, but got{result}\"",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py",
-    "start_line": 13,
-    "end_line": 16,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 4
-  },
-  {
-    "topic": "Comprehensive error handling",
-    "comment": "The updated `test_get_parent_folder_error_handling()` function covers more error scenarios, including a generic `Exception` case.",
-    "confidence": "positive",
-    "reason": "Thorough error handling is important to ensure the function behaves correctly in various exceptional situations.",
-    "solution": "No changes needed, the updated error handling tests are comprehensive.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py",
-    "start_line": 20,
-    "end_line": 26,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 5
-  },
-  {
-    "topic": "Nested directory structure test",
-    "comment": "The updated `test_get_parent_folder_nested()` function tests the behavior of `get_parent_folder()` in a nested directory structure.",
-    "confidence": "positive",
-    "reason": "Testing the function in a nested directory structure is important to ensure it works correctly in different scenarios.",
-    "solution": "No changes needed, the nested directory structure test is a good addition.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py",
-    "start_line": 28,
-    "end_line": 33,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 5
-  },
-  {
-    "topic": "Boundary condition test",
-    "comment": "The previous `test_get_parent_folder_boundary_condition_long_path()` function testing a long path has been removed, which is a reasonable decision as the current implementation of `get_parent_folder()` should handle long paths without issues.",
-    "confidence": "positive",
-    "reason": "The previous test case was not necessary as the current implementation should handle long paths correctly.",
-    "solution": "No changes needed, the removal of the unnecessary boundary condition test is appropriate.",
-    "actual_code": "def test_get_parent_folder_boundary_condition_long_path():\n    long_path = \"/\" + \"a\" * 255\n    with mock.patch(\"os.getcwd\", return_value=long_path):\n        assert get_parent_folder() == long_path",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py",
-    "start_line": 29,
-    "end_line": 32,
-    "side": "LEFT",
-    "sentiment": "positive",
-    "severity_level": 4
-  },
-  {
-    "topic": "Unused imports",
-    "comment": "The imports `asyncio` and `nest_asyncio` are not used in the original test cases. Consider removing them if they are not required.",
-    "confidence": "moderate",
-    "reason": "Unused imports can clutter the code and make it less readable.",
-    "solution": "Remove the unused imports `asyncio` and `nest_asyncio` from the test file.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 4,
-    "end_line": 5,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 3
-  },
-  {
-    "topic": "Parameterized test",
-    "comment": "The new test `test_get_web_html_normal_cases` uses a parameterized approach, which is a good practice. It covers multiple test cases in a concise manner.",
-    "confidence": "positive",
-    "reason": "Parameterized tests improve code readability and maintainability by reducing duplication.",
-    "solution": "Keep the parameterized test approach.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 20,
-    "end_line": 75,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 5
-  },
-  {
-    "topic": "Handling empty HTML content",
-    "comment": "The new test `test_get_web_html_empty_html` ensures that the function handles empty HTML content correctly.",
-    "confidence": "positive",
-    "reason": "Handling edge cases like empty input is important for robust error handling.",
-    "solution": "Keep the test case for handling empty HTML content.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 85,
-    "end_line": 92,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 4
-  },
-  {
-    "topic": "Handling network errors",
-    "comment": "The new test `test_get_web_html_invalid_url` ensures that the function handles network errors correctly.",
-    "confidence": "positive",
-    "reason": "Handling network errors is important for a robust web scraping implementation.",
-    "solution": "Keep the test case for handling network errors.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 85,
-    "end_line": 91,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 4
-  },
-  {
-    "topic": "Handling large HTML content",
-    "comment": "The new test `test_get_web_html_large_content` ensures that the function can handle large HTML content without performance issues.",
-    "confidence": "positive",
-    "reason": "Testing the function's ability to handle large inputs is important for ensuring its scalability.",
-    "solution": "Keep the test case for handling large HTML content.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 93,
-    "end_line": 97,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 4
-  },
-  {
-    "topic": "Handling invalid HTML",
-    "comment": "The previous test `test_get_web_html_invalid_html` has been removed. Consider adding a new test case to ensure the function can handle invalid HTML content gracefully.",
-    "confidence": "moderate",
-    "reason": "Handling invalid HTML is important for a robust web scraping implementation.",
-    "solution": "Add a new test case to ensure the function can handle invalid HTML content without raising unexpected exceptions.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 98,
-    "end_line": 107,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 4
-  },
-  {
-    "topic": "Enabling critique mode",
-    "comment": "The `generate_tests` function call in the `examples/unittest/main.py` file has been updated to enable the critique mode and verbose output.",
-    "confidence": "positive",
-    "reason": "Enabling the critique mode and verbose output can provide more detailed feedback and insights during the testing process.",
-    "solution": "Keep the changes to enable critique mode and verbose output in the `generate_tests` function call.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "examples/unittest/main.py",
-    "start_line": 35,
-    "end_line": 37,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 4
-  },
-  {
-    "topic": "Refactoring",
-    "comment": "The `generate_tests` method has become quite long and complex. Consider breaking it down into smaller, more focused methods to improve readability and maintainability.",
-    "confidence": "important",
-    "reason": "Large methods can be difficult to understand and maintain, especially as the codebase grows.",
-    "solution": "Refactor the `generate_tests` method by extracting smaller, more focused methods for specific tasks, such as preparing the test file path, generating the AI tests, and writing the test file.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/unit_test.py",
-    "start_line": 0,
-    "end_line": 0,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Logging",
-    "comment": "The logging implementation could be improved by using a more structured approach, such as using the built-in `logging` module with appropriate log levels.",
-    "confidence": "moderate",
-    "reason": "The current logging implementation uses print statements, which can be less flexible and harder to manage than a structured logging approach.",
-    "solution": "Refactor the logging implementation to use the `logging` module, with appropriate log levels (e.g., DEBUG, INFO, WARNING, ERROR) and log messages that provide more context and details.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/unit_test.py",
-    "start_line": 0,
-    "end_line": 0,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 5
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "The `generate_tests_from_dir` method could benefit from more robust error handling, such as catching and handling specific exceptions.",
-    "confidence": "moderate",
-    "reason": "Catching and handling specific exceptions can help provide more informative error messages and improve the overall robustness of the application.",
-    "solution": "Modify the `generate_tests_from_dir` method to catch and handle specific exceptions, such as `FileNotFoundError` or `ValueError`, and provide more detailed error messages to help with debugging and troubleshooting.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/unit_test.py",
-    "start_line": 0,
-    "end_line": 0,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 6
-  },
-  {
-    "topic": "Separation of Concerns",
-    "comment": "The `UnitTestGenerator` class is responsible for both generating and running the tests. Consider separating these concerns into two different classes or modules.",
-    "confidence": "important",
-    "reason": "Separating the concerns of test generation and test execution can improve the overall design and maintainability of the codebase.",
-    "solution": "Create a separate `UnitTestRunner` class or module that is responsible for discovering and running the generated tests, while the `UnitTestGenerator` class focuses solely on generating the tests.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/unit_test.py",
-    "start_line": 0,
-    "end_line": 0,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Dependency Injection",
-    "comment": "The `UnitTestGenerator` class has several dependencies, such as the `LLMProvider` and the various prompt templates. Consider using dependency injection to improve the testability and flexibility of the class.",
-    "confidence": "moderate",
-    "reason": "Dependency injection can make the code more modular and easier to test, as it allows for easier substitution of dependencies.",
-    "solution": "Refactor the `UnitTestGenerator` class to accept its dependencies (e.g., `LLMProvider`, prompt templates) as constructor arguments, rather than creating them internally. This will improve the testability and flexibility of the class.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/unit_test.py",
-    "start_line": 0,
-    "end_line": 0,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 6
-  },
-  {
-    "topic": "Logging Configuration",
-    "comment": "The code sets all loggers to the ERROR level, which may be too restrictive. Consider providing more granular control over log levels.",
-    "confidence": "moderate",
-    "reason": "Setting all loggers to ERROR level may result in losing valuable information during development and debugging. It's generally better to have more fine-grained control over log levels for different components.",
-    "solution": "Instead of setting all loggers to ERROR, consider the following:\n1. Set a default log level (e.g., INFO) for all loggers using `logging.basicConfig()`.\n2. Selectively set the log level for specific loggers (e.g., 'LiteLLM', 'LiteLLM Router', 'LiteLLM Proxy') to a more appropriate level (e.g., DEBUG, INFO, or WARNING) based on the importance and verbosity of each component.\n3. Provide a way for users to easily adjust the log level, such as through an environment variable or a configuration file.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/llms/provider.py",
-    "start_line": 13,
-    "end_line": 28,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 6
-  },
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "11",
-    "end_line": "11",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_400/review.md b/.experiments/code_review/haiku/no_eval/pr_400/review.md
deleted file mode 100644
index a785248a..00000000
--- a/.experiments/code_review/haiku/no_eval/pr_400/review.md
+++ /dev/null
@@ -1,177 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/400
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 27
-- Critical: 1
-- Important: 4
-- Minor: 9
-- Files Affected: 8
-## 🏆 Code Quality
-[██████████████████░░] 90% (Excellent)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Configuration (1 issues)</strong></summary>
-
-### 1. Changes made to sensitive file
-📁 **File:** `config.json:11`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to config.json, which needs review
-💡 **Solution:** NA
-
-</details>
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Boundary Conditions (4 issues)</strong></summary>
-
-### 1. The test cases for boundary conditions (very long descriptions) look good. The execution time is also printed, which is a nice addition.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_description.py:45`
-⚖️ **Severity:** 8/10
-🔍 **Description:** Handling large inputs is an important aspect of the function's robustness.
-💡 **Solution:** No changes needed.
-
-### 2. The error handling tests cover various invalid input scenarios, which is good.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_description.py:31`
-⚖️ **Severity:** 8/10
-🔍 **Description:** Proper error handling is crucial for the function's reliability.
-💡 **Solution:** No changes needed.
-
-### 3. The `generate_tests` method has become quite long and complex. Consider breaking it down into smaller, more focused methods to improve readability and maintainability.
-📁 **File:** `kaizen/generator/unit_test.py:0`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Large methods can be difficult to understand and maintain, especially as the codebase grows.
-💡 **Solution:** Refactor the `generate_tests` method by extracting smaller, more focused methods for specific tasks, such as preparing the test file path, generating the AI tests, and writing the test file.
-
-### 4. The `UnitTestGenerator` class is responsible for both generating and running the tests. Consider separating these concerns into two different classes or modules.
-📁 **File:** `kaizen/generator/unit_test.py:0`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Separating the concerns of test generation and test execution can improve the overall design and maintainability of the codebase.
-💡 **Solution:** Create a separate `UnitTestRunner` class or module that is responsible for discovering and running the generated tests, while the `UnitTestGenerator` class focuses solely on generating the tests.
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (9 issues)</strong></summary>
-
-<details>
-<summary><strong>Collapsible Template (9 issues)</strong></summary>
-
-### 1. The collapsible template for the original description has been improved to include newlines for better readability.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_description.py:5`
-⚖️ **Severity:** 5/10
-🔍 **Description:** The previous template did not have proper newline formatting.
-💡 **Solution:** No changes needed.
-
-### 2. The imports `mock` and `pytest` are not used in the updated tests.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py:1`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Unused imports can make the code harder to read and maintain.
-💡 **Solution:** Remove the unused imports.
-
-**Current Code:**
-```python
-import os
-import pytest
-from unittest import mock
-from kaizen.helpers.output import get_parent_folder
-```
-
-**Suggested Code:**
-```python
-import os
-from kaizen.helpers.output import get_parent_folder
-```
-
-### 3. The `test_get_parent_folder_normal()` function does not need to mock `os.getcwd()` as the actual implementation can be used.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py:13`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Mocking should be used only when necessary, as it can make the tests more complex and harder to maintain.
-💡 **Solution:** Remove the mocking in `test_get_parent_folder_normal()` and use the actual implementation of `get_parent_folder()`.
-
-**Current Code:**
-```python
-    with mock.patch('os.getcwd', return_value='/home/user/project'):
-        expected = '/home/user/project'
-        result = get_parent_folder()
-        assert result == expected, f"Expected{expected}, but got{result}"
-```
-
-**Suggested Code:**
-```python
-    expected = os.path.dirname(os.getcwd())
-    result = get_parent_folder()
-    assert result == expected, f"Expected{expected}, but got{result}"
-```
-
-### 4. The imports `asyncio` and `nest_asyncio` are not used in the original test cases. Consider removing them if they are not required.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_web_html.py:4`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Unused imports can clutter the code and make it less readable.
-💡 **Solution:** Remove the unused imports `asyncio` and `nest_asyncio` from the test file.
-
-### 5. The previous test `test_get_web_html_invalid_html` has been removed. Consider adding a new test case to ensure the function can handle invalid HTML content gracefully.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_web_html.py:98`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Handling invalid HTML is important for a robust web scraping implementation.
-💡 **Solution:** Add a new test case to ensure the function can handle invalid HTML content without raising unexpected exceptions.
-
-### 6. The logging implementation could be improved by using a more structured approach, such as using the built-in `logging` module with appropriate log levels.
-📁 **File:** `kaizen/generator/unit_test.py:0`
-⚖️ **Severity:** 5/10
-🔍 **Description:** The current logging implementation uses print statements, which can be less flexible and harder to manage than a structured logging approach.
-💡 **Solution:** Refactor the logging implementation to use the `logging` module, with appropriate log levels (e.g., DEBUG, INFO, WARNING, ERROR) and log messages that provide more context and details.
-
-### 7. The `generate_tests_from_dir` method could benefit from more robust error handling, such as catching and handling specific exceptions.
-📁 **File:** `kaizen/generator/unit_test.py:0`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Catching and handling specific exceptions can help provide more informative error messages and improve the overall robustness of the application.
-💡 **Solution:** Modify the `generate_tests_from_dir` method to catch and handle specific exceptions, such as `FileNotFoundError` or `ValueError`, and provide more detailed error messages to help with debugging and troubleshooting.
-
-### 8. The `UnitTestGenerator` class has several dependencies, such as the `LLMProvider` and the various prompt templates. Consider using dependency injection to improve the testability and flexibility of the class.
-📁 **File:** `kaizen/generator/unit_test.py:0`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Dependency injection can make the code more modular and easier to test, as it allows for easier substitution of dependencies.
-💡 **Solution:** Refactor the `UnitTestGenerator` class to accept its dependencies (e.g., `LLMProvider`, prompt templates) as constructor arguments, rather than creating them internally. This will improve the testability and flexibility of the class.
-
-### 9. The code sets all loggers to the ERROR level, which may be too restrictive. Consider providing more granular control over log levels.
-📁 **File:** `kaizen/llms/provider.py:13`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Setting all loggers to ERROR level may result in losing valuable information during development and debugging. It's generally better to have more fine-grained control over log levels for different components.
-💡 **Solution:** Instead of setting all loggers to ERROR, consider the following:
-1. Set a default log level (e.g., INFO) for all loggers using `logging.basicConfig()`.
-2. Selectively set the log level for specific loggers (e.g., 'LiteLLM', 'LiteLLM Router', 'LiteLLM Proxy') to a more appropriate level (e.g., DEBUG, INFO, or WARNING) based on the importance and verbosity of each component.
-3. Provide a way for users to easily adjust the log level, such as through an environment variable or a configuration file.
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (anthropic.claude-3-haiku-20240307-v1:0)
-{"prompt_tokens": 42996, "completion_tokens": 5792, "total_tokens": 48788}
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_440/issues.json b/.experiments/code_review/haiku/no_eval/pr_440/issues.json
deleted file mode 100644
index 8e24d702..00000000
--- a/.experiments/code_review/haiku/no_eval/pr_440/issues.json
+++ /dev/null
@@ -1,62 +0,0 @@
-[
-  {
-    "topic": "Unnecessary comment removal",
-    "comment": "The commented-out line `# analyzer.setup_repository(\"./github_app/\")` should be removed, as the line below it already sets up the repository.",
-    "confidence": "moderate",
-    "reason": "Removing unnecessary comments improves code readability and maintainability.",
-    "solution": "Remove the commented-out line `# analyzer.setup_repository(\"./github_app/\")` in `examples/ragify_codebase/main.py`.",
-    "actual_code": "# analyzer.setup_repository(\"./github_app/\")",
-    "fixed_code": "",
-    "file_name": "examples/ragify_codebase/main.py",
-    "start_line": 7,
-    "end_line": 7,
-    "side": "LEFT",
-    "sentiment": "positive",
-    "severity_level": 3
-  },
-  {
-    "topic": "Duplicate embedding storage",
-    "comment": "The TODO comment `# TODO: DONT PUSH DUPLICATE` suggests that the code is storing duplicate embeddings, which could lead to performance and storage issues.",
-    "confidence": "important",
-    "reason": "Storing duplicate embeddings can waste storage space and slow down the retrieval process.",
-    "solution": "Implement a mechanism to check for and avoid storing duplicate embeddings in the database.",
-    "actual_code": "# TODO: DONT PUSH DUPLICATE",
-    "fixed_code": "",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 156,
-    "end_line": 156,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Dependency version update",
-    "comment": "The `llama-index-core` dependency version has been updated from `0.10.47` to `0.10.65`. This update should be reviewed to ensure compatibility with the rest of the codebase.",
-    "confidence": "moderate",
-    "reason": "Dependency version updates can introduce breaking changes, so it's important to review the changes and ensure they don't introduce any issues.",
-    "solution": "Review the changelog and release notes for the `llama-index-core` version update to understand the changes and ensure they don't introduce any issues in the codebase.",
-    "actual_code": "llama-index-core = \"0.10.65\"",
-    "fixed_code": "",
-    "file_name": "pyproject.toml",
-    "start_line": 27,
-    "end_line": 27,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  },
-  {
-    "topic": "Dependency removal",
-    "comment": "The `llama-index-llms-openai` dependency has been removed. This change should be reviewed to ensure that it doesn't impact the functionality of the codebase.",
-    "confidence": "moderate",
-    "reason": "Removing dependencies can have unintended consequences, so it's important to review the impact of the change.",
-    "solution": "Review the codebase to ensure that the removal of the `llama-index-llms-openai` dependency doesn't break any functionality.",
-    "actual_code": "llama-index-llms-openai = \"^0.1.22\"",
-    "fixed_code": "",
-    "file_name": "pyproject.toml",
-    "start_line": 28,
-    "end_line": 28,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_440/review.md b/.experiments/code_review/haiku/no_eval/pr_440/review.md
deleted file mode 100644
index 62d5b74d..00000000
--- a/.experiments/code_review/haiku/no_eval/pr_440/review.md
+++ /dev/null
@@ -1,118 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/440
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 4
-- Critical: 0
-- Important: 1
-- Minor: 3
-- Files Affected: 3
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Duplicate embedding storage (1 issues)</strong></summary>
-
-### 1. The TODO comment `# TODO: DONT PUSH DUPLICATE` suggests that the code is storing duplicate embeddings, which could lead to performance and storage issues.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:156`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Storing duplicate embeddings can waste storage space and slow down the retrieval process.
-💡 **Solution:** Implement a mechanism to check for and avoid storing duplicate embeddings in the database.
-
-**Current Code:**
-```python
-# TODO: DONT PUSH DUPLICATE
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (3 issues)</strong></summary>
-
-<details>
-<summary><strong>Unnecessary comment removal (3 issues)</strong></summary>
-
-### 1. The commented-out line `# analyzer.setup_repository("./github_app/")` should be removed, as the line below it already sets up the repository.
-📁 **File:** `examples/ragify_codebase/main.py:7`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Removing unnecessary comments improves code readability and maintainability.
-💡 **Solution:** Remove the commented-out line `# analyzer.setup_repository("./github_app/")` in `examples/ragify_codebase/main.py`.
-
-**Current Code:**
-```python
-# analyzer.setup_repository("./github_app/")
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 2. The `llama-index-core` dependency version has been updated from `0.10.47` to `0.10.65`. This update should be reviewed to ensure compatibility with the rest of the codebase.
-📁 **File:** `pyproject.toml:27`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Dependency version updates can introduce breaking changes, so it's important to review the changes and ensure they don't introduce any issues.
-💡 **Solution:** Review the changelog and release notes for the `llama-index-core` version update to understand the changes and ensure they don't introduce any issues in the codebase.
-
-**Current Code:**
-```python
-llama-index-core = "0.10.65"
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 3. The `llama-index-llms-openai` dependency has been removed. This change should be reviewed to ensure that it doesn't impact the functionality of the codebase.
-📁 **File:** `pyproject.toml:28`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Removing dependencies can have unintended consequences, so it's important to review the impact of the change.
-💡 **Solution:** Review the codebase to ensure that the removal of the `llama-index-llms-openai` dependency doesn't break any functionality.
-
-**Current Code:**
-```python
-llama-index-llms-openai = "^0.1.22"
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (anthropic.claude-3-haiku-20240307-v1:0)
-{"prompt_tokens": 1579, "completion_tokens": 944, "total_tokens": 2523}
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_476/comments.json b/.experiments/code_review/haiku/no_eval/pr_476/comments.json
deleted file mode 100644
index a9d40eac..00000000
--- a/.experiments/code_review/haiku/no_eval/pr_476/comments.json
+++ /dev/null
@@ -1,16 +0,0 @@
-[
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_476/issues.json b/.experiments/code_review/haiku/no_eval/pr_476/issues.json
deleted file mode 100644
index 15365ce7..00000000
--- a/.experiments/code_review/haiku/no_eval/pr_476/issues.json
+++ /dev/null
@@ -1,106 +0,0 @@
-[
-  {
-    "topic": "Sorting PR Files",
-    "comment": "The PR files are now being sorted before passing them to the description generator. This is a good improvement for maintaining consistent file order in the review.",
-    "confidence": "important",
-    "reason": "Sorting the files ensures a consistent order in the review, making it easier for the reviewer to understand the changes.",
-    "solution": "The `sort_files` function looks good and should effectively sort the files in alphabetical order.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "github_app/github_helper/pull_requests.py",
-    "start_line": 184,
-    "end_line": 194,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 4
-  },
-  {
-    "topic": "Generating Tests",
-    "comment": "The new `generate_tests` function is a good addition, as it provides a way to generate test cases based on the changed files in the PR.",
-    "confidence": "important",
-    "reason": "Generating tests based on the PR files can help ensure the changes don't break existing functionality.",
-    "solution": "The current implementation of `generate_tests` is simple and effective. It returns a list of file names, which can be used to create test cases.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "github_app/github_helper/pull_requests.py",
-    "start_line": 199,
-    "end_line": 200,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 4
-  },
-  {
-    "topic": "Printing Diff and PR Files",
-    "comment": "The changes in the `main.py` file to print the `diff_text` and `pr_files` are useful for debugging and understanding the input data.",
-    "confidence": "moderate",
-    "reason": "Printing the diff and PR files can help developers better understand the changes being reviewed.",
-    "solution": "The changes look good and should provide helpful information during the review process.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "examples/code_review/main.py",
-    "start_line": 21,
-    "end_line": 22,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 3
-  },
-  {
-    "topic": "Passing Code Quality to Review Description",
-    "comment": "The change to pass the `code_quality` parameter to the `create_pr_review_text` function is a good improvement, as it allows the review description to include information about the overall code quality.",
-    "confidence": "important",
-    "reason": "Providing information about the code quality in the review description can give the developer a better understanding of the overall state of the codebase.",
-    "solution": "The change looks good and should provide valuable information in the review description.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "examples/code_review/main.py",
-    "start_line": 36,
-    "end_line": 36,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 4
-  },
-  {
-    "topic": "Printing Raw Issues",
-    "comment": "The change to print the `review_data.issues` instead of the `topics` variable is an improvement, as it provides more detailed information about the identified issues.",
-    "confidence": "moderate",
-    "reason": "Printing the raw issues can give the developer a better understanding of the specific problems found during the review.",
-    "solution": "The change looks good and should provide more useful information in the output.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "examples/code_review/main.py",
-    "start_line": 39,
-    "end_line": 39,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 3
-  },
-  {
-    "topic": "Removing Unused Configuration",
-    "comment": "The removal of the `enable_observability_logging` configuration option is a good cleanup, as it removes an unused feature from the configuration file.",
-    "confidence": "moderate",
-    "reason": "Removing unused configuration options helps keep the codebase clean and maintainable.",
-    "solution": "The change looks good and should help simplify the configuration file.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "config.json",
-    "start_line": 4,
-    "end_line": 4,
-    "side": "LEFT",
-    "sentiment": "positive",
-    "severity_level": 3
-  },
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_476/review.md b/.experiments/code_review/haiku/no_eval/pr_476/review.md
deleted file mode 100644
index 86156ba1..00000000
--- a/.experiments/code_review/haiku/no_eval/pr_476/review.md
+++ /dev/null
@@ -1,103 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/476
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 7
-- Critical: 1
-- Important: 3
-- Minor: 3
-- Files Affected: 3
-## 🏆 Code Quality
-[██████████████████░░] 90% (Excellent)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Configuration (1 issues)</strong></summary>
-
-### 1. Changes made to sensitive file
-📁 **File:** `config.json:1`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to config.json, which needs review
-💡 **Solution:** NA
-
-</details>
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Sorting PR Files (3 issues)</strong></summary>
-
-### 1. The PR files are now being sorted before passing them to the description generator. This is a good improvement for maintaining consistent file order in the review.
-📁 **File:** `github_app/github_helper/pull_requests.py:184`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Sorting the files ensures a consistent order in the review, making it easier for the reviewer to understand the changes.
-💡 **Solution:** The `sort_files` function looks good and should effectively sort the files in alphabetical order.
-
-### 2. The new `generate_tests` function is a good addition, as it provides a way to generate test cases based on the changed files in the PR.
-📁 **File:** `github_app/github_helper/pull_requests.py:199`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Generating tests based on the PR files can help ensure the changes don't break existing functionality.
-💡 **Solution:** The current implementation of `generate_tests` is simple and effective. It returns a list of file names, which can be used to create test cases.
-
-### 3. The change to pass the `code_quality` parameter to the `create_pr_review_text` function is a good improvement, as it allows the review description to include information about the overall code quality.
-📁 **File:** `examples/code_review/main.py:36`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Providing information about the code quality in the review description can give the developer a better understanding of the overall state of the codebase.
-💡 **Solution:** The change looks good and should provide valuable information in the review description.
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (3 issues)</strong></summary>
-
-<details>
-<summary><strong>Printing Diff and PR Files (3 issues)</strong></summary>
-
-### 1. The changes in the `main.py` file to print the `diff_text` and `pr_files` are useful for debugging and understanding the input data.
-📁 **File:** `examples/code_review/main.py:21`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Printing the diff and PR files can help developers better understand the changes being reviewed.
-💡 **Solution:** The changes look good and should provide helpful information during the review process.
-
-### 2. The change to print the `review_data.issues` instead of the `topics` variable is an improvement, as it provides more detailed information about the identified issues.
-📁 **File:** `examples/code_review/main.py:39`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Printing the raw issues can give the developer a better understanding of the specific problems found during the review.
-💡 **Solution:** The change looks good and should provide more useful information in the output.
-
-### 3. The removal of the `enable_observability_logging` configuration option is a good cleanup, as it removes an unused feature from the configuration file.
-📁 **File:** `config.json:4`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Removing unused configuration options helps keep the codebase clean and maintainable.
-💡 **Solution:** The change looks good and should help simplify the configuration file.
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (anthropic.claude-3-haiku-20240307-v1:0)
-{"prompt_tokens": 4511, "completion_tokens": 1305, "total_tokens": 5816}
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_5/comments.json b/.experiments/code_review/haiku/no_eval/pr_5/comments.json
deleted file mode 100644
index 1130b98e..00000000
--- a/.experiments/code_review/haiku/no_eval/pr_5/comments.json
+++ /dev/null
@@ -1,32 +0,0 @@
-[
-  {
-    "topic": "Error Handling",
-    "comment": "Potential for API call to fail without retry mechanism",
-    "confidence": "critical",
-    "reason": "The `process_applicant` function makes a call to the `completion` function, which could fail for various reasons (e.g., network issues, API downtime). Without a retry mechanism, the function will silently fail, leading to incomplete or inaccurate results.",
-    "solution": "Implement a retry mechanism with exponential backoff to handle transient failures in the `completion` function call. This will improve the reliability and robustness of the application.",
-    "actual_code": "response = completion(\n    model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n)",
-    "fixed_code": "num_retries = 3\nretry_delay = 1\nfor _ in range(num_retries):\n    try:\n        response = completion(\n            model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n        )\n        break\n    except Exception as e:\n        print(f\"Error calling completion function:{e}. Retrying in{retry_delay}seconds...\")\n        time.sleep(retry_delay)\n        retry_delay *= 2\nelse:\n    print(\"Failed to call completion function after multiple retries. Skipping this applicant.\")\n    return{key: \"\" for key in[\"feedback\", \"review\", \"should_interview\", \"rating\", \"input_tokens\", \"output_tokens\"]}",
-    "file_name": "main.py",
-    "start_line": 66,
-    "end_line": 68,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 9
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "Silent failure without logging",
-    "confidence": "critical",
-    "reason": "In the `process_applicant` function, when a `json.JSONDecodeError` is raised, the function silently returns a default result without any logging or error reporting. This makes it difficult to diagnose and troubleshoot issues that may occur during the processing of applicants.",
-    "solution": "Add proper error logging to the `except` block to capture the error and provide more visibility into the failure. This will help with debugging and monitoring the application's performance.",
-    "actual_code": "except json.JSONDecodeError:\n    result ={\n        key: \"\" for key in[\n            \"feedback\",\n            \"review\",\n            \"should_interview\",\n            \"rating\",\n            \"input_tokens\",\n            \"output_tokens\",\n        ]\n}",
-    "fixed_code": "except json.JSONDecodeError as e:\n    print(f\"Failed to parse content for applicant:{e}\")\n    result ={\n        key: \"\" for key in[\n            \"feedback\",\n            \"review\",\n            \"should_interview\",\n            \"rating\",\n            \"input_tokens\",\n            \"output_tokens\",\n        ]\n}",
-    "file_name": "main.py",
-    "start_line": 82,
-    "end_line": 94,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 8
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_5/issues.json b/.experiments/code_review/haiku/no_eval/pr_5/issues.json
deleted file mode 100644
index 5482c08e..00000000
--- a/.experiments/code_review/haiku/no_eval/pr_5/issues.json
+++ /dev/null
@@ -1,107 +0,0 @@
-[
-  {
-    "topic": "Error Handling",
-    "comment": "Potential for API call to fail without retry mechanism",
-    "confidence": "critical",
-    "reason": "The `process_applicant` function makes a call to the `completion` function, which could fail for various reasons (e.g., network issues, API downtime). Without a retry mechanism, the function will silently fail, leading to incomplete or inaccurate results.",
-    "solution": "Implement a retry mechanism with exponential backoff to handle transient failures in the `completion` function call. This will improve the reliability and robustness of the application.",
-    "actual_code": "response = completion(\n    model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n)",
-    "fixed_code": "num_retries = 3\nretry_delay = 1\nfor _ in range(num_retries):\n    try:\n        response = completion(\n            model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n        )\n        break\n    except Exception as e:\n        print(f\"Error calling completion function:{e}. Retrying in{retry_delay}seconds...\")\n        time.sleep(retry_delay)\n        retry_delay *= 2\nelse:\n    print(\"Failed to call completion function after multiple retries. Skipping this applicant.\")\n    return{key: \"\" for key in[\"feedback\", \"review\", \"should_interview\", \"rating\", \"input_tokens\", \"output_tokens\"]}",
-    "file_name": "main.py",
-    "start_line": 66,
-    "end_line": 68,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 9
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "Silent failure without logging",
-    "confidence": "critical",
-    "reason": "In the `process_applicant` function, when a `json.JSONDecodeError` is raised, the function silently returns a default result without any logging or error reporting. This makes it difficult to diagnose and troubleshoot issues that may occur during the processing of applicants.",
-    "solution": "Add proper error logging to the `except` block to capture the error and provide more visibility into the failure. This will help with debugging and monitoring the application's performance.",
-    "actual_code": "except json.JSONDecodeError:\n    result ={\n        key: \"\" for key in[\n            \"feedback\",\n            \"review\",\n            \"should_interview\",\n            \"rating\",\n            \"input_tokens\",\n            \"output_tokens\",\n        ]\n}",
-    "fixed_code": "except json.JSONDecodeError as e:\n    print(f\"Failed to parse content for applicant:{e}\")\n    result ={\n        key: \"\" for key in[\n            \"feedback\",\n            \"review\",\n            \"should_interview\",\n            \"rating\",\n            \"input_tokens\",\n            \"output_tokens\",\n        ]\n}",
-    "file_name": "main.py",
-    "start_line": 82,
-    "end_line": 94,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Performance",
-    "comment": "Inefficient way to print progress",
-    "confidence": "important",
-    "reason": "The `process_applicants` function uses a print statement with carriage return (`\r`) to update the progress bar. This approach can be inefficient, especially for large datasets, as it requires continuously overwriting the same line of output.",
-    "solution": "Use a dedicated progress reporting library, such as `tqdm`, which provides a more efficient and visually appealing progress bar. This will improve the overall performance and user experience of the application.",
-    "actual_code": "progress = (index + 1) / total\nprint(f\"\\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}\", end=\"\", flush=True)",
-    "fixed_code": "if use_tqdm:\n    progress_bar = tqdm(total=total, desc=\"Processing applicants\")\n    progress_bar.update(1)\nelse:\n    progress = (index + 1) / total\n    print(f\"\\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}\", end=\"\", flush=True)",
-    "file_name": "main.py",
-    "start_line": 120,
-    "end_line": 122,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 6
-  },
-  {
-    "topic": "Redundant Code",
-    "comment": "Redundant code: The following line is unnecessary",
-    "confidence": "moderate",
-    "reason": "The `if len(df) == 0` check in the `main` function is redundant, as the `process_applicants` function already handles the case where the DataFrame is empty.",
-    "solution": "Remove the unnecessary `if` statement, as it does not provide any additional value to the code.",
-    "actual_code": "if len(df) == 0:\n    return",
-    "fixed_code": "",
-    "file_name": "main.py",
-    "start_line": 142,
-    "end_line": 143,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 3
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "Division by zero potential if total_tokens is zero",
-    "confidence": "important",
-    "reason": "In the `main` function, the code calculates the total tokens used and prints a summary. However, if the total tokens is zero, the division operation will result in a division by zero error, which can cause the application to crash.",
-    "solution": "Add a check to ensure that the total tokens is not zero before performing the division operation. If the total tokens is zero, handle the case gracefully by printing a message or skipping the division step.",
-    "actual_code": "print(f\"Total tokens used:{total_tokens:,}\")\nprint(f\"  - Input tokens:{total_input_tokens:,}\")\nprint(f\"  - Output tokens:{total_output_tokens:,}\")",
-    "fixed_code": "if total_tokens > 0:\n    print(f\"Total tokens used:{total_tokens:,}\")\n    print(f\"  - Input tokens:{total_input_tokens:,}\")\n    print(f\"  - Output tokens:{total_output_tokens:,}\")\nelse:\n    print(\"Total tokens used: 0\")",
-    "file_name": "main.py",
-    "start_line": 158,
-    "end_line": 163,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "No error handling for file not found",
-    "confidence": "important",
-    "reason": "The `main` function does not handle the case where the input file specified by the user does not exist. This can lead to a `FileNotFoundError` being raised, which will cause the application to crash without any meaningful error message.",
-    "solution": "Add a try-except block to handle the `FileNotFoundError` and provide a user-friendly error message when the input file is not found.",
-    "actual_code": "main(input_file)",
-    "fixed_code": "try:\n    main(input_file)\nexcept FileNotFoundError:\n    print(f\"Error: The file '{input_file}' does not exist. Please check the file path and try again.\")",
-    "file_name": "main.py",
-    "start_line": 174,
-    "end_line": 175,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Unused Import",
-    "comment": "Unused import",
-    "confidence": "low",
-    "reason": "The `random` module is imported but not used in the code.",
-    "solution": "Remove the unused import statement.",
-    "actual_code": "import random  # Unused import",
-    "fixed_code": "",
-    "file_name": "main.py",
-    "start_line": 8,
-    "end_line": 8,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 1
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/haiku/no_eval/pr_5/review.md b/.experiments/code_review/haiku/no_eval/pr_5/review.md
deleted file mode 100644
index 5e96731d..00000000
--- a/.experiments/code_review/haiku/no_eval/pr_5/review.md
+++ /dev/null
@@ -1,211 +0,0 @@
-PR URL: https://github.com/sauravpanda/applicant-screening/pull/5
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 7
-- Critical: 2
-- Important: 3
-- Minor: 1
-- Files Affected: 1
-## 🏆 Code Quality
-[███████████████░░░░░] 75% (Fair)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Error Handling (2 issues)</strong></summary>
-
-### 1. Potential for API call to fail without retry mechanism
-📁 **File:** `main.py:66`
-⚖️ **Severity:** 9/10
-🔍 **Description:** The `process_applicant` function makes a call to the `completion` function, which could fail for various reasons (e.g., network issues, API downtime). Without a retry mechanism, the function will silently fail, leading to incomplete or inaccurate results.
-💡 **Solution:** Implement a retry mechanism with exponential backoff to handle transient failures in the `completion` function call. This will improve the reliability and robustness of the application.
-
-**Current Code:**
-```python
-response = completion(
-    model=os.environ.get("model", "anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1"), messages=messages
-)
-```
-
-**Suggested Code:**
-```python
-num_retries = 3
-retry_delay = 1
-for _ in range(num_retries):
-    try:
-        response = completion(
-            model=os.environ.get("model", "anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1"), messages=messages
-        )
-        break
-    except Exception as e:
-        print(f"Error calling completion function:{e}. Retrying in{retry_delay}seconds...")
-        time.sleep(retry_delay)
-        retry_delay *= 2
-else:
-    print("Failed to call completion function after multiple retries. Skipping this applicant.")
-    return{key: "" for key in["feedback", "review", "should_interview", "rating", "input_tokens", "output_tokens"]}
-```
-
-### 2. Silent failure without logging
-📁 **File:** `main.py:82`
-⚖️ **Severity:** 8/10
-🔍 **Description:** In the `process_applicant` function, when a `json.JSONDecodeError` is raised, the function silently returns a default result without any logging or error reporting. This makes it difficult to diagnose and troubleshoot issues that may occur during the processing of applicants.
-💡 **Solution:** Add proper error logging to the `except` block to capture the error and provide more visibility into the failure. This will help with debugging and monitoring the application's performance.
-
-**Current Code:**
-```python
-except json.JSONDecodeError:
-    result ={
-        key: "" for key in[
-            "feedback",
-            "review",
-            "should_interview",
-            "rating",
-            "input_tokens",
-            "output_tokens",
-        ]
-}
-```
-
-**Suggested Code:**
-```python
-except json.JSONDecodeError as e:
-    print(f"Failed to parse content for applicant:{e}")
-    result ={
-        key: "" for key in[
-            "feedback",
-            "review",
-            "should_interview",
-            "rating",
-            "input_tokens",
-            "output_tokens",
-        ]
-}
-```
-
-</details>
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Performance (3 issues)</strong></summary>
-
-### 1. Inefficient way to print progress
-📁 **File:** `main.py:120`
-⚖️ **Severity:** 6/10
-🔍 **Description:** The `process_applicants` function uses a print statement with carriage return (`
`) to update the progress bar. This approach can be inefficient, especially for large datasets, as it requires continuously overwriting the same line of output.
-💡 **Solution:** Use a dedicated progress reporting library, such as `tqdm`, which provides a more efficient and visually appealing progress bar. This will improve the overall performance and user experience of the application.
-
-**Current Code:**
-```python
-progress = (index + 1) / total
-print(f"\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}", end="", flush=True)
-```
-
-**Suggested Code:**
-```python
-if use_tqdm:
-    progress_bar = tqdm(total=total, desc="Processing applicants")
-    progress_bar.update(1)
-else:
-    progress = (index + 1) / total
-    print(f"\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}", end="", flush=True)
-```
-
-### 2. Division by zero potential if total_tokens is zero
-📁 **File:** `main.py:158`
-⚖️ **Severity:** 7/10
-🔍 **Description:** In the `main` function, the code calculates the total tokens used and prints a summary. However, if the total tokens is zero, the division operation will result in a division by zero error, which can cause the application to crash.
-💡 **Solution:** Add a check to ensure that the total tokens is not zero before performing the division operation. If the total tokens is zero, handle the case gracefully by printing a message or skipping the division step.
-
-**Current Code:**
-```python
-print(f"Total tokens used:{total_tokens:,}")
-print(f"  - Input tokens:{total_input_tokens:,}")
-print(f"  - Output tokens:{total_output_tokens:,}")
-```
-
-**Suggested Code:**
-```python
-if total_tokens > 0:
-    print(f"Total tokens used:{total_tokens:,}")
-    print(f"  - Input tokens:{total_input_tokens:,}")
-    print(f"  - Output tokens:{total_output_tokens:,}")
-else:
-    print("Total tokens used: 0")
-```
-
-### 3. No error handling for file not found
-📁 **File:** `main.py:174`
-⚖️ **Severity:** 7/10
-🔍 **Description:** The `main` function does not handle the case where the input file specified by the user does not exist. This can lead to a `FileNotFoundError` being raised, which will cause the application to crash without any meaningful error message.
-💡 **Solution:** Add a try-except block to handle the `FileNotFoundError` and provide a user-friendly error message when the input file is not found.
-
-**Current Code:**
-```python
-main(input_file)
-```
-
-**Suggested Code:**
-```python
-try:
-    main(input_file)
-except FileNotFoundError:
-    print(f"Error: The file '{input_file}' does not exist. Please check the file path and try again.")
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (2 issues)</strong></summary>
-
-<details>
-<summary><strong>Redundant Code (1 issues)</strong></summary>
-
-### 1. Redundant code: The following line is unnecessary
-📁 **File:** `main.py:142`
-⚖️ **Severity:** 3/10
-🔍 **Description:** The `if len(df) == 0` check in the `main` function is redundant, as the `process_applicants` function already handles the case where the DataFrame is empty.
-💡 **Solution:** Remove the unnecessary `if` statement, as it does not provide any additional value to the code.
-
-**Current Code:**
-```python
-if len(df) == 0:
-    return
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (anthropic.claude-3-haiku-20240307-v1:0)
-{"prompt_tokens": 7009, "completion_tokens": 2247, "total_tokens": 9256}
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_222/comments.json b/.experiments/code_review/llama-405b/no_eval/pr_222/comments.json
deleted file mode 100644
index f054c504..00000000
--- a/.experiments/code_review/llama-405b/no_eval/pr_222/comments.json
+++ /dev/null
@@ -1,117 +0,0 @@
-[
-  {
-    "topic": "Error handling",
-    "comment": "Error handling is missing in some critical sections of the code.",
-    "confidence": "critical",
-    "reason": "Error handling is crucial for preventing crashes and providing useful error messages.",
-    "solution": "Add try-except blocks to handle potential errors.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "",
-    "start_line": 0,
-    "end_line": 0,
-    "side": "",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "SQL Injection",
-    "comment": "Potential SQL injection vulnerability",
-    "confidence": "critical",
-    "reason": "Using string formatting to construct SQL queries can lead to SQL injection attacks",
-    "solution": "Use parameterized queries or an ORM to prevent SQL injection",
-    "actual_code": "query = f\"\"\"SELECT ... FROM{self.table_name}e\"\"\"",
-    "fixed_code": "query = \"\"\"SELECT ... FROM %s e\"\"\" % self.table_name",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 19,
-    "end_line": 37,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 9
-  },
-  {
-    "topic": "Version bump",
-    "comment": "Verify that the version bump is intentional and follows the project's versioning scheme",
-    "confidence": "critical",
-    "reason": "Inconsistent versioning can cause confusion and break dependencies",
-    "solution": "Verify the version bump and update the project's versioning scheme if necessary",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "pyproject.toml",
-    "start_line": 3,
-    "end_line": 3,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 9
-  },
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Docker",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to Dockerfile, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_name": "Dockerfile",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Docker",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to docker-compose.yml, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "15",
-    "end_line": "15",
-    "side": "RIGHT",
-    "file_name": "docker-compose.yml",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Version Control",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to .gitignore, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "164",
-    "end_line": "164",
-    "side": "RIGHT",
-    "file_name": ".gitignore",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Database",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to db_setup/init.sql, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_name": "db_setup/init.sql",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_222/issues.json b/.experiments/code_review/llama-405b/no_eval/pr_222/issues.json
deleted file mode 100644
index 599c0667..00000000
--- a/.experiments/code_review/llama-405b/no_eval/pr_222/issues.json
+++ /dev/null
@@ -1,282 +0,0 @@
-[
-  {
-    "topic": "Code organization",
-    "comment": "The code is well-organized, but some files have too many responsibilities.",
-    "confidence": "important",
-    "reason": "Separation of concerns is crucial for maintainability.",
-    "solution": "Consider breaking down large files into smaller ones.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "",
-    "start_line": 0,
-    "end_line": 0,
-    "side": "",
-    "sentiment": "neutral",
-    "severity_level": 5
-  },
-  {
-    "topic": "Type hints",
-    "comment": "Type hints are missing in some function definitions.",
-    "confidence": "moderate",
-    "reason": "Type hints improve code readability and help catch type-related errors.",
-    "solution": "Add type hints for function parameters and return types.",
-    "actual_code": "def chunk_code(code: str, language: str) -> ParsedBody:",
-    "fixed_code": "def chunk_code(code: str, language: str) -> Dict[str, Dict[str, Any]]:",
-    "file_name": "kaizen/retriever/code_chunker.py",
-    "start_line": 7,
-    "end_line": 7,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 3
-  },
-  {
-    "topic": "Error handling",
-    "comment": "Error handling is missing in some critical sections of the code.",
-    "confidence": "critical",
-    "reason": "Error handling is crucial for preventing crashes and providing useful error messages.",
-    "solution": "Add try-except blocks to handle potential errors.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "",
-    "start_line": 0,
-    "end_line": 0,
-    "side": "",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Code duplication",
-    "comment": "Some code is duplicated across multiple files.",
-    "confidence": "moderate",
-    "reason": "Code duplication makes maintenance harder and increases the chance of bugs.",
-    "solution": "Extract duplicated code into reusable functions or classes.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "",
-    "start_line": 0,
-    "end_line": 0,
-    "side": "",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Type Hints",
-    "comment": "Missing type hints for function return types",
-    "confidence": "moderate",
-    "reason": "Type hints improve code readability and help catch type-related errors",
-    "solution": "Add type hints for function return types",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 13,
-    "end_line": 52,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  },
-  {
-    "topic": "SQL Injection",
-    "comment": "Potential SQL injection vulnerability",
-    "confidence": "critical",
-    "reason": "Using string formatting to construct SQL queries can lead to SQL injection attacks",
-    "solution": "Use parameterized queries or an ORM to prevent SQL injection",
-    "actual_code": "query = f\"\"\"SELECT ... FROM{self.table_name}e\"\"\"",
-    "fixed_code": "query = \"\"\"SELECT ... FROM %s e\"\"\" % self.table_name",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 19,
-    "end_line": 37,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 9
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "Missing error handling for database operations",
-    "confidence": "important",
-    "reason": "Database operations can fail due to various reasons, and error handling is necessary to prevent crashes",
-    "solution": "Add try-except blocks to handle database operation errors",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 39,
-    "end_line": 52,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 6
-  },
-  {
-    "topic": "Code Organization",
-    "comment": "AbstractionFeedback class has a single responsibility, but its methods are not well-organized",
-    "confidence": "moderate",
-    "reason": "Well-organized code is easier to read and maintain",
-    "solution": "Consider reorganizing the methods into separate classes or modules",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/retriever/feedback_system.py",
-    "start_line": 4,
-    "end_line": 18,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "The code does not handle potential errors that may occur when connecting to the database or executing queries.",
-    "confidence": "important",
-    "reason": "Error handling is crucial to prevent the program from crashing and to provide meaningful error messages instead.",
-    "solution": "Add try-except blocks to handle potential errors when connecting to the database or executing queries.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 250,
-    "end_line": 270,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 6
-  },
-  {
-    "topic": "Code Duplication",
-    "comment": "The code has duplicated logic for storing code in the database and storing function relationships.",
-    "confidence": "moderate",
-    "reason": "Code duplication makes the code harder to maintain and modify.",
-    "solution": "Extract the duplicated logic into a separate function.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 295,
-    "end_line": 313,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 4
-  },
-  {
-    "topic": "Code Readability",
-    "comment": "The code has long and complex functions that are hard to read and understand.",
-    "confidence": "moderate",
-    "reason": "Long and complex functions make the code harder to maintain and modify.",
-    "solution": "Break down the long and complex functions into smaller and simpler functions.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 100,
-    "end_line": 200,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 5
-  },
-  {
-    "topic": "Unused imports",
-    "comment": "Remove unused imports in tree_sitter_utils.py and test_chunker.py",
-    "confidence": "moderate",
-    "reason": "Unused imports can clutter the codebase and make it harder to maintain",
-    "solution": "Remove the unnecessary imports",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 1,
-    "end_line": 1,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  },
-  {
-    "topic": "Error handling",
-    "comment": "Catch specific exceptions in LanguageLoader.load_language",
-    "confidence": "important",
-    "reason": "Broad exception catching can mask bugs and make debugging harder",
-    "solution": "Catch specific exceptions, such as ImportError or ModuleNotFoundError",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 15,
-    "end_line": 15,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 7
-  },
-  {
-    "topic": "Version bump",
-    "comment": "Verify that the version bump is intentional and follows the project's versioning scheme",
-    "confidence": "critical",
-    "reason": "Inconsistent versioning can cause confusion and break dependencies",
-    "solution": "Verify the version bump and update the project's versioning scheme if necessary",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "pyproject.toml",
-    "start_line": 3,
-    "end_line": 3,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 9
-  },
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Docker",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to Dockerfile, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_name": "Dockerfile",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Docker",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to docker-compose.yml, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "15",
-    "end_line": "15",
-    "side": "RIGHT",
-    "file_name": "docker-compose.yml",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Version Control",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to .gitignore, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "164",
-    "end_line": "164",
-    "side": "RIGHT",
-    "file_name": ".gitignore",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Database",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to db_setup/init.sql, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_name": "db_setup/init.sql",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_222/review.md b/.experiments/code_review/llama-405b/no_eval/pr_222/review.md
deleted file mode 100644
index 6fd1770a..00000000
--- a/.experiments/code_review/llama-405b/no_eval/pr_222/review.md
+++ /dev/null
@@ -1,195 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/222
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 19
-- Critical: 8
-- Important: 4
-- Minor: 7
-- Files Affected: 12
-## 🏆 Code Quality
-[████████████████░░░░] 80% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Error handling (8 issues)</strong></summary>
-
-### 1. Error handling is missing in some critical sections of the code.
-📁 **File:** `:0`
-⚖️ **Severity:** 8/10
-🔍 **Description:** Error handling is crucial for preventing crashes and providing useful error messages.
-💡 **Solution:** Add try-except blocks to handle potential errors.
-
-### 2. Potential SQL injection vulnerability
-📁 **File:** `kaizen/retriever/custom_vector_store.py:19`
-⚖️ **Severity:** 9/10
-🔍 **Description:** Using string formatting to construct SQL queries can lead to SQL injection attacks
-💡 **Solution:** Use parameterized queries or an ORM to prevent SQL injection
-
-**Current Code:**
-```python
-query = f"""SELECT ... FROM{self.table_name}e"""
-```
-
-**Suggested Code:**
-```python
-query = """SELECT ... FROM %s e""" % self.table_name
-```
-
-### 3. Verify that the version bump is intentional and follows the project's versioning scheme
-📁 **File:** `pyproject.toml:3`
-⚖️ **Severity:** 9/10
-🔍 **Description:** Inconsistent versioning can cause confusion and break dependencies
-💡 **Solution:** Verify the version bump and update the project's versioning scheme if necessary
-
-### 4. Changes made to sensitive file
-📁 **File:** `config.json:4`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to config.json, which needs review
-💡 **Solution:** NA
-
-### 5. Changes made to sensitive file
-📁 **File:** `Dockerfile:4`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to Dockerfile, which needs review
-💡 **Solution:** NA
-
-### 6. Changes made to sensitive file
-📁 **File:** `docker-compose.yml:15`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to docker-compose.yml, which needs review
-💡 **Solution:** NA
-
-### 7. Changes made to sensitive file
-📁 **File:** `.gitignore:164`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to .gitignore, which needs review
-💡 **Solution:** NA
-
-### 8. Changes made to sensitive file
-📁 **File:** `db_setup/init.sql:1`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to db_setup/init.sql, which needs review
-💡 **Solution:** NA
-
-</details>
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Code organization (4 issues)</strong></summary>
-
-### 1. The code is well-organized, but some files have too many responsibilities.
-📁 **File:** `:0`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Separation of concerns is crucial for maintainability.
-💡 **Solution:** Consider breaking down large files into smaller ones.
-
-### 2. Missing error handling for database operations
-📁 **File:** `kaizen/retriever/custom_vector_store.py:39`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Database operations can fail due to various reasons, and error handling is necessary to prevent crashes
-💡 **Solution:** Add try-except blocks to handle database operation errors
-
-### 3. The code does not handle potential errors that may occur when connecting to the database or executing queries.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:250`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Error handling is crucial to prevent the program from crashing and to provide meaningful error messages instead.
-💡 **Solution:** Add try-except blocks to handle potential errors when connecting to the database or executing queries.
-
-### 4. Catch specific exceptions in LanguageLoader.load_language
-📁 **File:** `kaizen/retriever/tree_sitter_utils.py:15`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Broad exception catching can mask bugs and make debugging harder
-💡 **Solution:** Catch specific exceptions, such as ImportError or ModuleNotFoundError
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (7 issues)</strong></summary>
-
-<details>
-<summary><strong>Type hints (7 issues)</strong></summary>
-
-### 1. Type hints are missing in some function definitions.
-📁 **File:** `kaizen/retriever/code_chunker.py:7`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Type hints improve code readability and help catch type-related errors.
-💡 **Solution:** Add type hints for function parameters and return types.
-
-**Current Code:**
-```python
-def chunk_code(code: str, language: str) -> ParsedBody:
-```
-
-**Suggested Code:**
-```python
-def chunk_code(code: str, language: str) -> Dict[str, Dict[str, Any]]:
-```
-
-### 2. Some code is duplicated across multiple files.
-📁 **File:** `:0`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Code duplication makes maintenance harder and increases the chance of bugs.
-💡 **Solution:** Extract duplicated code into reusable functions or classes.
-
-### 3. Missing type hints for function return types
-📁 **File:** `kaizen/retriever/custom_vector_store.py:13`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Type hints improve code readability and help catch type-related errors
-💡 **Solution:** Add type hints for function return types
-
-### 4. AbstractionFeedback class has a single responsibility, but its methods are not well-organized
-📁 **File:** `kaizen/retriever/feedback_system.py:4`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Well-organized code is easier to read and maintain
-💡 **Solution:** Consider reorganizing the methods into separate classes or modules
-
-### 5. The code has duplicated logic for storing code in the database and storing function relationships.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:295`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Code duplication makes the code harder to maintain and modify.
-💡 **Solution:** Extract the duplicated logic into a separate function.
-
-### 6. The code has long and complex functions that are hard to read and understand.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:100`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Long and complex functions make the code harder to maintain and modify.
-💡 **Solution:** Break down the long and complex functions into smaller and simpler functions.
-
-### 7. Remove unused imports in tree_sitter_utils.py and test_chunker.py
-📁 **File:** `kaizen/retriever/tree_sitter_utils.py:1`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Unused imports can clutter the codebase and make it harder to maintain
-💡 **Solution:** Remove the unnecessary imports
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (azure_ai/Meta-Llama-3-405B-Instruct)
-{"prompt_tokens": 21487, "completion_tokens": 2711, "total_tokens": 24198}
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_232/issues.json b/.experiments/code_review/llama-405b/no_eval/pr_232/issues.json
deleted file mode 100644
index 1e332158..00000000
--- a/.experiments/code_review/llama-405b/no_eval/pr_232/issues.json
+++ /dev/null
@@ -1,137 +0,0 @@
-[
-  {
-    "topic": "Unused imports",
-    "comment": "There are several unused imports in the code, such as `redirect` in `page.tsx` and `Switch` and `Label` in `queryinput.tsx`. These imports should be removed to declutter the code.",
-    "confidence": "moderate",
-    "reason": "Unused imports can make the code harder to read and maintain.",
-    "solution": "Remove unused imports.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "page.tsx, queryinput.tsx",
-    "start_line": 0,
-    "end_line": 0,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  },
-  {
-    "topic": "Type annotations",
-    "comment": "Some variables and functions are missing type annotations, such as the `initialSpaces` prop in `queryinput.tsx`. Adding type annotations can improve code readability and prevent type-related errors.",
-    "confidence": "moderate",
-    "reason": "Type annotations can help catch type-related errors at compile-time.",
-    "solution": "Add type annotations for variables and functions.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "queryinput.tsx",
-    "start_line": 10,
-    "end_line": 10,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  },
-  {
-    "topic": "Code organization",
-    "comment": "Some files, such as `page.tsx`, contain multiple unrelated components. It would be better to separate these components into their own files to improve code organization and reusability.",
-    "confidence": "low",
-    "reason": "Separating components into their own files can improve code organization and reusability.",
-    "solution": "Separate unrelated components into their own files.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "page.tsx",
-    "start_line": 0,
-    "end_line": 0,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 3
-  },
-  {
-    "topic": "Functionality",
-    "comment": "The code seems to be implementing a memory creation feature with a dialog box. However, there are some potential issues with the functionality.",
-    "confidence": "important",
-    "reason": "The `handleSubmit` function is not properly handling errors. It should be improved to handle errors in a more robust way.",
-    "solution": "Add try-catch blocks to handle errors in the `handleSubmit` function.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "menu.tsx",
-    "start_line": 213,
-    "end_line": 233,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 6
-  },
-  {
-    "topic": "Code organization",
-    "comment": "The code is not well-organized. There are many functions and variables defined inside the `DialogContentContainer` component.",
-    "confidence": "moderate",
-    "reason": "It would be better to separate the concerns of the component into smaller functions or utilities.",
-    "solution": "Extract some of the functions and variables into separate files or utilities.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "menu.tsx",
-    "start_line": 163,
-    "end_line": 346,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Type annotations",
-    "comment": "Some of the variables and functions are missing type annotations.",
-    "confidence": "low",
-    "reason": "Adding type annotations would improve the code quality and make it easier to understand.",
-    "solution": "Add type annotations for the variables and functions.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "menu.tsx",
-    "start_line": 163,
-    "end_line": 346,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 2
-  },
-  {
-    "topic": "Unused imports",
-    "comment": "The import of `useEffect` from 'react' is not used in the code.",
-    "confidence": "moderate",
-    "reason": "The import is not used anywhere in the code.",
-    "solution": "Remove the unused import.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 3,
-    "end_line": 3,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  },
-  {
-    "topic": "Type annotations",
-    "comment": "The type annotation for the `handleInputChange` function is missing.",
-    "confidence": "moderate",
-    "reason": "The function is not annotated with a type.",
-    "solution": "Add the type annotation for the function.",
-    "actual_code": "const handleInputChange = (e: React.ChangeEvent<HTMLInputElement>) =>{",
-    "fixed_code": "const handleInputChange: React.ChangeEventHandler<HTMLInputElement> = (e) =>{",
-    "file_name": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 41,
-    "end_line": 41,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  },
-  {
-    "topic": "Code organization",
-    "comment": "The code is not organized in a logical manner.",
-    "confidence": "low",
-    "reason": "The code is not separated into clear sections or functions.",
-    "solution": "Organize the code into clear sections or functions.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 1,
-    "end_line": 107,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 3
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_232/review.md b/.experiments/code_review/llama-405b/no_eval/pr_232/review.md
deleted file mode 100644
index ef2f635f..00000000
--- a/.experiments/code_review/llama-405b/no_eval/pr_232/review.md
+++ /dev/null
@@ -1,100 +0,0 @@
-PR URL: https://github.com/supermemoryai/supermemory/pull/232
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 9
-- Critical: 0
-- Important: 1
-- Minor: 5
-- Files Affected: 5
-## 🏆 Code Quality
-[████████████████░░░░] 80% (Good)
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Functionality (1 issues)</strong></summary>
-
-### 1. The code seems to be implementing a memory creation feature with a dialog box. However, there are some potential issues with the functionality.
-📁 **File:** `menu.tsx:213`
-⚖️ **Severity:** 6/10
-🔍 **Description:** The `handleSubmit` function is not properly handling errors. It should be improved to handle errors in a more robust way.
-💡 **Solution:** Add try-catch blocks to handle errors in the `handleSubmit` function.
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (8 issues)</strong></summary>
-
-<details>
-<summary><strong>Unused imports (5 issues)</strong></summary>
-
-### 1. There are several unused imports in the code, such as `redirect` in `page.tsx` and `Switch` and `Label` in `queryinput.tsx`. These imports should be removed to declutter the code.
-📁 **File:** `page.tsx, queryinput.tsx:0`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Unused imports can make the code harder to read and maintain.
-💡 **Solution:** Remove unused imports.
-
-### 2. Some variables and functions are missing type annotations, such as the `initialSpaces` prop in `queryinput.tsx`. Adding type annotations can improve code readability and prevent type-related errors.
-📁 **File:** `queryinput.tsx:10`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Type annotations can help catch type-related errors at compile-time.
-💡 **Solution:** Add type annotations for variables and functions.
-
-### 3. The code is not well-organized. There are many functions and variables defined inside the `DialogContentContainer` component.
-📁 **File:** `menu.tsx:163`
-⚖️ **Severity:** 4/10
-🔍 **Description:** It would be better to separate the concerns of the component into smaller functions or utilities.
-💡 **Solution:** Extract some of the functions and variables into separate files or utilities.
-
-### 4. The import of `useEffect` from 'react' is not used in the code.
-📁 **File:** `packages/ui/shadcn/combobox.tsx:3`
-⚖️ **Severity:** 5/10
-🔍 **Description:** The import is not used anywhere in the code.
-💡 **Solution:** Remove the unused import.
-
-### 5. The type annotation for the `handleInputChange` function is missing.
-📁 **File:** `packages/ui/shadcn/combobox.tsx:41`
-⚖️ **Severity:** 5/10
-🔍 **Description:** The function is not annotated with a type.
-💡 **Solution:** Add the type annotation for the function.
-
-**Current Code:**
-```python
-const handleInputChange = (e: React.ChangeEvent<HTMLInputElement>) =>{
-```
-
-**Suggested Code:**
-```python
-const handleInputChange: React.ChangeEventHandler<HTMLInputElement> = (e) =>{
-```
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (azure_ai/Meta-Llama-3-405B-Instruct)
-{"prompt_tokens": 15756, "completion_tokens": 1539, "total_tokens": 17295}
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_252/issues.json b/.experiments/code_review/llama-405b/no_eval/pr_252/issues.json
deleted file mode 100644
index 4a42628f..00000000
--- a/.experiments/code_review/llama-405b/no_eval/pr_252/issues.json
+++ /dev/null
@@ -1,62 +0,0 @@
-[
-  {
-    "topic": "Code Organization",
-    "comment": "The `WorkSummaryGenerator` class has multiple responsibilities, including generating work summaries, Twitter posts, and LinkedIn posts. Consider breaking this down into separate classes or functions for better organization and maintainability.",
-    "confidence": "important",
-    "reason": "Separation of Concerns (SoC) principle",
-    "solution": "Refactor the `WorkSummaryGenerator` class into separate classes or functions for each responsibility.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/reviewer/work_summarizer.py",
-    "start_line": 0,
-    "end_line": 0,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 6
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "The `generate_twitter_post` and `generate_linkedin_post` methods do not handle potential errors that may occur during the generation process. Consider adding try-except blocks to handle and log any exceptions.",
-    "confidence": "important",
-    "reason": "Error handling and logging",
-    "solution": "Add try-except blocks to handle and log any exceptions during the generation process.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/reviewer/work_summarizer.py",
-    "start_line": 58,
-    "end_line": 74,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 7
-  },
-  {
-    "topic": "Code Style",
-    "comment": "The `kaizen/llms/prompts/code_review_prompts.py` file has inconsistent indentation. Consider using a consistent number of spaces for indentation throughout the file.",
-    "confidence": "moderate",
-    "reason": "Code style and readability",
-    "solution": "Use a consistent number of spaces for indentation throughout the file.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/llms/prompts/code_review_prompts.py",
-    "start_line": 0,
-    "end_line": 0,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Code Duplication",
-    "comment": "The `generate_twitter_post` and `generate_linkedin_post` methods have similar code structures. Consider extracting a common method to avoid code duplication.",
-    "confidence": "moderate",
-    "reason": "Don't Repeat Yourself (DRY) principle",
-    "solution": "Extract a common method to avoid code duplication.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/reviewer/work_summarizer.py",
-    "start_line": 58,
-    "end_line": 74,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_252/review.md b/.experiments/code_review/llama-405b/no_eval/pr_252/review.md
deleted file mode 100644
index 732aa03c..00000000
--- a/.experiments/code_review/llama-405b/no_eval/pr_252/review.md
+++ /dev/null
@@ -1,78 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/252
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 4
-- Critical: 0
-- Important: 2
-- Minor: 2
-- Files Affected: 2
-## 🏆 Code Quality
-[████████████████░░░░] 80% (Good)
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Code Organization (2 issues)</strong></summary>
-
-### 1. The `WorkSummaryGenerator` class has multiple responsibilities, including generating work summaries, Twitter posts, and LinkedIn posts. Consider breaking this down into separate classes or functions for better organization and maintainability.
-📁 **File:** `kaizen/reviewer/work_summarizer.py:0`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Separation of Concerns (SoC) principle
-💡 **Solution:** Refactor the `WorkSummaryGenerator` class into separate classes or functions for each responsibility.
-
-### 2. The `generate_twitter_post` and `generate_linkedin_post` methods do not handle potential errors that may occur during the generation process. Consider adding try-except blocks to handle and log any exceptions.
-📁 **File:** `kaizen/reviewer/work_summarizer.py:58`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Error handling and logging
-💡 **Solution:** Add try-except blocks to handle and log any exceptions during the generation process.
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (2 issues)</strong></summary>
-
-<details>
-<summary><strong>Code Style (2 issues)</strong></summary>
-
-### 1. The `kaizen/llms/prompts/code_review_prompts.py` file has inconsistent indentation. Consider using a consistent number of spaces for indentation throughout the file.
-📁 **File:** `kaizen/llms/prompts/code_review_prompts.py:0`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Code style and readability
-💡 **Solution:** Use a consistent number of spaces for indentation throughout the file.
-
-### 2. The `generate_twitter_post` and `generate_linkedin_post` methods have similar code structures. Consider extracting a common method to avoid code duplication.
-📁 **File:** `kaizen/reviewer/work_summarizer.py:58`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Don't Repeat Yourself (DRY) principle
-💡 **Solution:** Extract a common method to avoid code duplication.
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (azure_ai/Meta-Llama-3-405B-Instruct)
-{"prompt_tokens": 3881, "completion_tokens": 757, "total_tokens": 4638}
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_335/issues.json b/.experiments/code_review/llama-405b/no_eval/pr_335/issues.json
deleted file mode 100644
index 5b84d52d..00000000
--- a/.experiments/code_review/llama-405b/no_eval/pr_335/issues.json
+++ /dev/null
@@ -1,47 +0,0 @@
-[
-  {
-    "topic": "Import Statements",
-    "comment": "Unused import statements are present in the code.",
-    "confidence": "moderate",
-    "reason": "The import statements for 'output' and 'parser' are not used anywhere in the code.",
-    "solution": "Remove unused import statements to declutter the code.",
-    "actual_code": "from kaizen.helpers import output, parser",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/pr_description.py",
-    "start_line": 6,
-    "end_line": 6,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 5
-  },
-  {
-    "topic": "Function Definition",
-    "comment": "The function '__init__' has an unused parameter 'llm_provider'.",
-    "confidence": "high",
-    "reason": "The parameter 'llm_provider' is not used anywhere in the function.",
-    "solution": "Remove unused function parameters to simplify the code.",
-    "actual_code": "def __init__(self, llm_provider: LLMProvider):",
-    "fixed_code": "def __init__(self):",
-    "file_name": "kaizen/generator/pr_description.py",
-    "start_line": 26,
-    "end_line": 26,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 6
-  },
-  {
-    "topic": "Variable Naming",
-    "comment": "Variable names are not following PEP 8 conventions.",
-    "confidence": "low",
-    "reason": "Variable names should be in lowercase with words separated by underscores.",
-    "solution": "Rename variables to follow PEP 8 conventions.",
-    "actual_code": "PULL_REQUEST_TITLE ={PULL_REQUEST_TITLE}",
-    "fixed_code": "pull_request_title ={pull_request_title}",
-    "file_name": "kaizen/llms/prompts/pr_desc_prompts.py",
-    "start_line": 27,
-    "end_line": 27,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 3
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_400/comments.json b/.experiments/code_review/llama-405b/no_eval/pr_400/comments.json
deleted file mode 100644
index 33197e40..00000000
--- a/.experiments/code_review/llama-405b/no_eval/pr_400/comments.json
+++ /dev/null
@@ -1,16 +0,0 @@
-[
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "11",
-    "end_line": "11",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_400/issues.json b/.experiments/code_review/llama-405b/no_eval/pr_400/issues.json
deleted file mode 100644
index 318f90b8..00000000
--- a/.experiments/code_review/llama-405b/no_eval/pr_400/issues.json
+++ /dev/null
@@ -1,316 +0,0 @@
-[
-  {
-    "topic": "Error Handling",
-    "comment": "The `create_pr_description` function does not handle cases where `desc` or `original_desc` are not strings.",
-    "confidence": "important",
-    "reason": "The function should check the type of the input parameters and raise a meaningful error if they are not strings.",
-    "solution": "Add type checking for `desc` and `original_desc` and raise a `TypeError` if they are not strings.",
-    "actual_code": "",
-    "fixed_code": "if not isinstance(desc, str) or not isinstance(original_desc, str):\n    raise TypeError('desc and original_desc must be strings')",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
-    "start_line": 31,
-    "end_line": 31,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 6
-  },
-  {
-    "topic": "Performance",
-    "comment": "The `create_pr_description` function may have performance issues for large input strings.",
-    "confidence": "moderate",
-    "reason": "The function uses string concatenation, which can be inefficient for large strings.",
-    "solution": "Consider using a more efficient string concatenation method, such as using a list and joining the strings at the end.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
-    "start_line": 51,
-    "end_line": 51,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Code Organization",
-    "comment": "The test file has a mix of test cases and helper functions.",
-    "confidence": "moderate",
-    "reason": "It's better to separate test cases and helper functions into different files or modules.",
-    "solution": "Consider moving the helper functions to a separate file or module.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
-    "start_line": 1,
-    "end_line": 1,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 3
-  },
-  {
-    "topic": "Code Duplication",
-    "comment": "The `PR_COLLAPSIBLE_TEMPLATE` is duplicated in multiple tests.",
-    "confidence": "high",
-    "reason": "Code duplication can make maintenance harder.",
-    "solution": "Extract the template into a separate constant or function.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 0,
-    "end_line": 0,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 6
-  },
-  {
-    "topic": "Test Coverage",
-    "comment": "There are no tests for the `create_pr_review_text` function with an empty input.",
-    "confidence": "medium",
-    "reason": "Empty input can cause unexpected behavior.",
-    "solution": "Add a test for the `create_pr_review_text` function with an empty input.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 0,
-    "end_line": 0,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 5
-  },
-  {
-    "topic": "Code Readability",
-    "comment": "The `test_create_pr_review_text_mixed_reviews` test has a long and complex expected output.",
-    "confidence": "low",
-    "reason": "Long and complex expected outputs can make tests harder to understand.",
-    "solution": "Consider breaking the expected output into smaller, more manageable parts.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 96,
-    "end_line": 172,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 3
-  },
-  {
-    "topic": "Functionality",
-    "comment": "The code seems to be implementing the required functionality correctly.",
-    "confidence": "moderate",
-    "reason": "The code is using the correct libraries and functions to achieve the desired outcome.",
-    "solution": "",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 1,
-    "end_line": 247,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 5
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "The code is missing error handling for potential exceptions that may occur during file operations.",
-    "confidence": "high",
-    "reason": "The code is not handling potential exceptions that may occur during file operations, which can lead to unexpected behavior.",
-    "solution": "Add try-except blocks to handle potential exceptions during file operations.",
-    "actual_code": "",
-    "fixed_code": "try:\n    with open(file_path, 'r') as f:\n        content = f.read()\nexcept FileNotFoundError:\n    print(f'File{file_path}not found.')",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 40,
-    "end_line": 45,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Code Quality",
-    "comment": "The code has some redundant comments and docstrings that can be removed.",
-    "confidence": "low",
-    "reason": "The code has some redundant comments and docstrings that are not providing any additional information.",
-    "solution": "Remove redundant comments and docstrings.",
-    "actual_code": "# Correct implementation of get_parent_folder()",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py",
-    "start_line": 8,
-    "end_line": 8,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 2
-  },
-  {
-    "topic": "Import Statements",
-    "comment": "Unused import statement",
-    "confidence": "moderate",
-    "reason": "The import statement 'from kaizen.helpers.output import get_web_html' is not used in the code.",
-    "solution": "Remove the unused import statement.",
-    "actual_code": "from kaizen.helpers.output import get_web_html",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 4,
-    "end_line": 4,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 5
-  },
-  {
-    "topic": "Function Definition",
-    "comment": "Function 'test_get_web_html_normal_cases' is too long and complex.",
-    "confidence": "high",
-    "reason": "The function has too many lines of code and is difficult to understand.",
-    "solution": "Break down the function into smaller, more manageable functions.",
-    "actual_code": "async def test_get_web_html_normal_cases(mock_get_html, mock_nest_asyncio, html_content, expected_output):",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 76,
-    "end_line": 103,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "Error handling is missing in function 'test_get_web_html_invalid_url'.",
-    "confidence": "high",
-    "reason": "The function does not handle potential errors that may occur.",
-    "solution": "Add try-except blocks to handle potential errors.",
-    "actual_code": "async def test_get_web_html_invalid_url(mock_get_html, mock_nest_asyncio):",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 85,
-    "end_line": 91,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 9
-  },
-  {
-    "topic": "Code Duplication",
-    "comment": "Code duplication in functions 'test_get_web_html_normal_cases' and 'test_get_web_html_invalid_url'.",
-    "confidence": "moderate",
-    "reason": "The functions have similar code that can be extracted into a separate function.",
-    "solution": "Extract the common code into a separate function.",
-    "actual_code": "mock_get_html.return_value = html_content",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 77,
-    "end_line": 77,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 6
-  },
-  {
-    "topic": "Import organization",
-    "comment": "Imports are not organized alphabetically.",
-    "confidence": "moderate",
-    "reason": "Following PEP 8 guidelines for import organization improves readability.",
-    "solution": "Organize imports alphabetically.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/unit_test.py",
-    "start_line": 1,
-    "end_line": 10,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 3
-  },
-  {
-    "topic": "Docstrings",
-    "comment": "Missing docstrings for classes and methods.",
-    "confidence": "important",
-    "reason": "Docstrings provide essential documentation for users and maintainers.",
-    "solution": "Add docstrings to classes and methods.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/unit_test.py",
-    "start_line": 11,
-    "end_line": 275,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 6
-  },
-  {
-    "topic": "Type hints",
-    "comment": "Missing type hints for method parameters and return types.",
-    "confidence": "important",
-    "reason": "Type hints improve code readability and enable static type checking.",
-    "solution": "Add type hints for method parameters and return types.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/unit_test.py",
-    "start_line": 11,
-    "end_line": 275,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 6
-  },
-  {
-    "topic": "Error handling",
-    "comment": "Insufficient error handling in methods.",
-    "confidence": "important",
-    "reason": "Proper error handling ensures the program remains stable and provides useful error messages.",
-    "solution": "Implement try-except blocks to handle potential errors.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/unit_test.py",
-    "start_line": 11,
-    "end_line": 275,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Code organization",
-    "comment": "Some methods are too long and complex.",
-    "confidence": "moderate",
-    "reason": "Breaking down long methods into smaller ones improves readability and maintainability.",
-    "solution": "Refactor long methods into smaller, more focused ones.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/unit_test.py",
-    "start_line": 11,
-    "end_line": 275,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Logging Configuration",
-    "comment": "The logging configuration is not properly set up.",
-    "confidence": "important",
-    "reason": "The logging level is set to ERROR for all loggers, but the LOGLEVEL environment variable is set to INFO.",
-    "solution": "Set the logging level consistently throughout the application.",
-    "actual_code": "set_all_loggers_to_ERROR()",
-    "fixed_code": "set_all_loggers_to_INFO()",
-    "file_name": "kaizen/llms/provider.py",
-    "start_line": 13,
-    "end_line": 28,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 6
-  },
-  {
-    "topic": "Code Organization",
-    "comment": "The code is not properly organized.",
-    "confidence": "moderate",
-    "reason": "The set_all_loggers_to_ERROR function is defined in the middle of the file.",
-    "solution": "Move the function definition to the top of the file.",
-    "actual_code": "def set_all_loggers_to_ERROR():",
-    "fixed_code": "def set_all_loggers_to_ERROR():\n    # ...",
-    "file_name": "kaizen/llms/provider.py",
-    "start_line": 13,
-    "end_line": 28,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 4
-  },
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "11",
-    "end_line": "11",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_400/review.md b/.experiments/code_review/llama-405b/no_eval/pr_400/review.md
deleted file mode 100644
index c168419b..00000000
--- a/.experiments/code_review/llama-405b/no_eval/pr_400/review.md
+++ /dev/null
@@ -1,197 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/400
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 21
-- Critical: 1
-- Important: 5
-- Minor: 8
-- Files Affected: 8
-## 🏆 Code Quality
-[████████████████░░░░] 80% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Configuration (1 issues)</strong></summary>
-
-### 1. Changes made to sensitive file
-📁 **File:** `config.json:11`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to config.json, which needs review
-💡 **Solution:** NA
-
-</details>
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Error Handling (5 issues)</strong></summary>
-
-### 1. The `create_pr_description` function does not handle cases where `desc` or `original_desc` are not strings.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_description.py:31`
-⚖️ **Severity:** 6/10
-🔍 **Description:** The function should check the type of the input parameters and raise a meaningful error if they are not strings.
-💡 **Solution:** Add type checking for `desc` and `original_desc` and raise a `TypeError` if they are not strings.
-
-**Current Code:**
-```python
-
-```
-
-**Suggested Code:**
-```python
-if not isinstance(desc, str) or not isinstance(original_desc, str):
-    raise TypeError('desc and original_desc must be strings')
-```
-
-### 2. Missing docstrings for classes and methods.
-📁 **File:** `kaizen/generator/unit_test.py:11`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Docstrings provide essential documentation for users and maintainers.
-💡 **Solution:** Add docstrings to classes and methods.
-
-### 3. Missing type hints for method parameters and return types.
-📁 **File:** `kaizen/generator/unit_test.py:11`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Type hints improve code readability and enable static type checking.
-💡 **Solution:** Add type hints for method parameters and return types.
-
-### 4. Insufficient error handling in methods.
-📁 **File:** `kaizen/generator/unit_test.py:11`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Proper error handling ensures the program remains stable and provides useful error messages.
-💡 **Solution:** Implement try-except blocks to handle potential errors.
-
-### 5. The logging configuration is not properly set up.
-📁 **File:** `kaizen/llms/provider.py:13`
-⚖️ **Severity:** 6/10
-🔍 **Description:** The logging level is set to ERROR for all loggers, but the LOGLEVEL environment variable is set to INFO.
-💡 **Solution:** Set the logging level consistently throughout the application.
-
-**Current Code:**
-```python
-set_all_loggers_to_ERROR()
-```
-
-**Suggested Code:**
-```python
-set_all_loggers_to_INFO()
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (10 issues)</strong></summary>
-
-<details>
-<summary><strong>Performance (8 issues)</strong></summary>
-
-### 1. The `create_pr_description` function may have performance issues for large input strings.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_description.py:51`
-⚖️ **Severity:** 4/10
-🔍 **Description:** The function uses string concatenation, which can be inefficient for large strings.
-💡 **Solution:** Consider using a more efficient string concatenation method, such as using a list and joining the strings at the end.
-
-### 2. The test file has a mix of test cases and helper functions.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_description.py:1`
-⚖️ **Severity:** 3/10
-🔍 **Description:** It's better to separate test cases and helper functions into different files or modules.
-💡 **Solution:** Consider moving the helper functions to a separate file or module.
-
-### 3. The code seems to be implementing the required functionality correctly.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_test_files.py:1`
-⚖️ **Severity:** 5/10
-🔍 **Description:** The code is using the correct libraries and functions to achieve the desired outcome.
-💡 **Solution:** 
-
-### 4. Unused import statement
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_web_html.py:4`
-⚖️ **Severity:** 5/10
-🔍 **Description:** The import statement 'from kaizen.helpers.output import get_web_html' is not used in the code.
-💡 **Solution:** Remove the unused import statement.
-
-**Current Code:**
-```python
-from kaizen.helpers.output import get_web_html
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 5. Code duplication in functions 'test_get_web_html_normal_cases' and 'test_get_web_html_invalid_url'.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_web_html.py:77`
-⚖️ **Severity:** 6/10
-🔍 **Description:** The functions have similar code that can be extracted into a separate function.
-💡 **Solution:** Extract the common code into a separate function.
-
-**Current Code:**
-```python
-mock_get_html.return_value = html_content
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 6. Imports are not organized alphabetically.
-📁 **File:** `kaizen/generator/unit_test.py:1`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Following PEP 8 guidelines for import organization improves readability.
-💡 **Solution:** Organize imports alphabetically.
-
-### 7. Some methods are too long and complex.
-📁 **File:** `kaizen/generator/unit_test.py:11`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Breaking down long methods into smaller ones improves readability and maintainability.
-💡 **Solution:** Refactor long methods into smaller, more focused ones.
-
-### 8. The code is not properly organized.
-📁 **File:** `kaizen/llms/provider.py:13`
-⚖️ **Severity:** 4/10
-🔍 **Description:** The set_all_loggers_to_ERROR function is defined in the middle of the file.
-💡 **Solution:** Move the function definition to the top of the file.
-
-**Current Code:**
-```python
-def set_all_loggers_to_ERROR():
-```
-
-**Suggested Code:**
-```python
-def set_all_loggers_to_ERROR():
-    # ...
-```
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (azure_ai/Meta-Llama-3-405B-Instruct)
-{"prompt_tokens": 36841, "completion_tokens": 3640, "total_tokens": 40481}
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_440/issues.json b/.experiments/code_review/llama-405b/no_eval/pr_440/issues.json
deleted file mode 100644
index 172e9fbf..00000000
--- a/.experiments/code_review/llama-405b/no_eval/pr_440/issues.json
+++ /dev/null
@@ -1,47 +0,0 @@
-[
-  {
-    "topic": "Unnecessary Comment",
-    "comment": "The comment 'TODO: DONT PUSH DUPLICATE' is unnecessary and should be removed.",
-    "confidence": "moderate",
-    "reason": "The comment does not provide any useful information and is not relevant to the code.",
-    "solution": "Remove the comment.",
-    "actual_code": "# TODO: DONT PUSH DUPLICATE",
-    "fixed_code": "",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 157,
-    "end_line": 157,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 2
-  },
-  {
-    "topic": "Dependency Update",
-    "comment": "The dependency 'llama-index-core' has been updated to version '0.10.65'.",
-    "confidence": "important",
-    "reason": "The updated version may include security patches or bug fixes.",
-    "solution": "Review the changelog for the updated version to ensure compatibility with the current codebase.",
-    "actual_code": "llama-index-core = \"0.10.65\"",
-    "fixed_code": "",
-    "file_name": "pyproject.toml",
-    "start_line": 27,
-    "end_line": 27,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 5
-  },
-  {
-    "topic": "Removed Dependencies",
-    "comment": "The dependencies 'llama-index-llms-openai' and 'llama-index-core' (version '^0.10.47') have been removed.",
-    "confidence": "important",
-    "reason": "The removed dependencies may be required by other parts of the codebase.",
-    "solution": "Review the codebase to ensure that the removed dependencies are not required.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "pyproject.toml",
-    "start_line": 27,
-    "end_line": 28,
-    "side": "LEFT",
-    "sentiment": "negative",
-    "severity_level": 6
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_440/review.md b/.experiments/code_review/llama-405b/no_eval/pr_440/review.md
deleted file mode 100644
index f509ad53..00000000
--- a/.experiments/code_review/llama-405b/no_eval/pr_440/review.md
+++ /dev/null
@@ -1,92 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/440
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 3
-- Critical: 0
-- Important: 2
-- Minor: 1
-- Files Affected: 2
-## 🏆 Code Quality
-[████████████████░░░░] 80% (Good)
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Dependency Update (2 issues)</strong></summary>
-
-### 1. The dependency 'llama-index-core' has been updated to version '0.10.65'.
-📁 **File:** `pyproject.toml:27`
-⚖️ **Severity:** 5/10
-🔍 **Description:** The updated version may include security patches or bug fixes.
-💡 **Solution:** Review the changelog for the updated version to ensure compatibility with the current codebase.
-
-**Current Code:**
-```python
-llama-index-core = "0.10.65"
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 2. The dependencies 'llama-index-llms-openai' and 'llama-index-core' (version '^0.10.47') have been removed.
-📁 **File:** `pyproject.toml:27`
-⚖️ **Severity:** 6/10
-🔍 **Description:** The removed dependencies may be required by other parts of the codebase.
-💡 **Solution:** Review the codebase to ensure that the removed dependencies are not required.
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (1 issues)</strong></summary>
-
-<details>
-<summary><strong>Unnecessary Comment (1 issues)</strong></summary>
-
-### 1. The comment 'TODO: DONT PUSH DUPLICATE' is unnecessary and should be removed.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:157`
-⚖️ **Severity:** 2/10
-🔍 **Description:** The comment does not provide any useful information and is not relevant to the code.
-💡 **Solution:** Remove the comment.
-
-**Current Code:**
-```python
-# TODO: DONT PUSH DUPLICATE
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (azure_ai/Meta-Llama-3-405B-Instruct)
-{"prompt_tokens": 1362, "completion_tokens": 562, "total_tokens": 1924}
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_476/comments.json b/.experiments/code_review/llama-405b/no_eval/pr_476/comments.json
deleted file mode 100644
index a9d40eac..00000000
--- a/.experiments/code_review/llama-405b/no_eval/pr_476/comments.json
+++ /dev/null
@@ -1,16 +0,0 @@
-[
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_476/issues.json b/.experiments/code_review/llama-405b/no_eval/pr_476/issues.json
deleted file mode 100644
index a394f97b..00000000
--- a/.experiments/code_review/llama-405b/no_eval/pr_476/issues.json
+++ /dev/null
@@ -1,61 +0,0 @@
-[
-  {
-    "topic": "Error Handling",
-    "comment": "Broad exception handling can mask bugs and make debugging difficult.",
-    "confidence": "important",
-    "reason": "The `except Exception` block in `github_app/github_helper/pull_requests.py` (line 140) catches all exceptions, which can make it challenging to identify and fix specific issues.",
-    "solution": "Catch specific exceptions that can occur during the execution of the code, and provide meaningful error messages to aid in debugging.",
-    "actual_code": "except Exception:",
-    "fixed_code": "except requests.exceptions.RequestException as e:",
-    "file_name": "github_app/github_helper/pull_requests.py",
-    "start_line": 140,
-    "end_line": 141,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 6
-  },
-  {
-    "topic": "Code Organization",
-    "comment": "The `sort_files` function is not necessary and can be replaced with a built-in sorting function.",
-    "confidence": "moderate",
-    "reason": "The `sort_files` function in `github_app/github_helper/pull_requests.py` (line 184) is not necessary and can be replaced with the built-in `sorted` function.",
-    "solution": "Use the built-in `sorted` function to sort the files, which is more efficient and Pythonic.",
-    "actual_code": "def sort_files(files):",
-    "fixed_code": "sorted_files = sorted(files, key=lambda x: x['filename'])",
-    "file_name": "github_app/github_helper/pull_requests.py",
-    "start_line": 184,
-    "end_line": 194,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Code Quality",
-    "comment": "The `generate_tests` function is not necessary and can be replaced with a list comprehension.",
-    "confidence": "moderate",
-    "reason": "The `generate_tests` function in `github_app/github_helper/pull_requests.py` (line 199) is not necessary and can be replaced with a list comprehension.",
-    "solution": "Use a list comprehension to generate the tests, which is more efficient and Pythonic.",
-    "actual_code": "def generate_tests(pr_files):",
-    "fixed_code": "tests =[f['filename'] for f in pr_files]",
-    "file_name": "github_app/github_helper/pull_requests.py",
-    "start_line": 199,
-    "end_line": 200,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_476/review.md b/.experiments/code_review/llama-405b/no_eval/pr_476/review.md
deleted file mode 100644
index 8f093042..00000000
--- a/.experiments/code_review/llama-405b/no_eval/pr_476/review.md
+++ /dev/null
@@ -1,115 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/476
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 4
-- Critical: 1
-- Important: 1
-- Minor: 2
-- Files Affected: 2
-## 🏆 Code Quality
-[████████████████░░░░] 80% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Configuration (1 issues)</strong></summary>
-
-### 1. Changes made to sensitive file
-📁 **File:** `config.json:1`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to config.json, which needs review
-💡 **Solution:** NA
-
-</details>
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Error Handling (1 issues)</strong></summary>
-
-### 1. Broad exception handling can mask bugs and make debugging difficult.
-📁 **File:** `github_app/github_helper/pull_requests.py:140`
-⚖️ **Severity:** 6/10
-🔍 **Description:** The `except Exception` block in `github_app/github_helper/pull_requests.py` (line 140) catches all exceptions, which can make it challenging to identify and fix specific issues.
-💡 **Solution:** Catch specific exceptions that can occur during the execution of the code, and provide meaningful error messages to aid in debugging.
-
-**Current Code:**
-```python
-except Exception:
-```
-
-**Suggested Code:**
-```python
-except requests.exceptions.RequestException as e:
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (2 issues)</strong></summary>
-
-<details>
-<summary><strong>Code Organization (2 issues)</strong></summary>
-
-### 1. The `sort_files` function is not necessary and can be replaced with a built-in sorting function.
-📁 **File:** `github_app/github_helper/pull_requests.py:184`
-⚖️ **Severity:** 4/10
-🔍 **Description:** The `sort_files` function in `github_app/github_helper/pull_requests.py` (line 184) is not necessary and can be replaced with the built-in `sorted` function.
-💡 **Solution:** Use the built-in `sorted` function to sort the files, which is more efficient and Pythonic.
-
-**Current Code:**
-```python
-def sort_files(files):
-```
-
-**Suggested Code:**
-```python
-sorted_files = sorted(files, key=lambda x: x['filename'])
-```
-
-### 2. The `generate_tests` function is not necessary and can be replaced with a list comprehension.
-📁 **File:** `github_app/github_helper/pull_requests.py:199`
-⚖️ **Severity:** 4/10
-🔍 **Description:** The `generate_tests` function in `github_app/github_helper/pull_requests.py` (line 199) is not necessary and can be replaced with a list comprehension.
-💡 **Solution:** Use a list comprehension to generate the tests, which is more efficient and Pythonic.
-
-**Current Code:**
-```python
-def generate_tests(pr_files):
-```
-
-**Suggested Code:**
-```python
-tests =[f['filename'] for f in pr_files]
-```
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (azure_ai/Meta-Llama-3-405B-Instruct)
-{"prompt_tokens": 4006, "completion_tokens": 655, "total_tokens": 4661}
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_5/comments.json b/.experiments/code_review/llama-405b/no_eval/pr_5/comments.json
deleted file mode 100644
index ce61c328..00000000
--- a/.experiments/code_review/llama-405b/no_eval/pr_5/comments.json
+++ /dev/null
@@ -1,47 +0,0 @@
-[
-  {
-    "topic": "API Call Failure",
-    "comment": "API call failure without retry mechanism.",
-    "confidence": "critical",
-    "reason": "The API call may fail without a retry mechanism, causing the program to crash.",
-    "solution": "Implement a retry mechanism for the API call.",
-    "actual_code": "response = completion(model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages)",
-    "fixed_code": "import time\ntry:\n    response = completion(model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages)\nexcept Exception as e:\n    print(f\"API call failed:{e}\")\n    time.sleep(1)\n    # retry the API call",
-    "file_name": "main.py",
-    "start_line": 65,
-    "end_line": 65,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Silent Failure",
-    "comment": "Silent failure without logging.",
-    "confidence": "critical",
-    "reason": "The program will fail silently without logging any errors.",
-    "solution": "Implement logging for errors.",
-    "actual_code": "except json.JSONDecodeError:\n    result ={key: \"\" for key in[\"feedback\", \"review\", \"should_interview\", \"rating\", \"input_tokens\", \"output_tokens\"]}",
-    "fixed_code": "import logging\nexcept json.JSONDecodeError as e:\n    logging.error(f\"JSON decode error:{e}\")\n    result ={key: \"\" for key in[\"feedback\", \"review\", \"should_interview\", \"rating\", \"input_tokens\", \"output_tokens\"]}",
-    "file_name": "main.py",
-    "start_line": 82,
-    "end_line": 84,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Division by Zero",
-    "comment": "Division by zero potential.",
-    "confidence": "critical",
-    "reason": "The code may divide by zero, causing a runtime error.",
-    "solution": "Add a check to prevent division by zero.",
-    "actual_code": "total_tokens = total_input_tokens + total_output_tokens",
-    "fixed_code": "if total_tokens == 0:\n    total_tokens = 1\n    print(\"Warning: total tokens is zero\")",
-    "file_name": "main.py",
-    "start_line": 156,
-    "end_line": 156,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 8
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_5/issues.json b/.experiments/code_review/llama-405b/no_eval/pr_5/issues.json
deleted file mode 100644
index a33e8f2e..00000000
--- a/.experiments/code_review/llama-405b/no_eval/pr_5/issues.json
+++ /dev/null
@@ -1,107 +0,0 @@
-[
-  {
-    "topic": "Unused Import",
-    "comment": "Unused import 'random' should be removed.",
-    "confidence": "moderate",
-    "reason": "The import 'random' is not used anywhere in the code.",
-    "solution": "Remove the line 'import random'.",
-    "actual_code": "import random",
-    "fixed_code": "",
-    "file_name": "main.py",
-    "start_line": 8,
-    "end_line": 8,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 3
-  },
-  {
-    "topic": "API Call Failure",
-    "comment": "API call failure without retry mechanism.",
-    "confidence": "critical",
-    "reason": "The API call may fail without a retry mechanism, causing the program to crash.",
-    "solution": "Implement a retry mechanism for the API call.",
-    "actual_code": "response = completion(model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages)",
-    "fixed_code": "import time\ntry:\n    response = completion(model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages)\nexcept Exception as e:\n    print(f\"API call failed:{e}\")\n    time.sleep(1)\n    # retry the API call",
-    "file_name": "main.py",
-    "start_line": 65,
-    "end_line": 65,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Silent Failure",
-    "comment": "Silent failure without logging.",
-    "confidence": "critical",
-    "reason": "The program will fail silently without logging any errors.",
-    "solution": "Implement logging for errors.",
-    "actual_code": "except json.JSONDecodeError:\n    result ={key: \"\" for key in[\"feedback\", \"review\", \"should_interview\", \"rating\", \"input_tokens\", \"output_tokens\"]}",
-    "fixed_code": "import logging\nexcept json.JSONDecodeError as e:\n    logging.error(f\"JSON decode error:{e}\")\n    result ={key: \"\" for key in[\"feedback\", \"review\", \"should_interview\", \"rating\", \"input_tokens\", \"output_tokens\"]}",
-    "file_name": "main.py",
-    "start_line": 82,
-    "end_line": 84,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Inefficient Progress Printing",
-    "comment": "Inefficient way to print progress.",
-    "confidence": "important",
-    "reason": "The progress printing is inefficient and may cause performance issues.",
-    "solution": "Use a more efficient way to print progress, such as using a progress bar library.",
-    "actual_code": "print(f\"\\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}\", end=\"\", flush=True)",
-    "fixed_code": "from tqdm import tqdm\nwith tqdm(total=total, desc=\"Processing applicants\") as pbar:\n    for index, row in df.iterrows():\n        # process applicant\n        pbar.update(1)",
-    "file_name": "main.py",
-    "start_line": 121,
-    "end_line": 121,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 5
-  },
-  {
-    "topic": "Redundant Code",
-    "comment": "Redundant code.",
-    "confidence": "moderate",
-    "reason": "The code is redundant and can be removed.",
-    "solution": "Remove the redundant code.",
-    "actual_code": "if len(df) == 0:\n    return",
-    "fixed_code": "",
-    "file_name": "main.py",
-    "start_line": 141,
-    "end_line": 142,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 3
-  },
-  {
-    "topic": "Division by Zero",
-    "comment": "Division by zero potential.",
-    "confidence": "critical",
-    "reason": "The code may divide by zero, causing a runtime error.",
-    "solution": "Add a check to prevent division by zero.",
-    "actual_code": "total_tokens = total_input_tokens + total_output_tokens",
-    "fixed_code": "if total_tokens == 0:\n    total_tokens = 1\n    print(\"Warning: total tokens is zero\")",
-    "file_name": "main.py",
-    "start_line": 156,
-    "end_line": 156,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "File Not Found",
-    "comment": "No error handling for file not found.",
-    "confidence": "important",
-    "reason": "The code does not handle the case where the file is not found.",
-    "solution": "Add error handling for file not found.",
-    "actual_code": "main(input_file)",
-    "fixed_code": "try:\n    main(input_file)\nexcept FileNotFoundError:\n    print(f\"Error: file '{input_file}' not found\")",
-    "file_name": "main.py",
-    "start_line": 174,
-    "end_line": 174,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 5
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/llama-405b/no_eval/pr_5/review.md b/.experiments/code_review/llama-405b/no_eval/pr_5/review.md
deleted file mode 100644
index adfbba95..00000000
--- a/.experiments/code_review/llama-405b/no_eval/pr_5/review.md
+++ /dev/null
@@ -1,193 +0,0 @@
-PR URL: https://github.com/sauravpanda/applicant-screening/pull/5
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 7
-- Critical: 3
-- Important: 2
-- Minor: 2
-- Files Affected: 1
-## 🏆 Code Quality
-[██████████████░░░░░░] 70% (Fair)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>API Call Failure (3 issues)</strong></summary>
-
-### 1. API call failure without retry mechanism.
-📁 **File:** `main.py:65`
-⚖️ **Severity:** 8/10
-🔍 **Description:** The API call may fail without a retry mechanism, causing the program to crash.
-💡 **Solution:** Implement a retry mechanism for the API call.
-
-**Current Code:**
-```python
-response = completion(model=os.environ.get("model", "anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1"), messages=messages)
-```
-
-**Suggested Code:**
-```python
-import time
-try:
-    response = completion(model=os.environ.get("model", "anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1"), messages=messages)
-except Exception as e:
-    print(f"API call failed:{e}")
-    time.sleep(1)
-    # retry the API call
-```
-
-### 2. Silent failure without logging.
-📁 **File:** `main.py:82`
-⚖️ **Severity:** 8/10
-🔍 **Description:** The program will fail silently without logging any errors.
-💡 **Solution:** Implement logging for errors.
-
-**Current Code:**
-```python
-except json.JSONDecodeError:
-    result ={key: "" for key in["feedback", "review", "should_interview", "rating", "input_tokens", "output_tokens"]}
-```
-
-**Suggested Code:**
-```python
-import logging
-except json.JSONDecodeError as e:
-    logging.error(f"JSON decode error:{e}")
-    result ={key: "" for key in["feedback", "review", "should_interview", "rating", "input_tokens", "output_tokens"]}
-```
-
-### 3. Division by zero potential.
-📁 **File:** `main.py:156`
-⚖️ **Severity:** 8/10
-🔍 **Description:** The code may divide by zero, causing a runtime error.
-💡 **Solution:** Add a check to prevent division by zero.
-
-**Current Code:**
-```python
-total_tokens = total_input_tokens + total_output_tokens
-```
-
-**Suggested Code:**
-```python
-if total_tokens == 0:
-    total_tokens = 1
-    print("Warning: total tokens is zero")
-```
-
-</details>
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Inefficient Progress Printing (2 issues)</strong></summary>
-
-### 1. Inefficient way to print progress.
-📁 **File:** `main.py:121`
-⚖️ **Severity:** 5/10
-🔍 **Description:** The progress printing is inefficient and may cause performance issues.
-💡 **Solution:** Use a more efficient way to print progress, such as using a progress bar library.
-
-**Current Code:**
-```python
-print(f"\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}", end="", flush=True)
-```
-
-**Suggested Code:**
-```python
-from tqdm import tqdm
-with tqdm(total=total, desc="Processing applicants") as pbar:
-    for index, row in df.iterrows():
-        # process applicant
-        pbar.update(1)
-```
-
-### 2. No error handling for file not found.
-📁 **File:** `main.py:174`
-⚖️ **Severity:** 5/10
-🔍 **Description:** The code does not handle the case where the file is not found.
-💡 **Solution:** Add error handling for file not found.
-
-**Current Code:**
-```python
-main(input_file)
-```
-
-**Suggested Code:**
-```python
-try:
-    main(input_file)
-except FileNotFoundError:
-    print(f"Error: file '{input_file}' not found")
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (2 issues)</strong></summary>
-
-<details>
-<summary><strong>Unused Import (2 issues)</strong></summary>
-
-### 1. Unused import 'random' should be removed.
-📁 **File:** `main.py:8`
-⚖️ **Severity:** 3/10
-🔍 **Description:** The import 'random' is not used anywhere in the code.
-💡 **Solution:** Remove the line 'import random'.
-
-**Current Code:**
-```python
-import random
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 2. Redundant code.
-📁 **File:** `main.py:141`
-⚖️ **Severity:** 3/10
-🔍 **Description:** The code is redundant and can be removed.
-💡 **Solution:** Remove the redundant code.
-
-**Current Code:**
-```python
-if len(df) == 0:
-    return
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (azure_ai/Meta-Llama-3-405B-Instruct)
-{"prompt_tokens": 6128, "completion_tokens": 1277, "total_tokens": 7405}
\ No newline at end of file
diff --git a/.experiments/code_review/main.py b/.experiments/code_review/main.py
index c04c609f..6350858c 100644
--- a/.experiments/code_review/main.py
+++ b/.experiments/code_review/main.py
@@ -35,7 +35,7 @@ def process_pr(pr_url, reeval_response=False):
     diff_text = get_diff_text(pr_diff, "")
     pr_files = get_pr_files(pr_files, "")
 
-    reviewer = CodeReviewer(llm_provider=LLMProvider(), default_model="best")
+    reviewer = CodeReviewer(llm_provider=LLMProvider(), default_model="default")
     review_data = reviewer.review_pull_request(
         diff_text=diff_text,
         pull_request_title=pr_title,
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_222/comments.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_222/comments.json
deleted file mode 100644
index 7edfffbe..00000000
--- a/.experiments/code_review/sonnet-3.5/no_eval/pr_222/comments.json
+++ /dev/null
@@ -1,72 +0,0 @@
-[
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Docker",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to Dockerfile, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_name": "Dockerfile",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Docker",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to docker-compose.yml, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "15",
-    "end_line": "15",
-    "side": "RIGHT",
-    "file_name": "docker-compose.yml",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Version Control",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to .gitignore, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "164",
-    "end_line": "164",
-    "side": "RIGHT",
-    "file_name": ".gitignore",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Database",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to db_setup/init.sql, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_name": "db_setup/init.sql",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_222/issues.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_222/issues.json
deleted file mode 100644
index 13cebd2d..00000000
--- a/.experiments/code_review/sonnet-3.5/no_eval/pr_222/issues.json
+++ /dev/null
@@ -1,312 +0,0 @@
-[
-  {
-    "topic": "Docker Configuration",
-    "comment": "Consider using multi-stage builds to reduce the final image size",
-    "confidence": "moderate",
-    "reason": "Multi-stage builds can significantly reduce the size of the final Docker image by excluding build dependencies",
-    "solution": "Implement a multi-stage build in the Dockerfile",
-    "actual_code": "RUN apt-get update && apt-get install -y \\\n    git \\\n    build-essential \\\n    && rm -rf /var/lib/apt/lists/*",
-    "fixed_code": "FROM python:3.9 AS builder\n\nRUN apt-get update && apt-get install -y \\\n    git \\\n    build-essential\n\n# ... (build steps)\n\nFROM python:3.9-slim\n\nCOPY --from=builder /app /app\n\n# ... (runtime configuration)",
-    "file_name": "Dockerfile",
-    "start_line": 7,
-    "end_line": 11,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Environment Variables",
-    "comment": "Consider using a more secure method for storing sensitive information",
-    "confidence": "important",
-    "reason": "Storing sensitive information like API keys directly in environment variables can be a security risk",
-    "solution": "Use a secret management system or encrypt sensitive values",
-    "actual_code": "OPENAI_API_KEY=\nOPENAI_ORGANIZATION=",
-    "fixed_code": "# Use a secret management system to securely store and retrieve API keys\n# OPENAI_API_KEY=\n# OPENAI_ORGANIZATION=",
-    "file_name": ".env.example",
-    "start_line": 10,
-    "end_line": 11,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Database Configuration",
-    "comment": "Ensure proper indexing for performance optimization",
-    "confidence": "important",
-    "reason": "Proper indexing is crucial for database performance, especially for frequently queried columns",
-    "solution": "Review and optimize index creation based on query patterns",
-    "actual_code": "CREATE INDEX idx_file_path ON files(file_path);\n\nCREATE INDEX idx_function_name ON function_abstractions(function_name);\n\nCREATE INDEX idx_node_type ON syntax_nodes(node_type);",
-    "fixed_code": "CREATE INDEX idx_file_path ON files(file_path);\nCREATE INDEX idx_function_name ON function_abstractions(function_name);\nCREATE INDEX idx_node_type ON syntax_nodes(node_type);\n-- Consider adding composite indexes based on common query patterns\n-- CREATE INDEX idx_file_repo ON files(repo_id, file_path);\n-- CREATE INDEX idx_function_file ON function_abstractions(file_id, function_name);",
-    "file_name": "db_setup/init.sql",
-    "start_line": 72,
-    "end_line": 78,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 5
-  },
-  {
-    "topic": "Code Embedding",
-    "comment": "Hardcoded embedding dimensions may limit flexibility",
-    "confidence": "moderate",
-    "reason": "Using a fixed embedding size of 1536 may not be suitable for all models or future changes",
-    "solution": "Consider making the embedding dimensions configurable",
-    "actual_code": "response = self.provider.embedding(\n    model=\"embedding\", input=[text], dimensions=1536, encoding_format=\"float\"\n)",
-    "fixed_code": "embedding_dim = self.config.get('embedding_dimensions', 1536)\nresponse = self.provider.embedding(\n    model=\"embedding\", input=[text], dimensions=embedding_dim, encoding_format=\"float\"\n)",
-    "file_name": "kaizen/llms/provider.py",
-    "start_line": 242,
-    "end_line": 244,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Type Hinting",
-    "comment": "Consider using more specific type hints for better code clarity and maintainability.",
-    "confidence": "important",
-    "reason": "Using more specific type hints can improve code readability and catch potential type-related errors early.",
-    "solution": "Replace 'List[float]' with 'np.ndarray' for the query_embedding parameter, and use 'List[Dict[str, Any]]' for the return type.",
-    "actual_code": "def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:",
-    "fixed_code": "def custom_query(self, query_embedding: np.ndarray, repo_id: int, similarity_top_k: int) -> List[Dict[str, Any]]:",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 13,
-    "end_line": 13,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "Add error handling for database operations to improve robustness.",
-    "confidence": "important",
-    "reason": "Database operations can fail due to various reasons, and proper error handling can prevent unexpected crashes and improve debugging.",
-    "solution": "Wrap the database operations in a try-except block and handle potential exceptions.",
-    "actual_code": "with self.get_client() as client:\n            with client.cursor() as cur:\n                cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))\n                results = cur.fetchall()",
-    "fixed_code": "try:\n    with self.get_client() as client:\n        with client.cursor() as cur:\n            cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))\n            results = cur.fetchall()\nexcept Exception as e:\n    # Log the error and handle it appropriately\n    print(f\"Database error:{e}\")\n    results =[]",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 39,
-    "end_line": 42,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Code Optimization",
-    "comment": "Consider using a list comprehension for creating the result list to improve performance and readability.",
-    "confidence": "moderate",
-    "reason": "List comprehensions are generally more efficient and concise than traditional for loops for creating lists.",
-    "solution": "Replace the for loop with a list comprehension.",
-    "actual_code": "return[\n{\n                \"id\": row[0],\n                \"text\": row[1],\n                \"metadata\": row[2] if isinstance(row[2], dict) else Json(row[2]),\n                \"similarity\": row[3]\n}\n            for row in results\n        ]",
-    "fixed_code": "return[{\n            \"id\": row[0],\n            \"text\": row[1],\n            \"metadata\": row[2] if isinstance(row[2], dict) else Json(row[2]),\n            \"similarity\": row[3]\n}for row in results]",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 44,
-    "end_line": 52,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 3
-  },
-  {
-    "topic": "Code Structure",
-    "comment": "Consider adding docstrings to methods for better documentation.",
-    "confidence": "moderate",
-    "reason": "Docstrings improve code readability and help other developers understand the purpose and usage of methods.",
-    "solution": "Add descriptive docstrings to the custom_query method and the AbstractionFeedback class methods.",
-    "actual_code": "def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:",
-    "fixed_code": "def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:\n    \"\"\"Perform a custom query on the vector store.\n\n    Args:\n        query_embedding (List[float]): The query embedding vector.\n        repo_id (int): The repository ID to filter results.\n        similarity_top_k (int): The number of top similar results to return.\n\n    Returns:\n        List[dict]: A list of dictionaries containing the query results.\n    \"\"\"",
-    "file_name": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 13,
-    "end_line": 13,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 4
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "Improve error handling in the parse_file method",
-    "confidence": "important",
-    "reason": "The current implementation catches all exceptions and logs them, but continues execution. This might lead to incomplete or inconsistent data.",
-    "solution": "Consider rethrowing specific exceptions or implementing a more granular error handling strategy.",
-    "actual_code": "except Exception as e:\n            logger.error(f\"Error processing file{file_path}:{str(e)}\")\n            logger.error(traceback.format_exc())",
-    "fixed_code": "except Exception as e:\n            logger.error(f\"Error processing file{file_path}:{str(e)}\")\n            logger.error(traceback.format_exc())\n            raise",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 108,
-    "end_line": 110,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Code Duplication",
-    "comment": "Repeated code for database connection string",
-    "confidence": "important",
-    "reason": "The database connection string is defined in multiple places, which violates the DRY principle and makes maintenance harder.",
-    "solution": "Extract the database connection string creation into a separate method or constant.",
-    "actual_code": "f\"postgresql://{os.environ['POSTGRES_USER']}:{os.environ['POSTGRES_PASSWORD']}@{os.environ['POSTGRES_HOST']}:{os.environ['POSTGRES_PORT']}/{os.environ['POSTGRES_DB']}\"",
-    "fixed_code": "self.db_connection_string = f\"postgresql://{os.environ['POSTGRES_USER']}:{os.environ['POSTGRES_PASSWORD']}@{os.environ['POSTGRES_HOST']}:{os.environ['POSTGRES_PORT']}/{os.environ['POSTGRES_DB']}\"\n        self.engine = create_engine(\n            self.db_connection_string,\n            pool_size=10,\n            max_overflow=20,\n        )",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 36,
-    "end_line": 37,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 6
-  },
-  {
-    "topic": "Code Optimization",
-    "comment": "Potential performance issue in store_function_relationships method",
-    "confidence": "moderate",
-    "reason": "The method executes a database query for each edge in the graph, which could be inefficient for large graphs.",
-    "solution": "Consider batching the inserts or using a more efficient bulk insert method if supported by the database.",
-    "actual_code": "for caller, callee in self.graph.edges():\n                query = text(\n                    \"\"\"\n                    INSERT INTO node_relationships (parent_node_id, child_node_id, relationship_type)\n                    VALUES (\n                        (SELECT node_id FROM syntax_nodes WHERE node_content LIKE :caller),\n                        (SELECT node_id FROM syntax_nodes WHERE node_content LIKE :callee),\n                        'calls'\n                    )\n                    ON CONFLICT DO NOTHING\n                \"\"\")\n                connection.execute(\n                    query,{\"caller\": f\"%{caller}%\", \"callee\": f\"%{callee}%\"}\n                )",
-    "fixed_code": "relationships =[(caller, callee) for caller, callee in self.graph.edges()]\n            query = text(\"\"\"\n                INSERT INTO node_relationships (parent_node_id, child_node_id, relationship_type)\n                VALUES (\n                    (SELECT node_id FROM syntax_nodes WHERE node_content LIKE :caller),\n                    (SELECT node_id FROM syntax_nodes WHERE node_content LIKE :callee),\n                    'calls'\n                )\n                ON CONFLICT DO NOTHING\n            \"\"\")\n            connection.execute(query,[{\"caller\": f\"%{caller}%\", \"callee\": f\"%{callee}%\"}for caller, callee in relationships])",
-    "file_name": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 298,
-    "end_line": 312,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 5
-  },
-  {
-    "topic": "Python Version Upgrade",
-    "comment": "The minimum Python version has been increased from 3.8.1 to 3.9.0.",
-    "confidence": "important",
-    "reason": "This change may break compatibility with environments using Python 3.8.x.",
-    "solution": "Ensure all development and production environments are updated to Python 3.9.0 or higher. Update CI/CD pipelines and deployment scripts accordingly.",
-    "actual_code": "python = \"^3.9.0\"",
-    "fixed_code": "",
-    "file_name": "pyproject.toml",
-    "start_line": 13,
-    "end_line": 13,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 7
-  },
-  {
-    "topic": "New Dependencies",
-    "comment": "Several new dependencies have been added, including llama-index and tree-sitter related packages.",
-    "confidence": "important",
-    "reason": "New dependencies may introduce compatibility issues or increase the project's complexity.",
-    "solution": "Review each new dependency for necessity and potential impact on the project. Ensure they are compatible with existing dependencies and the project's requirements.",
-    "actual_code": "llama-index-core = \"^0.10.47\"\nllama-index-llms-openai = \"^0.1.22\"\nllama-index-readers-file = \"^0.1.25\"\nllama-index-vector-stores-postgres = \"^0.1.11\"\nsqlalchemy = \"^2.0.31\"\nesprima = \"^4.0.1\"\nescodegen = \"^1.0.11\"\ntree-sitter = \"^0.22.3\"\nllama-index = \"^0.10.65\"\ntree-sitter-python = \"^0.21.0\"\ntree-sitter-javascript = \"^0.21.4\"\ntree-sitter-typescript = \"^0.21.2\"\ntree-sitter-rust = \"^0.21.2\"\nllama-index-llms-litellm = \"^0.1.4\"\nllama-index-embeddings-litellm = \"^0.1.1\"",
-    "fixed_code": "",
-    "file_name": "pyproject.toml",
-    "start_line": 27,
-    "end_line": 43,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 6
-  },
-  {
-    "topic": "Error Handling in LanguageLoader",
-    "comment": "The error handling in the LanguageLoader class could be improved for better debugging.",
-    "confidence": "moderate",
-    "reason": "The current error handling catches all exceptions and logs them, which might hide specific issues.",
-    "solution": "Consider catching specific exceptions (e.g., ImportError) separately and provide more detailed error messages.",
-    "actual_code": "except Exception as e:\n    logger.error(f\"Failed to load language{language}:{str(e)}\")\n    raise",
-    "fixed_code": "except ImportError as e:\n    logger.error(f\"Failed to import language module for{language}:{str(e)}\")\n    raise\nexcept Exception as e:\n    logger.error(f\"Unexpected error loading language{language}:{str(e)}\")\n    raise",
-    "file_name": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 28,
-    "end_line": 30,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 5
-  },
-  {
-    "topic": "Unused Import in Test File",
-    "comment": "The 'json' module is imported but not used in the test file.",
-    "confidence": "moderate",
-    "reason": "Unused imports can clutter the code and potentially confuse other developers.",
-    "solution": "Remove the unused import to improve code cleanliness.",
-    "actual_code": "import json",
-    "fixed_code": "",
-    "file_name": "tests/retriever/test_chunker.py",
-    "start_line": 2,
-    "end_line": 2,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 3
-  },
-  {
-    "topic": "Commented Out Code in Test File",
-    "comment": "There are several blocks of commented-out code in the test file.",
-    "confidence": "moderate",
-    "reason": "Commented-out code can make the file harder to read and maintain.",
-    "solution": "Remove commented-out code if it's no longer needed, or add a clear comment explaining why it's kept if it might be useful in the future.",
-    "actual_code": "# print(\"\\nFunctions:\")\n# for name, func in chunks[\"functions\"].items():\n#     print(f\"\\n{name}:\\n{func}\")\n\n# print(\"\\nClasses:\")\n# for name, class_info in chunks[\"classes\"].items():\n#     print(f\"\\n{name}:\")\n#     print(f\"Definition:\\n{class_info['definition']}\")\n#     print(\"Methods:\")\n#     for method_name, method in class_info[\"methods\"].items():\n#         print(f\"\\n{method_name}:\\n{method}\")\n\n# print(\"\\nOther Blocks:\")\n# for i, block in enumerate(chunks[\"other_blocks\"], 1):\n#     print(f\"\\nBlock{i}:\\n{block}\")",
-    "fixed_code": "",
-    "file_name": "tests/retriever/test_chunker.py",
-    "start_line": 81,
-    "end_line": 95,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 4
-  },
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Docker",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to Dockerfile, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_name": "Dockerfile",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Docker",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to docker-compose.yml, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "15",
-    "end_line": "15",
-    "side": "RIGHT",
-    "file_name": "docker-compose.yml",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Version Control",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to .gitignore, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "164",
-    "end_line": "164",
-    "side": "RIGHT",
-    "file_name": ".gitignore",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Database",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to db_setup/init.sql, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_name": "db_setup/init.sql",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_222/review.md b/.experiments/code_review/sonnet-3.5/no_eval/pr_222/review.md
deleted file mode 100644
index abca4169..00000000
--- a/.experiments/code_review/sonnet-3.5/no_eval/pr_222/review.md
+++ /dev/null
@@ -1,481 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/222
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 21
-- Critical: 5
-- Important: 8
-- Minor: 8
-- Files Affected: 12
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Configuration (5 issues)</strong></summary>
-
-### 1. Changes made to sensitive file
-📁 **File:** `config.json:4`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to config.json, which needs review
-💡 **Solution:** NA
-
-### 2. Changes made to sensitive file
-📁 **File:** `Dockerfile:4`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to Dockerfile, which needs review
-💡 **Solution:** NA
-
-### 3. Changes made to sensitive file
-📁 **File:** `docker-compose.yml:15`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to docker-compose.yml, which needs review
-💡 **Solution:** NA
-
-### 4. Changes made to sensitive file
-📁 **File:** `.gitignore:164`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to .gitignore, which needs review
-💡 **Solution:** NA
-
-### 5. Changes made to sensitive file
-📁 **File:** `db_setup/init.sql:1`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to db_setup/init.sql, which needs review
-💡 **Solution:** NA
-
-</details>
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Environment Variables (8 issues)</strong></summary>
-
-### 1. Consider using a more secure method for storing sensitive information
-📁 **File:** `.env.example:10`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Storing sensitive information like API keys directly in environment variables can be a security risk
-💡 **Solution:** Use a secret management system or encrypt sensitive values
-
-**Current Code:**
-```python
-OPENAI_API_KEY=
-OPENAI_ORGANIZATION=
-```
-
-**Suggested Code:**
-```python
-# Use a secret management system to securely store and retrieve API keys
-# OPENAI_API_KEY=
-# OPENAI_ORGANIZATION=
-```
-
-### 2. Ensure proper indexing for performance optimization
-📁 **File:** `db_setup/init.sql:72`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Proper indexing is crucial for database performance, especially for frequently queried columns
-💡 **Solution:** Review and optimize index creation based on query patterns
-
-**Current Code:**
-```python
-CREATE INDEX idx_file_path ON files(file_path);
-
-CREATE INDEX idx_function_name ON function_abstractions(function_name);
-
-CREATE INDEX idx_node_type ON syntax_nodes(node_type);
-```
-
-**Suggested Code:**
-```python
-CREATE INDEX idx_file_path ON files(file_path);
-CREATE INDEX idx_function_name ON function_abstractions(function_name);
-CREATE INDEX idx_node_type ON syntax_nodes(node_type);
--- Consider adding composite indexes based on common query patterns
--- CREATE INDEX idx_file_repo ON files(repo_id, file_path);
--- CREATE INDEX idx_function_file ON function_abstractions(file_id, function_name);
-```
-
-### 3. Consider using more specific type hints for better code clarity and maintainability.
-📁 **File:** `kaizen/retriever/custom_vector_store.py:13`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Using more specific type hints can improve code readability and catch potential type-related errors early.
-💡 **Solution:** Replace 'List[float]' with 'np.ndarray' for the query_embedding parameter, and use 'List[Dict[str, Any]]' for the return type.
-
-**Current Code:**
-```python
-def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:
-```
-
-**Suggested Code:**
-```python
-def custom_query(self, query_embedding: np.ndarray, repo_id: int, similarity_top_k: int) -> List[Dict[str, Any]]:
-```
-
-### 4. Add error handling for database operations to improve robustness.
-📁 **File:** `kaizen/retriever/custom_vector_store.py:39`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Database operations can fail due to various reasons, and proper error handling can prevent unexpected crashes and improve debugging.
-💡 **Solution:** Wrap the database operations in a try-except block and handle potential exceptions.
-
-**Current Code:**
-```python
-with self.get_client() as client:
-            with client.cursor() as cur:
-                cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))
-                results = cur.fetchall()
-```
-
-**Suggested Code:**
-```python
-try:
-    with self.get_client() as client:
-        with client.cursor() as cur:
-            cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))
-            results = cur.fetchall()
-except Exception as e:
-    # Log the error and handle it appropriately
-    print(f"Database error:{e}")
-    results =[]
-```
-
-### 5. Improve error handling in the parse_file method
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:108`
-⚖️ **Severity:** 7/10
-🔍 **Description:** The current implementation catches all exceptions and logs them, but continues execution. This might lead to incomplete or inconsistent data.
-💡 **Solution:** Consider rethrowing specific exceptions or implementing a more granular error handling strategy.
-
-**Current Code:**
-```python
-except Exception as e:
-            logger.error(f"Error processing file{file_path}:{str(e)}")
-            logger.error(traceback.format_exc())
-```
-
-**Suggested Code:**
-```python
-except Exception as e:
-            logger.error(f"Error processing file{file_path}:{str(e)}")
-            logger.error(traceback.format_exc())
-            raise
-```
-
-### 6. Repeated code for database connection string
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:36`
-⚖️ **Severity:** 6/10
-🔍 **Description:** The database connection string is defined in multiple places, which violates the DRY principle and makes maintenance harder.
-💡 **Solution:** Extract the database connection string creation into a separate method or constant.
-
-**Current Code:**
-```python
-f"postgresql://{os.environ['POSTGRES_USER']}:{os.environ['POSTGRES_PASSWORD']}@{os.environ['POSTGRES_HOST']}:{os.environ['POSTGRES_PORT']}/{os.environ['POSTGRES_DB']}"
-```
-
-**Suggested Code:**
-```python
-self.db_connection_string = f"postgresql://{os.environ['POSTGRES_USER']}:{os.environ['POSTGRES_PASSWORD']}@{os.environ['POSTGRES_HOST']}:{os.environ['POSTGRES_PORT']}/{os.environ['POSTGRES_DB']}"
-        self.engine = create_engine(
-            self.db_connection_string,
-            pool_size=10,
-            max_overflow=20,
-        )
-```
-
-### 7. The minimum Python version has been increased from 3.8.1 to 3.9.0.
-📁 **File:** `pyproject.toml:13`
-⚖️ **Severity:** 7/10
-🔍 **Description:** This change may break compatibility with environments using Python 3.8.x.
-💡 **Solution:** Ensure all development and production environments are updated to Python 3.9.0 or higher. Update CI/CD pipelines and deployment scripts accordingly.
-
-**Current Code:**
-```python
-python = "^3.9.0"
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 8. Several new dependencies have been added, including llama-index and tree-sitter related packages.
-📁 **File:** `pyproject.toml:27`
-⚖️ **Severity:** 6/10
-🔍 **Description:** New dependencies may introduce compatibility issues or increase the project's complexity.
-💡 **Solution:** Review each new dependency for necessity and potential impact on the project. Ensure they are compatible with existing dependencies and the project's requirements.
-
-**Current Code:**
-```python
-llama-index-core = "^0.10.47"
-llama-index-llms-openai = "^0.1.22"
-llama-index-readers-file = "^0.1.25"
-llama-index-vector-stores-postgres = "^0.1.11"
-sqlalchemy = "^2.0.31"
-esprima = "^4.0.1"
-escodegen = "^1.0.11"
-tree-sitter = "^0.22.3"
-llama-index = "^0.10.65"
-tree-sitter-python = "^0.21.0"
-tree-sitter-javascript = "^0.21.4"
-tree-sitter-typescript = "^0.21.2"
-tree-sitter-rust = "^0.21.2"
-llama-index-llms-litellm = "^0.1.4"
-llama-index-embeddings-litellm = "^0.1.1"
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (8 issues)</strong></summary>
-
-<details>
-<summary><strong>Docker Configuration (8 issues)</strong></summary>
-
-### 1. Consider using multi-stage builds to reduce the final image size
-📁 **File:** `Dockerfile:7`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Multi-stage builds can significantly reduce the size of the final Docker image by excluding build dependencies
-💡 **Solution:** Implement a multi-stage build in the Dockerfile
-
-**Current Code:**
-```python
-RUN apt-get update && apt-get install -y \
-    git \
-    build-essential \
-    && rm -rf /var/lib/apt/lists/*
-```
-
-**Suggested Code:**
-```python
-FROM python:3.9 AS builder
-
-RUN apt-get update && apt-get install -y \
-    git \
-    build-essential
-
-# ... (build steps)
-
-FROM python:3.9-slim
-
-COPY --from=builder /app /app
-
-# ... (runtime configuration)
-```
-
-### 2. Hardcoded embedding dimensions may limit flexibility
-📁 **File:** `kaizen/llms/provider.py:242`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Using a fixed embedding size of 1536 may not be suitable for all models or future changes
-💡 **Solution:** Consider making the embedding dimensions configurable
-
-**Current Code:**
-```python
-response = self.provider.embedding(
-    model="embedding", input=[text], dimensions=1536, encoding_format="float"
-)
-```
-
-**Suggested Code:**
-```python
-embedding_dim = self.config.get('embedding_dimensions', 1536)
-response = self.provider.embedding(
-    model="embedding", input=[text], dimensions=embedding_dim, encoding_format="float"
-)
-```
-
-### 3. Consider using a list comprehension for creating the result list to improve performance and readability.
-📁 **File:** `kaizen/retriever/custom_vector_store.py:44`
-⚖️ **Severity:** 3/10
-🔍 **Description:** List comprehensions are generally more efficient and concise than traditional for loops for creating lists.
-💡 **Solution:** Replace the for loop with a list comprehension.
-
-**Current Code:**
-```python
-return[
-{
-                "id": row[0],
-                "text": row[1],
-                "metadata": row[2] if isinstance(row[2], dict) else Json(row[2]),
-                "similarity": row[3]
-}
-            for row in results
-        ]
-```
-
-**Suggested Code:**
-```python
-return[{
-            "id": row[0],
-            "text": row[1],
-            "metadata": row[2] if isinstance(row[2], dict) else Json(row[2]),
-            "similarity": row[3]
-}for row in results]
-```
-
-### 4. Consider adding docstrings to methods for better documentation.
-📁 **File:** `kaizen/retriever/custom_vector_store.py:13`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Docstrings improve code readability and help other developers understand the purpose and usage of methods.
-💡 **Solution:** Add descriptive docstrings to the custom_query method and the AbstractionFeedback class methods.
-
-**Current Code:**
-```python
-def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:
-```
-
-**Suggested Code:**
-```python
-def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:
-    """Perform a custom query on the vector store.
-
-    Args:
-        query_embedding (List[float]): The query embedding vector.
-        repo_id (int): The repository ID to filter results.
-        similarity_top_k (int): The number of top similar results to return.
-
-    Returns:
-        List[dict]: A list of dictionaries containing the query results.
-    """
-```
-
-### 5. Potential performance issue in store_function_relationships method
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:298`
-⚖️ **Severity:** 5/10
-🔍 **Description:** The method executes a database query for each edge in the graph, which could be inefficient for large graphs.
-💡 **Solution:** Consider batching the inserts or using a more efficient bulk insert method if supported by the database.
-
-**Current Code:**
-```python
-for caller, callee in self.graph.edges():
-                query = text(
-                    """
-                    INSERT INTO node_relationships (parent_node_id, child_node_id, relationship_type)
-                    VALUES (
-                        (SELECT node_id FROM syntax_nodes WHERE node_content LIKE :caller),
-                        (SELECT node_id FROM syntax_nodes WHERE node_content LIKE :callee),
-                        'calls'
-                    )
-                    ON CONFLICT DO NOTHING
-                """)
-                connection.execute(
-                    query,{"caller": f"%{caller}%", "callee": f"%{callee}%"}
-                )
-```
-
-**Suggested Code:**
-```python
-relationships =[(caller, callee) for caller, callee in self.graph.edges()]
-            query = text("""
-                INSERT INTO node_relationships (parent_node_id, child_node_id, relationship_type)
-                VALUES (
-                    (SELECT node_id FROM syntax_nodes WHERE node_content LIKE :caller),
-                    (SELECT node_id FROM syntax_nodes WHERE node_content LIKE :callee),
-                    'calls'
-                )
-                ON CONFLICT DO NOTHING
-            """)
-            connection.execute(query,[{"caller": f"%{caller}%", "callee": f"%{callee}%"}for caller, callee in relationships])
-```
-
-### 6. The error handling in the LanguageLoader class could be improved for better debugging.
-📁 **File:** `kaizen/retriever/tree_sitter_utils.py:28`
-⚖️ **Severity:** 5/10
-🔍 **Description:** The current error handling catches all exceptions and logs them, which might hide specific issues.
-💡 **Solution:** Consider catching specific exceptions (e.g., ImportError) separately and provide more detailed error messages.
-
-**Current Code:**
-```python
-except Exception as e:
-    logger.error(f"Failed to load language{language}:{str(e)}")
-    raise
-```
-
-**Suggested Code:**
-```python
-except ImportError as e:
-    logger.error(f"Failed to import language module for{language}:{str(e)}")
-    raise
-except Exception as e:
-    logger.error(f"Unexpected error loading language{language}:{str(e)}")
-    raise
-```
-
-### 7. The 'json' module is imported but not used in the test file.
-📁 **File:** `tests/retriever/test_chunker.py:2`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Unused imports can clutter the code and potentially confuse other developers.
-💡 **Solution:** Remove the unused import to improve code cleanliness.
-
-**Current Code:**
-```python
-import json
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 8. There are several blocks of commented-out code in the test file.
-📁 **File:** `tests/retriever/test_chunker.py:81`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Commented-out code can make the file harder to read and maintain.
-💡 **Solution:** Remove commented-out code if it's no longer needed, or add a clear comment explaining why it's kept if it might be useful in the future.
-
-**Current Code:**
-```python
-# print("\nFunctions:")
-# for name, func in chunks["functions"].items():
-#     print(f"\n{name}:\n{func}")
-
-# print("\nClasses:")
-# for name, class_info in chunks["classes"].items():
-#     print(f"\n{name}:")
-#     print(f"Definition:\n{class_info['definition']}")
-#     print("Methods:")
-#     for method_name, method in class_info["methods"].items():
-#         print(f"\n{method_name}:\n{method}")
-
-# print("\nOther Blocks:")
-# for i, block in enumerate(chunks["other_blocks"], 1):
-#     print(f"\nBlock{i}:\n{block}")
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (anthropic.claude-3-5-sonnet-20240620-v1:0)
-{"prompt_tokens": 24844, "completion_tokens": 5105, "total_tokens": 29949}
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_232/issues.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_232/issues.json
deleted file mode 100644
index 5a4abbcf..00000000
--- a/.experiments/code_review/sonnet-3.5/no_eval/pr_232/issues.json
+++ /dev/null
@@ -1,227 +0,0 @@
-[
-  {
-    "topic": "Type Definition",
-    "comment": "Improved type definition for MemoriesPage props",
-    "confidence": "important",
-    "reason": "Using a separate type definition improves code readability and maintainability",
-    "solution": "The change is already implemented correctly",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 40,
-    "end_line": 45,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 3
-  },
-  {
-    "topic": "Error Message",
-    "comment": "Improved error message for space deletion",
-    "confidence": "moderate",
-    "reason": "More specific error message provides better user feedback",
-    "solution": "The change is already implemented correctly",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 73,
-    "end_line": 73,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 2
-  },
-  {
-    "topic": "Component Naming",
-    "comment": "Renamed components for better clarity",
-    "confidence": "important",
-    "reason": "More descriptive component names improve code readability",
-    "solution": "The changes are already implemented correctly",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 231,
-    "end_line": 231,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 3
-  },
-  {
-    "topic": "UI Improvements",
-    "comment": "Enhanced UI elements with better styling and layout",
-    "confidence": "moderate",
-    "reason": "Improved visual consistency and user experience",
-    "solution": "The changes are already implemented correctly",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 140,
-    "end_line": 152,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 3
-  },
-  {
-    "topic": "Import Optimization",
-    "comment": "Removed unused import",
-    "confidence": "important",
-    "reason": "Removing unused imports improves code cleanliness and potentially reduces bundle size",
-    "solution": "The change is already implemented correctly",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "apps/web/app/(dash)/home/page.tsx",
-    "start_line": 6,
-    "end_line": 6,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 3
-  },
-  {
-    "topic": "Import Cleanup",
-    "comment": "Removed unused imports",
-    "confidence": "important",
-    "reason": "Removing unused imports improves code cleanliness and potentially reduces bundle size",
-    "solution": "The change is already implemented correctly",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "apps/web/app/(dash)/home/queryinput.tsx",
-    "start_line": 3,
-    "end_line": 3,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 3
-  },
-  {
-    "topic": "Code Structure",
-    "comment": "Improved code organization by extracting dialog content into a separate component",
-    "confidence": "important",
-    "reason": "Separating concerns improves readability and maintainability",
-    "solution": "The change is already implemented correctly",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "apps/web/app/(dash)/menu.tsx",
-    "start_line": 163,
-    "end_line": 346,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 7
-  },
-  {
-    "topic": "State Management",
-    "comment": "Moved state management for spaces and selectedSpaces into the DialogContentContainer component",
-    "confidence": "important",
-    "reason": "Localizing state management to the component that uses it improves encapsulation",
-    "solution": "The change is already implemented correctly",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "apps/web/app/(dash)/menu.tsx",
-    "start_line": 168,
-    "end_line": 170,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 6
-  },
-  {
-    "topic": "Prop Drilling",
-    "comment": "Consider using context or state management library to avoid prop drilling",
-    "confidence": "moderate",
-    "reason": "The setDialogClose function is passed down as a prop, which could lead to prop drilling in larger components",
-    "solution": "Implement React Context or use a state management library like Redux for managing global state",
-    "actual_code": "function DialogContentContainer({\n\tsetDialogClose,\n}:{\n\tsetDialogClose: () => void;\n}){",
-    "fixed_code": "const DialogContext = React.createContext();\n\nfunction DialogContentContainer(){\n  const{setDialogClose}= useContext(DialogContext);",
-    "file_name": "apps/web/app/(dash)/menu.tsx",
-    "start_line": 163,
-    "end_line": 167,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "Improve error handling in the handleSubmit function",
-    "confidence": "important",
-    "reason": "The current implementation throws an error but then continues execution",
-    "solution": "Remove the return statement after throwing the error",
-    "actual_code": "throw new Error(`Memory creation failed: ${cont.error}`);\nreturn cont;",
-    "fixed_code": "throw new Error(`Memory creation failed: ${cont.error}`);",
-    "file_name": "apps/web/app/(dash)/menu.tsx",
-    "start_line": 230,
-    "end_line": 231,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Unused Import",
-    "comment": "The useEffect import is removed but not replaced with any other import.",
-    "confidence": "important",
-    "reason": "Removing unused imports improves code cleanliness and potentially reduces bundle size.",
-    "solution": "Ensure all necessary hooks are imported and remove any unused imports.",
-    "actual_code": "import{useState}from \"react\";",
-    "fixed_code": "import{useState, useEffect}from \"react\";",
-    "file_name": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 3,
-    "end_line": 3,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 3
-  },
-  {
-    "topic": "Type Safety",
-    "comment": "The component definition has been changed from a typed functional component to a regular function without explicit typing.",
-    "confidence": "important",
-    "reason": "Removing explicit typing can lead to potential type-related bugs and reduces code readability.",
-    "solution": "Maintain explicit typing for the component to ensure type safety and improve code clarity.",
-    "actual_code": "const ComboboxWithCreate = ({",
-    "fixed_code": "const ComboboxWithCreate: React.FC<ComboboxWithCreateProps> = ({",
-    "file_name": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 32,
-    "end_line": 32,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 6
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "The new handleKeyDown function doesn't handle potential undefined values when accessing selectedSpaces.",
-    "confidence": "moderate",
-    "reason": "Not checking for undefined values can lead to runtime errors if selectedSpaces is not properly initialized.",
-    "solution": "Add a null check before accessing selectedSpaces.length.",
-    "actual_code": "if (\n\t\t\te.key === \"Backspace\" &&\n\t\t\tinputValue === \"\" &&\n\t\t\tselectedSpaces.length > 0\n\t\t){\n\t\t\tsetSelectedSpaces((prev) => prev.slice(0, -1));\n\t\t}",
-    "fixed_code": "if (\n\t\t\te.key === \"Backspace\" &&\n\t\t\tinputValue === \"\" &&\n\t\t\tselectedSpaces?.length > 0\n\t\t){\n\t\t\tsetSelectedSpaces((prev) => prev.slice(0, -1));\n\t\t}",
-    "file_name": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 46,
-    "end_line": 52,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 5
-  },
-  {
-    "topic": "Performance Optimization",
-    "comment": "The filteredOptions array is being recalculated on every render, which could be inefficient for large arrays.",
-    "confidence": "moderate",
-    "reason": "Recalculating filtered options on every render can lead to unnecessary computations and potential performance issues.",
-    "solution": "Consider using useMemo to memoize the filteredOptions calculation.",
-    "actual_code": "const filteredOptions = options.filter(\n\t\t(option) => !selectedSpaces.includes(parseInt(option.value)),\n\t);",
-    "fixed_code": "const filteredOptions = useMemo(() => options.filter(\n\t\t(option) => !selectedSpaces.includes(parseInt(option.value)),\n\t),[options, selectedSpaces]);",
-    "file_name": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 55,
-    "end_line": 57,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Accessibility",
-    "comment": "The button for removing selected spaces lacks an aria-label for better accessibility.",
-    "confidence": "moderate",
-    "reason": "Missing aria-labels can make it difficult for screen reader users to understand the purpose of interactive elements.",
-    "solution": "Add an appropriate aria-label to the button for removing selected spaces.",
-    "actual_code": "<button\n\t\t\t\t\t\t\tkey={spaceId}\n\t\t\t\t\t\t\ttype=\"button\"\n\t\t\t\t\t\t\tonClick={() =>\n\t\t\t\t\t\t\t\tsetSelectedSpaces((prev) => prev.filter((id) => id !== spaceId))\n\t\t\t\t\t\t\t}\n\t\t\t\t\t\t\tclassName=\"relative group rounded-md py-1 px-2 bg-[#3C464D] max-w-32\"\n\t\t\t\t\t\t>",
-    "fixed_code": "<button\n\t\t\t\t\t\t\tkey={spaceId}\n\t\t\t\t\t\t\ttype=\"button\"\n\t\t\t\t\t\t\tonClick={() =>\n\t\t\t\t\t\t\t\tsetSelectedSpaces((prev) => prev.filter((id) => id !== spaceId))\n\t\t\t\t\t\t\t}\n\t\t\t\t\t\t\tclassName=\"relative group rounded-md py-1 px-2 bg-[#3C464D] max-w-32\"\n\t\t\t\t\t\t\taria-label={`Remove ${options.find((opt) => opt.value === spaceId.toString())?.label}`}\n\t\t\t\t\t\t>",
-    "file_name": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 65,
-    "end_line": 72,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_232/review.md b/.experiments/code_review/sonnet-3.5/no_eval/pr_232/review.md
deleted file mode 100644
index 0c5cbfbc..00000000
--- a/.experiments/code_review/sonnet-3.5/no_eval/pr_232/review.md
+++ /dev/null
@@ -1,253 +0,0 @@
-PR URL: https://github.com/supermemoryai/supermemory/pull/232
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 15
-- Critical: 0
-- Important: 9
-- Minor: 6
-- Files Affected: 5
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Type Definition (9 issues)</strong></summary>
-
-### 1. Improved type definition for MemoriesPage props
-📁 **File:** `apps/web/app/(dash)/(memories)/content.tsx:40`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Using a separate type definition improves code readability and maintainability
-💡 **Solution:** The change is already implemented correctly
-
-### 2. Renamed components for better clarity
-📁 **File:** `apps/web/app/(dash)/(memories)/content.tsx:231`
-⚖️ **Severity:** 3/10
-🔍 **Description:** More descriptive component names improve code readability
-💡 **Solution:** The changes are already implemented correctly
-
-### 3. Removed unused import
-📁 **File:** `apps/web/app/(dash)/home/page.tsx:6`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Removing unused imports improves code cleanliness and potentially reduces bundle size
-💡 **Solution:** The change is already implemented correctly
-
-### 4. Removed unused imports
-📁 **File:** `apps/web/app/(dash)/home/queryinput.tsx:3`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Removing unused imports improves code cleanliness and potentially reduces bundle size
-💡 **Solution:** The change is already implemented correctly
-
-### 5. Improved code organization by extracting dialog content into a separate component
-📁 **File:** `apps/web/app/(dash)/menu.tsx:163`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Separating concerns improves readability and maintainability
-💡 **Solution:** The change is already implemented correctly
-
-### 6. Moved state management for spaces and selectedSpaces into the DialogContentContainer component
-📁 **File:** `apps/web/app/(dash)/menu.tsx:168`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Localizing state management to the component that uses it improves encapsulation
-💡 **Solution:** The change is already implemented correctly
-
-### 7. Improve error handling in the handleSubmit function
-📁 **File:** `apps/web/app/(dash)/menu.tsx:230`
-⚖️ **Severity:** 7/10
-🔍 **Description:** The current implementation throws an error but then continues execution
-💡 **Solution:** Remove the return statement after throwing the error
-
-**Current Code:**
-```python
-throw new Error(`Memory creation failed: ${cont.error}`);
-return cont;
-```
-
-**Suggested Code:**
-```python
-throw new Error(`Memory creation failed: ${cont.error}`);
-```
-
-### 8. The useEffect import is removed but not replaced with any other import.
-📁 **File:** `packages/ui/shadcn/combobox.tsx:3`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Removing unused imports improves code cleanliness and potentially reduces bundle size.
-💡 **Solution:** Ensure all necessary hooks are imported and remove any unused imports.
-
-**Current Code:**
-```python
-import{useState}from "react";
-```
-
-**Suggested Code:**
-```python
-import{useState, useEffect}from "react";
-```
-
-### 9. The component definition has been changed from a typed functional component to a regular function without explicit typing.
-📁 **File:** `packages/ui/shadcn/combobox.tsx:32`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Removing explicit typing can lead to potential type-related bugs and reduces code readability.
-💡 **Solution:** Maintain explicit typing for the component to ensure type safety and improve code clarity.
-
-**Current Code:**
-```python
-const ComboboxWithCreate = ({
-```
-
-**Suggested Code:**
-```python
-const ComboboxWithCreate: React.FC<ComboboxWithCreateProps> = ({
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (6 issues)</strong></summary>
-
-<details>
-<summary><strong>Error Message (6 issues)</strong></summary>
-
-### 1. Improved error message for space deletion
-📁 **File:** `apps/web/app/(dash)/(memories)/content.tsx:73`
-⚖️ **Severity:** 2/10
-🔍 **Description:** More specific error message provides better user feedback
-💡 **Solution:** The change is already implemented correctly
-
-### 2. Enhanced UI elements with better styling and layout
-📁 **File:** `apps/web/app/(dash)/(memories)/content.tsx:140`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Improved visual consistency and user experience
-💡 **Solution:** The changes are already implemented correctly
-
-### 3. Consider using context or state management library to avoid prop drilling
-📁 **File:** `apps/web/app/(dash)/menu.tsx:163`
-⚖️ **Severity:** 4/10
-🔍 **Description:** The setDialogClose function is passed down as a prop, which could lead to prop drilling in larger components
-💡 **Solution:** Implement React Context or use a state management library like Redux for managing global state
-
-**Current Code:**
-```python
-function DialogContentContainer({
-	setDialogClose,
-}:{
-	setDialogClose: () => void;
-}){
-```
-
-**Suggested Code:**
-```python
-const DialogContext = React.createContext();
-
-function DialogContentContainer(){
-  const{setDialogClose}= useContext(DialogContext);
-```
-
-### 4. The new handleKeyDown function doesn't handle potential undefined values when accessing selectedSpaces.
-📁 **File:** `packages/ui/shadcn/combobox.tsx:46`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Not checking for undefined values can lead to runtime errors if selectedSpaces is not properly initialized.
-💡 **Solution:** Add a null check before accessing selectedSpaces.length.
-
-**Current Code:**
-```python
-if (
-			e.key === "Backspace" &&
-			inputValue === "" &&
-			selectedSpaces.length > 0
-		){
-			setSelectedSpaces((prev) => prev.slice(0, -1));
-		}
-```
-
-**Suggested Code:**
-```python
-if (
-			e.key === "Backspace" &&
-			inputValue === "" &&
-			selectedSpaces?.length > 0
-		){
-			setSelectedSpaces((prev) => prev.slice(0, -1));
-		}
-```
-
-### 5. The filteredOptions array is being recalculated on every render, which could be inefficient for large arrays.
-📁 **File:** `packages/ui/shadcn/combobox.tsx:55`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Recalculating filtered options on every render can lead to unnecessary computations and potential performance issues.
-💡 **Solution:** Consider using useMemo to memoize the filteredOptions calculation.
-
-**Current Code:**
-```python
-const filteredOptions = options.filter(
-		(option) => !selectedSpaces.includes(parseInt(option.value)),
-	);
-```
-
-**Suggested Code:**
-```python
-const filteredOptions = useMemo(() => options.filter(
-		(option) => !selectedSpaces.includes(parseInt(option.value)),
-	),[options, selectedSpaces]);
-```
-
-### 6. The button for removing selected spaces lacks an aria-label for better accessibility.
-📁 **File:** `packages/ui/shadcn/combobox.tsx:65`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Missing aria-labels can make it difficult for screen reader users to understand the purpose of interactive elements.
-💡 **Solution:** Add an appropriate aria-label to the button for removing selected spaces.
-
-**Current Code:**
-```python
-<button
-							key={spaceId}
-							type="button"
-							onClick={() =>
-								setSelectedSpaces((prev) => prev.filter((id) => id !== spaceId))
-							}
-							className="relative group rounded-md py-1 px-2 bg-[#3C464D] max-w-32"
-						>
-```
-
-**Suggested Code:**
-```python
-<button
-							key={spaceId}
-							type="button"
-							onClick={() =>
-								setSelectedSpaces((prev) => prev.filter((id) => id !== spaceId))
-							}
-							className="relative group rounded-md py-1 px-2 bg-[#3C464D] max-w-32"
-							aria-label={`Remove ${options.find((opt) => opt.value === spaceId.toString())?.label}`}
-						>
-```
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (anthropic.claude-3-5-sonnet-20240620-v1:0)
-{"prompt_tokens": 17838, "completion_tokens": 3363, "total_tokens": 21201}
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_252/issues.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_252/issues.json
deleted file mode 100644
index 0b1ff3f5..00000000
--- a/.experiments/code_review/sonnet-3.5/no_eval/pr_252/issues.json
+++ /dev/null
@@ -1,62 +0,0 @@
-[
-  {
-    "topic": "Code Structure",
-    "comment": "The `generate_linkedin_post` method call has been split across multiple lines, which improves readability for long function calls.",
-    "confidence": "moderate",
-    "reason": "Multi-line function calls can improve code readability, especially for functions with long parameter lists.",
-    "solution": "The current implementation is good. No changes needed.",
-    "actual_code": "linkedin_post = work_summary_generator.generate_linkedin_post(\n    summary, user=\"oss_example\"\n)",
-    "fixed_code": "",
-    "file_name": "examples/work_summarizer/main.py",
-    "start_line": 60,
-    "end_line": 62,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 2
-  },
-  {
-    "topic": "Code Consistency",
-    "comment": "The `generate_twitter_post` method call is not formatted consistently with the `generate_linkedin_post` call.",
-    "confidence": "moderate",
-    "reason": "Consistent formatting improves code readability and maintainability.",
-    "solution": "Consider formatting the `generate_twitter_post` call similarly to the `generate_linkedin_post` call for consistency.",
-    "actual_code": "twitter_post = work_summary_generator.generate_twitter_post(summary, user=\"oss_example\")",
-    "fixed_code": "twitter_post = work_summary_generator.generate_twitter_post(\n    summary, user=\"oss_example\"\n)",
-    "file_name": "examples/work_summarizer/main.py",
-    "start_line": 59,
-    "end_line": 59,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 3
-  },
-  {
-    "topic": "Code Duplication",
-    "comment": "The print statement for LinkedIn post is duplicated.",
-    "confidence": "important",
-    "reason": "Code duplication can lead to maintenance issues and inconsistencies.",
-    "solution": "Remove the duplicated print statement for the LinkedIn post.",
-    "actual_code": "print(f\" LinkedIn Post: \\n{linkedin_post}\\n\")",
-    "fixed_code": "",
-    "file_name": "examples/work_summarizer/main.py",
-    "start_line": 68,
-    "end_line": 68,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 6
-  },
-  {
-    "topic": "Code Documentation",
-    "comment": "The new `severity_level` field in the code review prompt is not explained in detail.",
-    "confidence": "moderate",
-    "reason": "Clear documentation helps users understand how to use the severity level correctly.",
-    "solution": "Add a brief explanation of what each severity level represents (e.g., what constitutes a level 1 vs. level 10 issue).",
-    "actual_code": "For \"severity_level\" score in range of 1 to 10, 1 being not severe and 10 being critical.",
-    "fixed_code": "For \"severity_level\" score in range of 1 to 10:\n1-3: Minor issues (style, small optimizations)\n4-6: Moderate issues (potential bugs, performance concerns)\n7-8: Major issues (definite bugs, security vulnerabilities)\n9-10: Critical issues (severe security risks, system-breaking bugs)",
-    "file_name": "kaizen/llms/prompts/code_review_prompts.py",
-    "start_line": 100,
-    "end_line": 100,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_252/review.md b/.experiments/code_review/sonnet-3.5/no_eval/pr_252/review.md
deleted file mode 100644
index d9e48407..00000000
--- a/.experiments/code_review/sonnet-3.5/no_eval/pr_252/review.md
+++ /dev/null
@@ -1,126 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/252
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 4
-- Critical: 0
-- Important: 1
-- Minor: 3
-- Files Affected: 2
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Code Duplication (1 issues)</strong></summary>
-
-### 1. The print statement for LinkedIn post is duplicated.
-📁 **File:** `examples/work_summarizer/main.py:68`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Code duplication can lead to maintenance issues and inconsistencies.
-💡 **Solution:** Remove the duplicated print statement for the LinkedIn post.
-
-**Current Code:**
-```python
-print(f" LinkedIn Post: \n{linkedin_post}\n")
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (3 issues)</strong></summary>
-
-<details>
-<summary><strong>Code Structure (3 issues)</strong></summary>
-
-### 1. The `generate_linkedin_post` method call has been split across multiple lines, which improves readability for long function calls.
-📁 **File:** `examples/work_summarizer/main.py:60`
-⚖️ **Severity:** 2/10
-🔍 **Description:** Multi-line function calls can improve code readability, especially for functions with long parameter lists.
-💡 **Solution:** The current implementation is good. No changes needed.
-
-**Current Code:**
-```python
-linkedin_post = work_summary_generator.generate_linkedin_post(
-    summary, user="oss_example"
-)
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 2. The `generate_twitter_post` method call is not formatted consistently with the `generate_linkedin_post` call.
-📁 **File:** `examples/work_summarizer/main.py:59`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Consistent formatting improves code readability and maintainability.
-💡 **Solution:** Consider formatting the `generate_twitter_post` call similarly to the `generate_linkedin_post` call for consistency.
-
-**Current Code:**
-```python
-twitter_post = work_summary_generator.generate_twitter_post(summary, user="oss_example")
-```
-
-**Suggested Code:**
-```python
-twitter_post = work_summary_generator.generate_twitter_post(
-    summary, user="oss_example"
-)
-```
-
-### 3. The new `severity_level` field in the code review prompt is not explained in detail.
-📁 **File:** `kaizen/llms/prompts/code_review_prompts.py:100`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Clear documentation helps users understand how to use the severity level correctly.
-💡 **Solution:** Add a brief explanation of what each severity level represents (e.g., what constitutes a level 1 vs. level 10 issue).
-
-**Current Code:**
-```python
-For "severity_level" score in range of 1 to 10, 1 being not severe and 10 being critical.
-```
-
-**Suggested Code:**
-```python
-For "severity_level" score in range of 1 to 10:
-1-3: Minor issues (style, small optimizations)
-4-6: Moderate issues (potential bugs, performance concerns)
-7-8: Major issues (definite bugs, security vulnerabilities)
-9-10: Critical issues (severe security risks, system-breaking bugs)
-```
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (anthropic.claude-3-5-sonnet-20240620-v1:0)
-{"prompt_tokens": 4436, "completion_tokens": 952, "total_tokens": 5388}
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_335/issues.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_335/issues.json
deleted file mode 100644
index 5b6fae70..00000000
--- a/.experiments/code_review/sonnet-3.5/no_eval/pr_335/issues.json
+++ /dev/null
@@ -1,62 +0,0 @@
-[
-  {
-    "topic": "Code Refactoring",
-    "comment": "Removal of reevaluation logic and associated prompts",
-    "confidence": "important",
-    "reason": "The code has been simplified by removing the reevaluation logic, which could potentially improve performance and reduce complexity.",
-    "solution": "Ensure that the removal of reevaluation doesn't impact the quality of PR descriptions. Consider adding unit tests to verify the new behavior.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/pr_description.py",
-    "start_line": 86,
-    "end_line": 88,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 5
-  },
-  {
-    "topic": "API Change",
-    "comment": "Change from chat_completion_with_json to chat_completion",
-    "confidence": "important",
-    "reason": "The API call has been changed, which might affect the format of the response and how it's processed.",
-    "solution": "Ensure that the new chat_completion method returns the expected format. Update any dependent code that might be affected by this change.",
-    "actual_code": "resp, usage = self.provider.chat_completion(prompt, user=user)",
-    "fixed_code": "resp, usage = self.provider.chat_completion(prompt, user=user)\ndesc = parser.extract_code_from_markdown(resp)",
-    "file_name": "kaizen/generator/pr_description.py",
-    "start_line": 79,
-    "end_line": 80,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 6
-  },
-  {
-    "topic": "Code Simplification",
-    "comment": "Removal of reeval_response parameter from multiple methods",
-    "confidence": "moderate",
-    "reason": "The reeval_response parameter has been removed from several method signatures, simplifying the code.",
-    "solution": "Verify that the removal of this parameter doesn't affect any calling code. Update any documentation or comments related to these methods.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/pr_description.py",
-    "start_line": 52,
-    "end_line": 52,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 3
-  },
-  {
-    "topic": "File Reorganization",
-    "comment": "Move of PR description prompts to a new file",
-    "confidence": "important",
-    "reason": "PR description prompts have been moved from code_review_prompts.py to pr_desc_prompts.py, improving code organization.",
-    "solution": "Update any import statements in other files that might be affected by this change. Ensure that all necessary prompts have been moved correctly.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/llms/prompts/pr_desc_prompts.py",
-    "start_line": 1,
-    "end_line": 92,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 4
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_335/review.md b/.experiments/code_review/sonnet-3.5/no_eval/pr_335/review.md
deleted file mode 100644
index 8c444640..00000000
--- a/.experiments/code_review/sonnet-3.5/no_eval/pr_335/review.md
+++ /dev/null
@@ -1,89 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/335
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 4
-- Critical: 0
-- Important: 3
-- Minor: 1
-- Files Affected: 2
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Code Refactoring (3 issues)</strong></summary>
-
-### 1. Removal of reevaluation logic and associated prompts
-📁 **File:** `kaizen/generator/pr_description.py:86`
-⚖️ **Severity:** 5/10
-🔍 **Description:** The code has been simplified by removing the reevaluation logic, which could potentially improve performance and reduce complexity.
-💡 **Solution:** Ensure that the removal of reevaluation doesn't impact the quality of PR descriptions. Consider adding unit tests to verify the new behavior.
-
-### 2. Change from chat_completion_with_json to chat_completion
-📁 **File:** `kaizen/generator/pr_description.py:79`
-⚖️ **Severity:** 6/10
-🔍 **Description:** The API call has been changed, which might affect the format of the response and how it's processed.
-💡 **Solution:** Ensure that the new chat_completion method returns the expected format. Update any dependent code that might be affected by this change.
-
-**Current Code:**
-```python
-resp, usage = self.provider.chat_completion(prompt, user=user)
-```
-
-**Suggested Code:**
-```python
-resp, usage = self.provider.chat_completion(prompt, user=user)
-desc = parser.extract_code_from_markdown(resp)
-```
-
-### 3. Move of PR description prompts to a new file
-📁 **File:** `kaizen/llms/prompts/pr_desc_prompts.py:1`
-⚖️ **Severity:** 4/10
-🔍 **Description:** PR description prompts have been moved from code_review_prompts.py to pr_desc_prompts.py, improving code organization.
-💡 **Solution:** Update any import statements in other files that might be affected by this change. Ensure that all necessary prompts have been moved correctly.
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (1 issues)</strong></summary>
-
-<details>
-<summary><strong>Code Simplification (1 issues)</strong></summary>
-
-### 1. Removal of reeval_response parameter from multiple methods
-📁 **File:** `kaizen/generator/pr_description.py:52`
-⚖️ **Severity:** 3/10
-🔍 **Description:** The reeval_response parameter has been removed from several method signatures, simplifying the code.
-💡 **Solution:** Verify that the removal of this parameter doesn't affect any calling code. Update any documentation or comments related to these methods.
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (anthropic.claude-3-5-sonnet-20240620-v1:0)
-{"prompt_tokens": 7614, "completion_tokens": 826, "total_tokens": 8440}
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_400/comments.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_400/comments.json
deleted file mode 100644
index f470a30b..00000000
--- a/.experiments/code_review/sonnet-3.5/no_eval/pr_400/comments.json
+++ /dev/null
@@ -1,31 +0,0 @@
-[
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "11",
-    "end_line": "11",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  },
-  {
-    "topic": "Asynchronous Testing",
-    "comment": "The tests have been updated to use async/await syntax, which is correct for testing asynchronous functions.",
-    "confidence": "critical",
-    "reason": "Proper async testing is crucial for accurate results when testing asynchronous code.",
-    "solution": "The implementation is correct. Ensure all test functions are defined with 'async def'.",
-    "actual_code": "async def test_get_web_html_normal_cases(mock_get_html, mock_nest_asyncio, html_content, expected_output):",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 76,
-    "end_line": 76,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 8
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_400/issues.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_400/issues.json
deleted file mode 100644
index ccc761ea..00000000
--- a/.experiments/code_review/sonnet-3.5/no_eval/pr_400/issues.json
+++ /dev/null
@@ -1,406 +0,0 @@
-[
-  {
-    "topic": "Test Coverage",
-    "comment": "Improved test coverage with more comprehensive test cases",
-    "confidence": "important",
-    "reason": "The new test cases cover a wider range of scenarios, including edge cases and error handling",
-    "solution": "No changes needed. The improvements are good.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
-    "start_line": 7,
-    "end_line": 64,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 2
-  },
-  {
-    "topic": "Performance Testing",
-    "comment": "Added performance testing for large inputs, but removed arbitrary time limit",
-    "confidence": "moderate",
-    "reason": "Performance testing is important, but the removal of the 1-second boundary condition is a good change as it avoids potential flaky tests",
-    "solution": "Consider adding a more flexible performance assertion based on input size",
-    "actual_code": "print(f\"Execution time:{execution_time}seconds\")",
-    "fixed_code": "assert execution_time < len(desc + original_desc) * 0.0001, f\"Execution time ({execution_time}seconds) exceeded expected limit\"",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
-    "start_line": 61,
-    "end_line": 61,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "Added error handling tests for invalid input types",
-    "confidence": "important",
-    "reason": "Proper error handling is crucial for robust code",
-    "solution": "No changes needed. The added error handling tests are beneficial.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
-    "start_line": 31,
-    "end_line": 43,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 3
-  },
-  {
-    "topic": "Test Structure",
-    "comment": "Improved test structure using pytest.mark.parametrize",
-    "confidence": "important",
-    "reason": "Parameterized tests reduce code duplication and make it easier to add new test cases",
-    "solution": "No changes needed. The use of pytest.mark.parametrize is a good practice.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
-    "start_line": 7,
-    "end_line": 28,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 2
-  },
-  {
-    "topic": "Configuration",
-    "comment": "Updated .flake8 configuration to exclude unit test directory",
-    "confidence": "moderate",
-    "reason": "Excluding unit tests from flake8 checks can be beneficial, but it's important to maintain code quality in tests as well",
-    "solution": "Consider running flake8 on test files with a separate, less strict configuration",
-    "actual_code": "exclude = docs/*, venv/*, .kaizen/unit_test/*",
-    "fixed_code": "exclude = docs/*, venv/*",
-    "file_name": ".flake8",
-    "start_line": 2,
-    "end_line": 2,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 3
-  },
-  {
-    "topic": "Environment Configuration",
-    "comment": "Added LITELLM_LOG environment variable",
-    "confidence": "moderate",
-    "reason": "Adding logging configuration can improve debugging and monitoring",
-    "solution": "Ensure this change is documented in the project's README or documentation",
-    "actual_code": "LITELLM_LOG=\"ERROR\"",
-    "fixed_code": "",
-    "file_name": ".env.example",
-    "start_line": 6,
-    "end_line": 6,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 2
-  },
-  {
-    "topic": "Test Coverage",
-    "comment": "New test cases have been added to improve test coverage, which is a positive change.",
-    "confidence": "important",
-    "reason": "Comprehensive test coverage is crucial for maintaining code quality and catching potential bugs.",
-    "solution": "Continue to add test cases for edge cases and ensure all new functionality is covered by tests.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 18,
-    "end_line": 33,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 7
-  },
-  {
-    "topic": "Code Structure",
-    "comment": "The PR_COLLAPSIBLE_TEMPLATE has been updated to use a multi-line string, which improves readability.",
-    "confidence": "moderate",
-    "reason": "Multi-line strings are more readable and maintainable for long string templates.",
-    "solution": "Consider using f-strings for better performance and readability if Python 3.6+ is supported.",
-    "actual_code": "PR_COLLAPSIBLE_TEMPLATE = \"\"\"\n<details>\n<summary>Review Comment</summary>\n<p>{comment}</p>\n<p><strong>Reason:</strong>{reason}</p>\n<p><strong>Solution:</strong>{solution}</p>\n<p><strong>Confidence:</strong>{confidence}</p>\n<p><strong>Start Line:</strong>{start_line}</p>\n<p><strong>End Line:</strong>{end_line}</p>\n<p><strong>File Name:</strong>{file_name}</p>\n<p><strong>Severity:</strong>{severity}</p>\n</details>\n\"\"\"",
-    "fixed_code": "PR_COLLAPSIBLE_TEMPLATE = f\"\"\"\n<details>\n<summary>Review Comment</summary>\n<p>{{comment}}</p>\n<p><strong>Reason:</strong>{{reason}}</p>\n<p><strong>Solution:</strong>{{solution}}</p>\n<p><strong>Confidence:</strong>{{confidence}}</p>\n<p><strong>Start Line:</strong>{{start_line}}</p>\n<p><strong>End Line:</strong>{{end_line}}</p>\n<p><strong>File Name:</strong>{{file_name}}</p>\n<p><strong>Severity:</strong>{{severity}}</p>\n</details>\n\"\"\"",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 4,
-    "end_line": 16,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 3
-  },
-  {
-    "topic": "Test Case Design",
-    "comment": "New test cases for handling missing fields and empty lists have been added, which improves robustness.",
-    "confidence": "important",
-    "reason": "Testing edge cases and error handling is crucial for ensuring the reliability of the code.",
-    "solution": "Continue to add test cases for other potential edge cases and error scenarios.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 142,
-    "end_line": 276,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 7
-  },
-  {
-    "topic": "Test Coverage",
-    "comment": "The new tests provide better coverage and include more edge cases.",
-    "confidence": "important",
-    "reason": "Comprehensive test coverage is crucial for maintaining code quality and preventing regressions.",
-    "solution": "No changes needed. The new tests are well-structured and cover various scenarios.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 25,
-    "end_line": 246,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 2
-  },
-  {
-    "topic": "Code Organization",
-    "comment": "The test functions are well-organized and use pytest fixtures effectively.",
-    "confidence": "important",
-    "reason": "Good organization improves readability and maintainability of test code.",
-    "solution": "No changes needed. The use of fixtures and utility functions is appropriate.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 7,
-    "end_line": 23,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 2
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "The tests now include checks for file writing permission issues, which is a good practice.",
-    "confidence": "important",
-    "reason": "Testing error handling scenarios ensures the code behaves correctly under various conditions.",
-    "solution": "No changes needed. The error handling tests are well-implemented.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 175,
-    "end_line": 199,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 2
-  },
-  {
-    "topic": "Test Case Naming",
-    "comment": "Test function names are descriptive and follow a consistent naming convention.",
-    "confidence": "moderate",
-    "reason": "Clear and consistent naming improves code readability and helps understand test purposes.",
-    "solution": "No changes needed. The test function names are appropriate.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 25,
-    "end_line": 246,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 3
-  },
-  {
-    "topic": "Test Implementation",
-    "comment": "The test_get_parent_folder_normal function doesn't mock os.getcwd, which may lead to inconsistent results.",
-    "confidence": "important",
-    "reason": "Not mocking external dependencies can make tests unreliable and environment-dependent.",
-    "solution": "Mock os.getcwd to ensure consistent test results across different environments.",
-    "actual_code": "def test_get_parent_folder_normal():\n    expected = os.path.dirname(os.getcwd())\n    result = get_parent_folder()\n    assert result == expected, f\"Expected{expected}, but got{result}\"",
-    "fixed_code": "def test_get_parent_folder_normal():\n    with mock.patch('os.getcwd', return_value='/home/user/project'):\n        expected = '/home/user'\n        result = get_parent_folder()\n        assert result == expected, f\"Expected{expected}, but got{result}\"",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py",
-    "start_line": 13,
-    "end_line": 16,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Test Structure",
-    "comment": "The test structure has been significantly improved with the use of parametrized tests.",
-    "confidence": "important",
-    "reason": "Parametrized tests allow for more comprehensive testing with less code duplication.",
-    "solution": "The changes are already an improvement. Consider adding more test cases to cover edge cases.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 20,
-    "end_line": 75,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 3
-  },
-  {
-    "topic": "Asynchronous Testing",
-    "comment": "The tests have been updated to use async/await syntax, which is correct for testing asynchronous functions.",
-    "confidence": "critical",
-    "reason": "Proper async testing is crucial for accurate results when testing asynchronous code.",
-    "solution": "The implementation is correct. Ensure all test functions are defined with 'async def'.",
-    "actual_code": "async def test_get_web_html_normal_cases(mock_get_html, mock_nest_asyncio, html_content, expected_output):",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 76,
-    "end_line": 76,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 8
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "A new test case for invalid URL has been added, which is a good practice for error handling.",
-    "confidence": "important",
-    "reason": "Testing error scenarios is crucial for robust code.",
-    "solution": "The implementation is correct. Consider adding more error scenarios if applicable.",
-    "actual_code": "async def test_get_web_html_invalid_url(mock_get_html, mock_nest_asyncio):",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 85,
-    "end_line": 91,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 7
-  },
-  {
-    "topic": "Performance Testing",
-    "comment": "A new test case for large content has been added, which is good for performance testing.",
-    "confidence": "moderate",
-    "reason": "Testing with large inputs helps identify potential performance issues.",
-    "solution": "The implementation is good. Consider adding assertions for execution time if performance is critical.",
-    "actual_code": "async def test_get_web_html_large_content(mock_get_html, mock_nest_asyncio):",
-    "fixed_code": "",
-    "file_name": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 93,
-    "end_line": 102,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 5
-  },
-  {
-    "topic": "Configuration Update",
-    "comment": "New 'base_model' fields have been added to the configuration file.",
-    "confidence": "important",
-    "reason": "Keeping configuration up-to-date is crucial for proper system functionality.",
-    "solution": "Ensure that the code using this configuration is updated to handle the new 'base_model' field.",
-    "actual_code": "\"base_model\": \"azure/gpt-4o-mini\"",
-    "fixed_code": "",
-    "file_name": "config.json",
-    "start_line": 15,
-    "end_line": 15,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 6
-  },
-  {
-    "topic": "Test Generation",
-    "comment": "The test generation call has been updated with new parameters.",
-    "confidence": "moderate",
-    "reason": "New parameters may affect the test generation process and output.",
-    "solution": "Ensure that the 'enable_critique' and 'verbose' options are properly handled in the generator code.",
-    "actual_code": "generator.generate_tests(\n    file_path=\"kaizen/helpers/output.py\", enable_critique=True, verbose=True\n)",
-    "fixed_code": "",
-    "file_name": "examples/unittest/main.py",
-    "start_line": 35,
-    "end_line": 37,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Code Structure",
-    "comment": "The class has been significantly refactored with improved modularity and organization.",
-    "confidence": "important",
-    "reason": "The changes introduce better separation of concerns and more focused methods.",
-    "solution": "No immediate action required, but continue to monitor for potential further improvements.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/unit_test.py",
-    "start_line": 24,
-    "end_line": 40,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 3
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "The _read_file_content method lacks error handling for file operations.",
-    "confidence": "important",
-    "reason": "File operations can fail due to various reasons (e.g., permissions, file not found).",
-    "solution": "Add try-except block to handle potential IOError or FileNotFoundError.",
-    "actual_code": "def _read_file_content(self, file_path):\n    with open(file_path, \"r\") as file:\n        return file.read()",
-    "fixed_code": "def _read_file_content(self, file_path):\n    try:\n        with open(file_path, \"r\") as file:\n            return file.read()\n    except (IOError, FileNotFoundError) as e:\n        self.logger.error(f\"Error reading file{file_path}:{e}\")\n        raise",
-    "file_name": "kaizen/generator/unit_test.py",
-    "start_line": 108,
-    "end_line": 110,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Code Duplication",
-    "comment": "The update_usage method duplicates functionality already present in the LLMProvider class.",
-    "confidence": "moderate",
-    "reason": "Duplicating functionality can lead to maintenance issues and inconsistencies.",
-    "solution": "Consider removing the update_usage method and directly using the provider's method.",
-    "actual_code": "def update_usage(self, usage):\n    self.total_usage = self.provider.update_usage(self.total_usage, usage)\n    print(f\"@ Token usage: current_step:{usage}, total:{self.total_usage}\")",
-    "fixed_code": "",
-    "file_name": "kaizen/generator/unit_test.py",
-    "start_line": 273,
-    "end_line": 275,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 5
-  },
-  {
-    "topic": "Code Improvement",
-    "comment": "The _review_tests_by_critique method could benefit from a more descriptive variable name for 'counter'.",
-    "confidence": "low",
-    "reason": "Using more descriptive variable names enhances code readability.",
-    "solution": "Rename 'counter' to 'critique_attempt' or similar for better clarity.",
-    "actual_code": "counter = 1",
-    "fixed_code": "critique_attempt = 1",
-    "file_name": "kaizen/generator/unit_test.py",
-    "start_line": 180,
-    "end_line": 180,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 2
-  },
-  {
-    "topic": "Logging Configuration",
-    "comment": "The new logging configuration might override the existing logging setup",
-    "confidence": "important",
-    "reason": "The added function `set_all_loggers_to_ERROR()` sets all loggers to ERROR level, which may interfere with the existing logging configuration set by `logging.basicConfig()`",
-    "solution": "Consider removing the `set_all_loggers_to_ERROR()` function or adjusting it to respect the `LOGLEVEL` environment variable",
-    "actual_code": "set_all_loggers_to_ERROR()",
-    "fixed_code": "# Consider removing this line or adjusting the function to respect LOGLEVEL\n# set_all_loggers_to_ERROR()",
-    "file_name": "kaizen/llms/provider.py",
-    "start_line": 23,
-    "end_line": 23,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Code Organization",
-    "comment": "The logging configuration is mixed with import statements",
-    "confidence": "moderate",
-    "reason": "Logging setup is typically done after imports for better code organization",
-    "solution": "Move the logging configuration to a separate section after all imports",
-    "actual_code": "def set_all_loggers_to_ERROR():\n    print(\"All Loggers and their levels:\")\n    for name, logger in logging.Logger.manager.loggerDict.items():\n        if isinstance(logger, logging.Logger):\n            print(f\"Logger:{name}, Level:{logging.getLevelName(logger.level)}\")\n            logging.getLogger(name).setLevel(logging.ERROR)\n        else:\n            print(f\"PlaceHolder:{name}\")\n\n\nset_all_loggers_to_ERROR()\n\n# Set litellm log level to ERROR\nlogging.getLogger(\"LiteLLM\").setLevel(logging.ERROR)\nlogging.getLogger(\"LiteLLM Router\").setLevel(logging.ERROR)\nlogging.getLogger(\"LiteLLM Proxy\").setLevel(logging.ERROR)",
-    "fixed_code": "import logging\n\n# Rest of the imports...\n\n# Logging configuration\ndef set_all_loggers_to_ERROR():\n    print(\"All Loggers and their levels:\")\n    for name, logger in logging.Logger.manager.loggerDict.items():\n        if isinstance(logger, logging.Logger):\n            print(f\"Logger:{name}, Level:{logging.getLevelName(logger.level)}\")\n            logging.getLogger(name).setLevel(logging.ERROR)\n        else:\n            print(f\"PlaceHolder:{name}\")\n\nset_all_loggers_to_ERROR()\n\n# Set litellm log level to ERROR\nlogging.getLogger(\"LiteLLM\").setLevel(logging.ERROR)\nlogging.getLogger(\"LiteLLM Router\").setLevel(logging.ERROR)\nlogging.getLogger(\"LiteLLM Proxy\").setLevel(logging.ERROR)",
-    "file_name": "kaizen/llms/provider.py",
-    "start_line": 13,
-    "end_line": 28,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "11",
-    "end_line": "11",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_400/review.md b/.experiments/code_review/sonnet-3.5/no_eval/pr_400/review.md
deleted file mode 100644
index 73268d79..00000000
--- a/.experiments/code_review/sonnet-3.5/no_eval/pr_400/review.md
+++ /dev/null
@@ -1,435 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/400
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 27
-- Critical: 2
-- Important: 15
-- Minor: 9
-- Files Affected: 11
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Asynchronous Testing (2 issues)</strong></summary>
-
-### 1. The tests have been updated to use async/await syntax, which is correct for testing asynchronous functions.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_web_html.py:76`
-⚖️ **Severity:** 8/10
-🔍 **Description:** Proper async testing is crucial for accurate results when testing asynchronous code.
-💡 **Solution:** The implementation is correct. Ensure all test functions are defined with 'async def'.
-
-**Current Code:**
-```python
-async def test_get_web_html_normal_cases(mock_get_html, mock_nest_asyncio, html_content, expected_output):
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 2. Changes made to sensitive file
-📁 **File:** `config.json:11`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to config.json, which needs review
-💡 **Solution:** NA
-
-</details>
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Test Coverage (15 issues)</strong></summary>
-
-### 1. Improved test coverage with more comprehensive test cases
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_description.py:7`
-⚖️ **Severity:** 2/10
-🔍 **Description:** The new test cases cover a wider range of scenarios, including edge cases and error handling
-💡 **Solution:** No changes needed. The improvements are good.
-
-### 2. Added error handling tests for invalid input types
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_description.py:31`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Proper error handling is crucial for robust code
-💡 **Solution:** No changes needed. The added error handling tests are beneficial.
-
-### 3. Improved test structure using pytest.mark.parametrize
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_description.py:7`
-⚖️ **Severity:** 2/10
-🔍 **Description:** Parameterized tests reduce code duplication and make it easier to add new test cases
-💡 **Solution:** No changes needed. The use of pytest.mark.parametrize is a good practice.
-
-### 4. New test cases have been added to improve test coverage, which is a positive change.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py:18`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Comprehensive test coverage is crucial for maintaining code quality and catching potential bugs.
-💡 **Solution:** Continue to add test cases for edge cases and ensure all new functionality is covered by tests.
-
-### 5. New test cases for handling missing fields and empty lists have been added, which improves robustness.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py:142`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Testing edge cases and error handling is crucial for ensuring the reliability of the code.
-💡 **Solution:** Continue to add test cases for other potential edge cases and error scenarios.
-
-### 6. The new tests provide better coverage and include more edge cases.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_test_files.py:25`
-⚖️ **Severity:** 2/10
-🔍 **Description:** Comprehensive test coverage is crucial for maintaining code quality and preventing regressions.
-💡 **Solution:** No changes needed. The new tests are well-structured and cover various scenarios.
-
-### 7. The test functions are well-organized and use pytest fixtures effectively.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_test_files.py:7`
-⚖️ **Severity:** 2/10
-🔍 **Description:** Good organization improves readability and maintainability of test code.
-💡 **Solution:** No changes needed. The use of fixtures and utility functions is appropriate.
-
-### 8. The tests now include checks for file writing permission issues, which is a good practice.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_test_files.py:175`
-⚖️ **Severity:** 2/10
-🔍 **Description:** Testing error handling scenarios ensures the code behaves correctly under various conditions.
-💡 **Solution:** No changes needed. The error handling tests are well-implemented.
-
-### 9. The test_get_parent_folder_normal function doesn't mock os.getcwd, which may lead to inconsistent results.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py:13`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Not mocking external dependencies can make tests unreliable and environment-dependent.
-💡 **Solution:** Mock os.getcwd to ensure consistent test results across different environments.
-
-**Current Code:**
-```python
-def test_get_parent_folder_normal():
-    expected = os.path.dirname(os.getcwd())
-    result = get_parent_folder()
-    assert result == expected, f"Expected{expected}, but got{result}"
-```
-
-**Suggested Code:**
-```python
-def test_get_parent_folder_normal():
-    with mock.patch('os.getcwd', return_value='/home/user/project'):
-        expected = '/home/user'
-        result = get_parent_folder()
-        assert result == expected, f"Expected{expected}, but got{result}"
-```
-
-### 10. The test structure has been significantly improved with the use of parametrized tests.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_web_html.py:20`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Parametrized tests allow for more comprehensive testing with less code duplication.
-💡 **Solution:** The changes are already an improvement. Consider adding more test cases to cover edge cases.
-
-### 11. A new test case for invalid URL has been added, which is a good practice for error handling.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_web_html.py:85`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Testing error scenarios is crucial for robust code.
-💡 **Solution:** The implementation is correct. Consider adding more error scenarios if applicable.
-
-**Current Code:**
-```python
-async def test_get_web_html_invalid_url(mock_get_html, mock_nest_asyncio):
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 12. New 'base_model' fields have been added to the configuration file.
-📁 **File:** `config.json:15`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Keeping configuration up-to-date is crucial for proper system functionality.
-💡 **Solution:** Ensure that the code using this configuration is updated to handle the new 'base_model' field.
-
-**Current Code:**
-```python
-"base_model": "azure/gpt-4o-mini"
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 13. The class has been significantly refactored with improved modularity and organization.
-📁 **File:** `kaizen/generator/unit_test.py:24`
-⚖️ **Severity:** 3/10
-🔍 **Description:** The changes introduce better separation of concerns and more focused methods.
-💡 **Solution:** No immediate action required, but continue to monitor for potential further improvements.
-
-### 14. The _read_file_content method lacks error handling for file operations.
-📁 **File:** `kaizen/generator/unit_test.py:108`
-⚖️ **Severity:** 7/10
-🔍 **Description:** File operations can fail due to various reasons (e.g., permissions, file not found).
-💡 **Solution:** Add try-except block to handle potential IOError or FileNotFoundError.
-
-**Current Code:**
-```python
-def _read_file_content(self, file_path):
-    with open(file_path, "r") as file:
-        return file.read()
-```
-
-**Suggested Code:**
-```python
-def _read_file_content(self, file_path):
-    try:
-        with open(file_path, "r") as file:
-            return file.read()
-    except (IOError, FileNotFoundError) as e:
-        self.logger.error(f"Error reading file{file_path}:{e}")
-        raise
-```
-
-### 15. The new logging configuration might override the existing logging setup
-📁 **File:** `kaizen/llms/provider.py:23`
-⚖️ **Severity:** 7/10
-🔍 **Description:** The added function `set_all_loggers_to_ERROR()` sets all loggers to ERROR level, which may interfere with the existing logging configuration set by `logging.basicConfig()`
-💡 **Solution:** Consider removing the `set_all_loggers_to_ERROR()` function or adjusting it to respect the `LOGLEVEL` environment variable
-
-**Current Code:**
-```python
-set_all_loggers_to_ERROR()
-```
-
-**Suggested Code:**
-```python
-# Consider removing this line or adjusting the function to respect LOGLEVEL
-# set_all_loggers_to_ERROR()
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (10 issues)</strong></summary>
-
-<details>
-<summary><strong>Performance Testing (9 issues)</strong></summary>
-
-### 1. Added performance testing for large inputs, but removed arbitrary time limit
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_description.py:61`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Performance testing is important, but the removal of the 1-second boundary condition is a good change as it avoids potential flaky tests
-💡 **Solution:** Consider adding a more flexible performance assertion based on input size
-
-**Current Code:**
-```python
-print(f"Execution time:{execution_time}seconds")
-```
-
-**Suggested Code:**
-```python
-assert execution_time < len(desc + original_desc) * 0.0001, f"Execution time ({execution_time}seconds) exceeded expected limit"
-```
-
-### 2. Updated .flake8 configuration to exclude unit test directory
-📁 **File:** `.flake8:2`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Excluding unit tests from flake8 checks can be beneficial, but it's important to maintain code quality in tests as well
-💡 **Solution:** Consider running flake8 on test files with a separate, less strict configuration
-
-**Current Code:**
-```python
-exclude = docs/*, venv/*, .kaizen/unit_test/*
-```
-
-**Suggested Code:**
-```python
-exclude = docs/*, venv/*
-```
-
-### 3. Added LITELLM_LOG environment variable
-📁 **File:** `.env.example:6`
-⚖️ **Severity:** 2/10
-🔍 **Description:** Adding logging configuration can improve debugging and monitoring
-💡 **Solution:** Ensure this change is documented in the project's README or documentation
-
-**Current Code:**
-```python
-LITELLM_LOG="ERROR"
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 4. The PR_COLLAPSIBLE_TEMPLATE has been updated to use a multi-line string, which improves readability.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py:4`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Multi-line strings are more readable and maintainable for long string templates.
-💡 **Solution:** Consider using f-strings for better performance and readability if Python 3.6+ is supported.
-
-**Current Code:**
-```python
-PR_COLLAPSIBLE_TEMPLATE = """
-<details>
-<summary>Review Comment</summary>
-<p>{comment}</p>
-<p><strong>Reason:</strong>{reason}</p>
-<p><strong>Solution:</strong>{solution}</p>
-<p><strong>Confidence:</strong>{confidence}</p>
-<p><strong>Start Line:</strong>{start_line}</p>
-<p><strong>End Line:</strong>{end_line}</p>
-<p><strong>File Name:</strong>{file_name}</p>
-<p><strong>Severity:</strong>{severity}</p>
-</details>
-"""
-```
-
-**Suggested Code:**
-```python
-PR_COLLAPSIBLE_TEMPLATE = f"""
-<details>
-<summary>Review Comment</summary>
-<p>{{comment}}</p>
-<p><strong>Reason:</strong>{{reason}}</p>
-<p><strong>Solution:</strong>{{solution}}</p>
-<p><strong>Confidence:</strong>{{confidence}}</p>
-<p><strong>Start Line:</strong>{{start_line}}</p>
-<p><strong>End Line:</strong>{{end_line}}</p>
-<p><strong>File Name:</strong>{{file_name}}</p>
-<p><strong>Severity:</strong>{{severity}}</p>
-</details>
-"""
-```
-
-### 5. Test function names are descriptive and follow a consistent naming convention.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_create_test_files.py:25`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Clear and consistent naming improves code readability and helps understand test purposes.
-💡 **Solution:** No changes needed. The test function names are appropriate.
-
-### 6. A new test case for large content has been added, which is good for performance testing.
-📁 **File:** `.kaizen/unit_test/kaizen/helpers/test_get_web_html.py:93`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Testing with large inputs helps identify potential performance issues.
-💡 **Solution:** The implementation is good. Consider adding assertions for execution time if performance is critical.
-
-**Current Code:**
-```python
-async def test_get_web_html_large_content(mock_get_html, mock_nest_asyncio):
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 7. The test generation call has been updated with new parameters.
-📁 **File:** `examples/unittest/main.py:35`
-⚖️ **Severity:** 4/10
-🔍 **Description:** New parameters may affect the test generation process and output.
-💡 **Solution:** Ensure that the 'enable_critique' and 'verbose' options are properly handled in the generator code.
-
-**Current Code:**
-```python
-generator.generate_tests(
-    file_path="kaizen/helpers/output.py", enable_critique=True, verbose=True
-)
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 8. The update_usage method duplicates functionality already present in the LLMProvider class.
-📁 **File:** `kaizen/generator/unit_test.py:273`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Duplicating functionality can lead to maintenance issues and inconsistencies.
-💡 **Solution:** Consider removing the update_usage method and directly using the provider's method.
-
-**Current Code:**
-```python
-def update_usage(self, usage):
-    self.total_usage = self.provider.update_usage(self.total_usage, usage)
-    print(f"@ Token usage: current_step:{usage}, total:{self.total_usage}")
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 9. The logging configuration is mixed with import statements
-📁 **File:** `kaizen/llms/provider.py:13`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Logging setup is typically done after imports for better code organization
-💡 **Solution:** Move the logging configuration to a separate section after all imports
-
-**Current Code:**
-```python
-def set_all_loggers_to_ERROR():
-    print("All Loggers and their levels:")
-    for name, logger in logging.Logger.manager.loggerDict.items():
-        if isinstance(logger, logging.Logger):
-            print(f"Logger:{name}, Level:{logging.getLevelName(logger.level)}")
-            logging.getLogger(name).setLevel(logging.ERROR)
-        else:
-            print(f"PlaceHolder:{name}")
-
-
-set_all_loggers_to_ERROR()
-
-# Set litellm log level to ERROR
-logging.getLogger("LiteLLM").setLevel(logging.ERROR)
-logging.getLogger("LiteLLM Router").setLevel(logging.ERROR)
-logging.getLogger("LiteLLM Proxy").setLevel(logging.ERROR)
-```
-
-**Suggested Code:**
-```python
-import logging
-
-# Rest of the imports...
-
-# Logging configuration
-def set_all_loggers_to_ERROR():
-    print("All Loggers and their levels:")
-    for name, logger in logging.Logger.manager.loggerDict.items():
-        if isinstance(logger, logging.Logger):
-            print(f"Logger:{name}, Level:{logging.getLevelName(logger.level)}")
-            logging.getLogger(name).setLevel(logging.ERROR)
-        else:
-            print(f"PlaceHolder:{name}")
-
-set_all_loggers_to_ERROR()
-
-# Set litellm log level to ERROR
-logging.getLogger("LiteLLM").setLevel(logging.ERROR)
-logging.getLogger("LiteLLM Router").setLevel(logging.ERROR)
-logging.getLogger("LiteLLM Proxy").setLevel(logging.ERROR)
-```
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (anthropic.claude-3-5-sonnet-20240620-v1:0)
-{"prompt_tokens": 42996, "completion_tokens": 6157, "total_tokens": 49153}
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_476/comments.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_476/comments.json
deleted file mode 100644
index a9d40eac..00000000
--- a/.experiments/code_review/sonnet-3.5/no_eval/pr_476/comments.json
+++ /dev/null
@@ -1,16 +0,0 @@
-[
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_476/issues.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_476/issues.json
deleted file mode 100644
index c12e22ed..00000000
--- a/.experiments/code_review/sonnet-3.5/no_eval/pr_476/issues.json
+++ /dev/null
@@ -1,106 +0,0 @@
-[
-  {
-    "topic": "Error Handling",
-    "comment": "Exception handling is too broad and prints a generic error message.",
-    "confidence": "important",
-    "reason": "Catching all exceptions and printing a generic error message can hide important errors and make debugging difficult.",
-    "solution": "Catch specific exceptions and provide more informative error messages.",
-    "actual_code": "except Exception:\n    print(\"Error\")",
-    "fixed_code": "except KeyError as e:\n    print(f\"Invalid confidence level:{e}\")\nexcept Exception as e:\n    print(f\"Unexpected error:{e}\")",
-    "file_name": "github_app/github_helper/pull_requests.py",
-    "start_line": 140,
-    "end_line": 141,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Code Efficiency",
-    "comment": "The sort_files function implements a manual insertion sort, which is inefficient for large lists.",
-    "confidence": "important",
-    "reason": "Insertion sort has O(n^2) time complexity, which can be slow for large numbers of files.",
-    "solution": "Use Python's built-in sorted() function with a key function for better performance.",
-    "actual_code": "def sort_files(files):\n    sorted_files =[]\n    for file in files:\n        min_index = len(sorted_files)\n        file_name = file[\"filename\"]\n        for i, sorted_file in enumerate(sorted_files):\n            if file_name < sorted_file[\"filename\"]:\n                min_index = i\n                break\n        sorted_files.insert(min_index, file)\n    return sorted_files",
-    "fixed_code": "def sort_files(files):\n    return sorted(files, key=lambda x: x[\"filename\"])",
-    "file_name": "github_app/github_helper/pull_requests.py",
-    "start_line": 184,
-    "end_line": 194,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 6
-  },
-  {
-    "topic": "Code Simplification",
-    "comment": "The generate_tests function can be simplified using a list comprehension.",
-    "confidence": "moderate",
-    "reason": "The current implementation is unnecessarily verbose for a simple operation.",
-    "solution": "Use a list comprehension to create the list of filenames.",
-    "actual_code": "def generate_tests(pr_files):\n    return[f[\"filename\"] for f in pr_files]",
-    "fixed_code": "def generate_tests(pr_files):\n    return[f[\"filename\"] for f in pr_files]",
-    "file_name": "github_app/github_helper/pull_requests.py",
-    "start_line": 199,
-    "end_line": 200,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 2
-  },
-  {
-    "topic": "Code Consistency",
-    "comment": "Inconsistent use of print statements for debugging.",
-    "confidence": "low",
-    "reason": "Some print statements are commented out while others are added, which may lead to inconsistent debugging output.",
-    "solution": "Decide on a consistent approach for debug logging, preferably using a proper logging system.",
-    "actual_code": "print(\"diff: \", diff_text)\nprint(\"pr_files\", pr_files)",
-    "fixed_code": "import logging\n\nlogging.debug(f\"diff:{diff_text}\")\nlogging.debug(f\"pr_files:{pr_files}\")",
-    "file_name": "examples/code_review/main.py",
-    "start_line": 21,
-    "end_line": 22,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity_level": 3
-  },
-  {
-    "topic": "Code Improvement",
-    "comment": "The create_pr_review_text function now includes a code_quality parameter, which is a good improvement.",
-    "confidence": "moderate",
-    "reason": "Including code quality in the review text provides more comprehensive feedback.",
-    "solution": "No change needed, this is a positive improvement.",
-    "actual_code": "review_desc = create_pr_review_text(topics, code_quality=review_data.code_quality)",
-    "fixed_code": "",
-    "file_name": "examples/code_review/main.py",
-    "start_line": 36,
-    "end_line": 36,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity_level": 1
-  },
-  {
-    "topic": "Configuration",
-    "comment": "Removal of 'enable_observability_logging' from config.json",
-    "confidence": "moderate",
-    "reason": "Removing configuration options without proper documentation or migration path can lead to issues for existing users.",
-    "solution": "If the feature is no longer supported, provide a migration guide or deprecation notice.",
-    "actual_code": "",
-    "fixed_code": "",
-    "file_name": "config.json",
-    "start_line": 4,
-    "end_line": 4,
-    "side": "LEFT",
-    "sentiment": "neutral",
-    "severity_level": 4
-  },
-  {
-    "topic": "Configuration",
-    "comment": "Changes made to sensitive file",
-    "confidence": "critical",
-    "reason": "Changes were made to config.json, which needs review",
-    "solution": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_name": "config.json",
-    "sentiment": "negative",
-    "severity_level": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_476/review.md b/.experiments/code_review/sonnet-3.5/no_eval/pr_476/review.md
deleted file mode 100644
index 0c63220f..00000000
--- a/.experiments/code_review/sonnet-3.5/no_eval/pr_476/review.md
+++ /dev/null
@@ -1,154 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/476
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 7
-- Critical: 1
-- Important: 2
-- Minor: 3
-- Files Affected: 3
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Configuration (1 issues)</strong></summary>
-
-### 1. Changes made to sensitive file
-📁 **File:** `config.json:1`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes were made to config.json, which needs review
-💡 **Solution:** NA
-
-</details>
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Error Handling (2 issues)</strong></summary>
-
-### 1. Exception handling is too broad and prints a generic error message.
-📁 **File:** `github_app/github_helper/pull_requests.py:140`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Catching all exceptions and printing a generic error message can hide important errors and make debugging difficult.
-💡 **Solution:** Catch specific exceptions and provide more informative error messages.
-
-**Current Code:**
-```python
-except Exception:
-    print("Error")
-```
-
-**Suggested Code:**
-```python
-except KeyError as e:
-    print(f"Invalid confidence level:{e}")
-except Exception as e:
-    print(f"Unexpected error:{e}")
-```
-
-### 2. The sort_files function implements a manual insertion sort, which is inefficient for large lists.
-📁 **File:** `github_app/github_helper/pull_requests.py:184`
-⚖️ **Severity:** 6/10
-🔍 **Description:** Insertion sort has O(n^2) time complexity, which can be slow for large numbers of files.
-💡 **Solution:** Use Python's built-in sorted() function with a key function for better performance.
-
-**Current Code:**
-```python
-def sort_files(files):
-    sorted_files =[]
-    for file in files:
-        min_index = len(sorted_files)
-        file_name = file["filename"]
-        for i, sorted_file in enumerate(sorted_files):
-            if file_name < sorted_file["filename"]:
-                min_index = i
-                break
-        sorted_files.insert(min_index, file)
-    return sorted_files
-```
-
-**Suggested Code:**
-```python
-def sort_files(files):
-    return sorted(files, key=lambda x: x["filename"])
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (4 issues)</strong></summary>
-
-<details>
-<summary><strong>Code Simplification (3 issues)</strong></summary>
-
-### 1. The generate_tests function can be simplified using a list comprehension.
-📁 **File:** `github_app/github_helper/pull_requests.py:199`
-⚖️ **Severity:** 2/10
-🔍 **Description:** The current implementation is unnecessarily verbose for a simple operation.
-💡 **Solution:** Use a list comprehension to create the list of filenames.
-
-**Current Code:**
-```python
-def generate_tests(pr_files):
-    return[f["filename"] for f in pr_files]
-```
-
-**Suggested Code:**
-```python
-def generate_tests(pr_files):
-    return[f["filename"] for f in pr_files]
-```
-
-### 2. The create_pr_review_text function now includes a code_quality parameter, which is a good improvement.
-📁 **File:** `examples/code_review/main.py:36`
-⚖️ **Severity:** 1/10
-🔍 **Description:** Including code quality in the review text provides more comprehensive feedback.
-💡 **Solution:** No change needed, this is a positive improvement.
-
-**Current Code:**
-```python
-review_desc = create_pr_review_text(topics, code_quality=review_data.code_quality)
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 3. Removal of 'enable_observability_logging' from config.json
-📁 **File:** `config.json:4`
-⚖️ **Severity:** 4/10
-🔍 **Description:** Removing configuration options without proper documentation or migration path can lead to issues for existing users.
-💡 **Solution:** If the feature is no longer supported, provide a migration guide or deprecation notice.
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (anthropic.claude-3-5-sonnet-20240620-v1:0)
-{"prompt_tokens": 4511, "completion_tokens": 1381, "total_tokens": 5892}
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_5/comments.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_5/comments.json
deleted file mode 100644
index eddca222..00000000
--- a/.experiments/code_review/sonnet-3.5/no_eval/pr_5/comments.json
+++ /dev/null
@@ -1,47 +0,0 @@
-[
-  {
-    "topic": "Error Handling",
-    "comment": "The API call in process_applicant() lacks a retry mechanism.",
-    "confidence": "critical",
-    "reason": "Without retries, temporary network issues could cause the application to fail.",
-    "solution": "Implement a retry mechanism with exponential backoff for the API call.",
-    "actual_code": "response = completion(\n    model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n)",
-    "fixed_code": "from tenacity import retry, stop_after_attempt, wait_exponential\n\n@retry(stop=stop_after_attempt(3), wait=wait_exponential(multiplier=1, min=4, max=10))\ndef make_completion_call(model, messages):\n    return completion(model=model, messages=messages)\n\nresponse = make_completion_call(\n    model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"),\n    messages=messages\n)",
-    "file_name": "main.py",
-    "start_line": 66,
-    "end_line": 68,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "Potential division by zero when calculating token percentages.",
-    "confidence": "critical",
-    "reason": "If total_tokens is zero, it could lead to a runtime error.",
-    "solution": "Add a check to avoid division by zero when calculating percentages.",
-    "actual_code": "print(\"\\nProcessing Summary:\")\nprint(f\"Total applicants processed:{len(df)}\")\nprint(f\"Total tokens used:{total_tokens:,}\")\nprint(f\"  - Input tokens:{total_input_tokens:,}\")\nprint(f\"  - Output tokens:{total_output_tokens:,}\")",
-    "fixed_code": "print(\"\\nProcessing Summary:\")\nprint(f\"Total applicants processed:{len(df)}\")\nprint(f\"Total tokens used:{total_tokens:,}\")\nif total_tokens > 0:\n    print(f\"  - Input tokens:{total_input_tokens:,}({total_input_tokens/total_tokens:.2%})\")\n    print(f\"  - Output tokens:{total_output_tokens:,}({total_output_tokens/total_tokens:.2%})\")\nelse:\n    print(\"  - No tokens used.\")",
-    "file_name": "main.py",
-    "start_line": 159,
-    "end_line": 163,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Error Logging",
-    "comment": "JSONDecodeError is caught but not logged, leading to silent failures.",
-    "confidence": "critical",
-    "reason": "Silent failures make debugging difficult and may hide important issues.",
-    "solution": "Add logging for the JSONDecodeError to track parsing failures.",
-    "actual_code": "except json.JSONDecodeError:\n    # Critical: Silent failure without logging\n    result ={\n        key: \"\"\n        for key in[\n            \"feedback\",\n            \"review\",\n            \"should_interview\",\n            \"rating\",\n            \"input_tokens\",\n            \"output_tokens\",\n        ]\n}",
-    "fixed_code": "import logging\n\nlogging.basicConfig(level=logging.ERROR)\nlogger = logging.getLogger(__name__)\n\ntry:\n    parsed_content = extract_json(content)\n    # ... existing code ...\nexcept json.JSONDecodeError as e:\n    logger.error(f\"Failed to parse JSON content:{e}\")\n    result ={\n        key: \"\"\n        for key in[\n            \"feedback\",\n            \"review\",\n            \"should_interview\",\n            \"rating\",\n            \"input_tokens\",\n            \"output_tokens\",\n        ]\n}",
-    "file_name": "main.py",
-    "start_line": 82,
-    "end_line": 94,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_5/issues.json b/.experiments/code_review/sonnet-3.5/no_eval/pr_5/issues.json
deleted file mode 100644
index 1c775bfe..00000000
--- a/.experiments/code_review/sonnet-3.5/no_eval/pr_5/issues.json
+++ /dev/null
@@ -1,107 +0,0 @@
-[
-  {
-    "topic": "Unused Import",
-    "comment": "The 'random' module is imported but never used in the code.",
-    "confidence": "important",
-    "reason": "Unused imports clutter the code and may lead to confusion.",
-    "solution": "Remove the unused import to improve code clarity.",
-    "actual_code": "import random  # Unused import",
-    "fixed_code": "# Remove this line",
-    "file_name": "main.py",
-    "start_line": 8,
-    "end_line": 8,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 3
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "The API call in process_applicant() lacks a retry mechanism.",
-    "confidence": "critical",
-    "reason": "Without retries, temporary network issues could cause the application to fail.",
-    "solution": "Implement a retry mechanism with exponential backoff for the API call.",
-    "actual_code": "response = completion(\n    model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n)",
-    "fixed_code": "from tenacity import retry, stop_after_attempt, wait_exponential\n\n@retry(stop=stop_after_attempt(3), wait=wait_exponential(multiplier=1, min=4, max=10))\ndef make_completion_call(model, messages):\n    return completion(model=model, messages=messages)\n\nresponse = make_completion_call(\n    model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"),\n    messages=messages\n)",
-    "file_name": "main.py",
-    "start_line": 66,
-    "end_line": 68,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 8
-  },
-  {
-    "topic": "Error Logging",
-    "comment": "JSONDecodeError is caught but not logged, leading to silent failures.",
-    "confidence": "critical",
-    "reason": "Silent failures make debugging difficult and may hide important issues.",
-    "solution": "Add logging for the JSONDecodeError to track parsing failures.",
-    "actual_code": "except json.JSONDecodeError:\n    # Critical: Silent failure without logging\n    result ={\n        key: \"\"\n        for key in[\n            \"feedback\",\n            \"review\",\n            \"should_interview\",\n            \"rating\",\n            \"input_tokens\",\n            \"output_tokens\",\n        ]\n}",
-    "fixed_code": "import logging\n\nlogging.basicConfig(level=logging.ERROR)\nlogger = logging.getLogger(__name__)\n\ntry:\n    parsed_content = extract_json(content)\n    # ... existing code ...\nexcept json.JSONDecodeError as e:\n    logger.error(f\"Failed to parse JSON content:{e}\")\n    result ={\n        key: \"\"\n        for key in[\n            \"feedback\",\n            \"review\",\n            \"should_interview\",\n            \"rating\",\n            \"input_tokens\",\n            \"output_tokens\",\n        ]\n}",
-    "file_name": "main.py",
-    "start_line": 82,
-    "end_line": 94,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Performance",
-    "comment": "Inefficient progress printing in non-tqdm mode.",
-    "confidence": "important",
-    "reason": "Frequent console updates can slow down processing, especially for large datasets.",
-    "solution": "Update progress less frequently, e.g., every 1% or 5% of completion.",
-    "actual_code": "print(f\"\\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}\", end=\"\", flush=True)",
-    "fixed_code": "if index % max(1, len(df) // 100) == 0:  # Update every 1%\n    print(f\"\\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}\", end=\"\", flush=True)",
-    "file_name": "main.py",
-    "start_line": 122,
-    "end_line": 122,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 5
-  },
-  {
-    "topic": "Redundant Code",
-    "comment": "Unnecessary check for empty DataFrame in main function.",
-    "confidence": "moderate",
-    "reason": "The check is redundant as an empty DataFrame would not affect the subsequent operations.",
-    "solution": "Remove the unnecessary check to simplify the code.",
-    "actual_code": "if len(df) == 0:\n    return",
-    "fixed_code": "# Remove these lines",
-    "file_name": "main.py",
-    "start_line": 142,
-    "end_line": 143,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 3
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "Potential division by zero when calculating token percentages.",
-    "confidence": "critical",
-    "reason": "If total_tokens is zero, it could lead to a runtime error.",
-    "solution": "Add a check to avoid division by zero when calculating percentages.",
-    "actual_code": "print(\"\\nProcessing Summary:\")\nprint(f\"Total applicants processed:{len(df)}\")\nprint(f\"Total tokens used:{total_tokens:,}\")\nprint(f\"  - Input tokens:{total_input_tokens:,}\")\nprint(f\"  - Output tokens:{total_output_tokens:,}\")",
-    "fixed_code": "print(\"\\nProcessing Summary:\")\nprint(f\"Total applicants processed:{len(df)}\")\nprint(f\"Total tokens used:{total_tokens:,}\")\nif total_tokens > 0:\n    print(f\"  - Input tokens:{total_input_tokens:,}({total_input_tokens/total_tokens:.2%})\")\n    print(f\"  - Output tokens:{total_output_tokens:,}({total_output_tokens/total_tokens:.2%})\")\nelse:\n    print(\"  - No tokens used.\")",
-    "file_name": "main.py",
-    "start_line": 159,
-    "end_line": 163,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 7
-  },
-  {
-    "topic": "Error Handling",
-    "comment": "No error handling for file not found in main function.",
-    "confidence": "important",
-    "reason": "The program may crash if the specified CSV file doesn't exist.",
-    "solution": "Add try-except block to handle FileNotFoundError.",
-    "actual_code": "main(input_file)",
-    "fixed_code": "try:\n    main(input_file)\nexcept FileNotFoundError:\n    print(f\"Error: The file '{input_file}' was not found.\")\nexcept Exception as e:\n    print(f\"An error occurred:{e}\")",
-    "file_name": "main.py",
-    "start_line": 175,
-    "end_line": 175,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity_level": 6
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/sonnet-3.5/no_eval/pr_5/review.md b/.experiments/code_review/sonnet-3.5/no_eval/pr_5/review.md
deleted file mode 100644
index a4a27b42..00000000
--- a/.experiments/code_review/sonnet-3.5/no_eval/pr_5/review.md
+++ /dev/null
@@ -1,234 +0,0 @@
-PR URL: https://github.com/sauravpanda/applicant-screening/pull/5
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 7
-- Critical: 3
-- Important: 3
-- Minor: 1
-- Files Affected: 1
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Error Handling (3 issues)</strong></summary>
-
-### 1. The API call in process_applicant() lacks a retry mechanism.
-📁 **File:** `main.py:66`
-⚖️ **Severity:** 8/10
-🔍 **Description:** Without retries, temporary network issues could cause the application to fail.
-💡 **Solution:** Implement a retry mechanism with exponential backoff for the API call.
-
-**Current Code:**
-```python
-response = completion(
-    model=os.environ.get("model", "anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1"), messages=messages
-)
-```
-
-**Suggested Code:**
-```python
-from tenacity import retry, stop_after_attempt, wait_exponential
-
-@retry(stop=stop_after_attempt(3), wait=wait_exponential(multiplier=1, min=4, max=10))
-def make_completion_call(model, messages):
-    return completion(model=model, messages=messages)
-
-response = make_completion_call(
-    model=os.environ.get("model", "anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1"),
-    messages=messages
-)
-```
-
-### 2. JSONDecodeError is caught but not logged, leading to silent failures.
-📁 **File:** `main.py:82`
-⚖️ **Severity:** 7/10
-🔍 **Description:** Silent failures make debugging difficult and may hide important issues.
-💡 **Solution:** Add logging for the JSONDecodeError to track parsing failures.
-
-**Current Code:**
-```python
-except json.JSONDecodeError:
-    # Critical: Silent failure without logging
-    result ={
-        key: ""
-        for key in[
-            "feedback",
-            "review",
-            "should_interview",
-            "rating",
-            "input_tokens",
-            "output_tokens",
-        ]
-}
-```
-
-**Suggested Code:**
-```python
-import logging
-
-logging.basicConfig(level=logging.ERROR)
-logger = logging.getLogger(__name__)
-
-try:
-    parsed_content = extract_json(content)
-    # ... existing code ...
-except json.JSONDecodeError as e:
-    logger.error(f"Failed to parse JSON content:{e}")
-    result ={
-        key: ""
-        for key in[
-            "feedback",
-            "review",
-            "should_interview",
-            "rating",
-            "input_tokens",
-            "output_tokens",
-        ]
-}
-```
-
-### 3. Potential division by zero when calculating token percentages.
-📁 **File:** `main.py:159`
-⚖️ **Severity:** 7/10
-🔍 **Description:** If total_tokens is zero, it could lead to a runtime error.
-💡 **Solution:** Add a check to avoid division by zero when calculating percentages.
-
-**Current Code:**
-```python
-print("\nProcessing Summary:")
-print(f"Total applicants processed:{len(df)}")
-print(f"Total tokens used:{total_tokens:,}")
-print(f"  - Input tokens:{total_input_tokens:,}")
-print(f"  - Output tokens:{total_output_tokens:,}")
-```
-
-**Suggested Code:**
-```python
-print("\nProcessing Summary:")
-print(f"Total applicants processed:{len(df)}")
-print(f"Total tokens used:{total_tokens:,}")
-if total_tokens > 0:
-    print(f"  - Input tokens:{total_input_tokens:,}({total_input_tokens/total_tokens:.2%})")
-    print(f"  - Output tokens:{total_output_tokens:,}({total_output_tokens/total_tokens:.2%})")
-else:
-    print("  - No tokens used.")
-```
-
-</details>
-
-## 🟠 Refinement Suggestions:
-These are not critical issues, but addressing them could further improve the code:
-
-<details>
-<summary><strong>Unused Import (3 issues)</strong></summary>
-
-### 1. The 'random' module is imported but never used in the code.
-📁 **File:** `main.py:8`
-⚖️ **Severity:** 3/10
-🔍 **Description:** Unused imports clutter the code and may lead to confusion.
-💡 **Solution:** Remove the unused import to improve code clarity.
-
-**Current Code:**
-```python
-import random  # Unused import
-```
-
-**Suggested Code:**
-```python
-# Remove this line
-```
-
-### 2. Inefficient progress printing in non-tqdm mode.
-📁 **File:** `main.py:122`
-⚖️ **Severity:** 5/10
-🔍 **Description:** Frequent console updates can slow down processing, especially for large datasets.
-💡 **Solution:** Update progress less frequently, e.g., every 1% or 5% of completion.
-
-**Current Code:**
-```python
-print(f"\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}", end="", flush=True)
-```
-
-**Suggested Code:**
-```python
-if index % max(1, len(df) // 100) == 0:  # Update every 1%
-    print(f"\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}", end="", flush=True)
-```
-
-### 3. No error handling for file not found in main function.
-📁 **File:** `main.py:175`
-⚖️ **Severity:** 6/10
-🔍 **Description:** The program may crash if the specified CSV file doesn't exist.
-💡 **Solution:** Add try-except block to handle FileNotFoundError.
-
-**Current Code:**
-```python
-main(input_file)
-```
-
-**Suggested Code:**
-```python
-try:
-    main(input_file)
-except FileNotFoundError:
-    print(f"Error: The file '{input_file}' was not found.")
-except Exception as e:
-    print(f"An error occurred:{e}")
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (1 issues)</strong></summary>
-
-<details>
-<summary><strong>Redundant Code (1 issues)</strong></summary>
-
-### 1. Unnecessary check for empty DataFrame in main function.
-📁 **File:** `main.py:142`
-⚖️ **Severity:** 3/10
-🔍 **Description:** The check is redundant as an empty DataFrame would not affect the subsequent operations.
-💡 **Solution:** Remove the unnecessary check to simplify the code.
-
-**Current Code:**
-```python
-if len(df) == 0:
-    return
-```
-
-**Suggested Code:**
-```python
-# Remove these lines
-```
-
-</details>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (anthropic.claude-3-5-sonnet-20240620-v1:0)
-{"prompt_tokens": 7009, "completion_tokens": 1992, "total_tokens": 9001}
\ No newline at end of file
diff --git a/.gitignore b/.gitignore
index 16d3c87f..ffc3f6ed 100644
--- a/.gitignore
+++ b/.gitignore
@@ -165,4 +165,6 @@ node_modules
 .next
 
 .cloudcode
-tree_sitter_languages/
\ No newline at end of file
+tree_sitter_languages/
+
+.experiments/code_review/.*
\ No newline at end of file
diff --git a/github_app/github_helper/pull_requests.py b/github_app/github_helper/pull_requests.py
index 3c6d7ea9..f2ecd386 100644
--- a/github_app/github_helper/pull_requests.py
+++ b/github_app/github_helper/pull_requests.py
@@ -147,7 +147,7 @@ def clean_keys(topics, min_confidence=None):
         rev = []
         for review in reviews:
             if not review.get("reasoning"):
-                review["reasoning"] = review["comment"]
+                review["reasoning"] = review["description"]
             if confidence_mapping[review["impact"]] >= min_value:
                 rev.append(review)
         new_topics[topic] = rev
@@ -165,7 +165,7 @@ def post_pull_request_comments(url, review, installation_id):
                 "path": review["file_path"],
                 "start_line": review["start_line"],
                 "line": review["end_line"],
-                "body": review["comment"],
+                "body": review["description"],
             }
         ],
     }
diff --git a/kaizen/formatters/code_review_formatter.py b/kaizen/formatters/code_review_formatter.py
index 67ae259b..5257a1e3 100644
--- a/kaizen/formatters/code_review_formatter.py
+++ b/kaizen/formatters/code_review_formatter.py
@@ -6,7 +6,7 @@ def create_pr_review_text(
 ) -> str:
     markdown_output = "# 🔍 Code Review Summary\n\n"
 
-    if sum(1 for review in reviews if review["impact"] == "critical") == 0:
+    if sum(1 for review in reviews if review.get("impact", "") == "critical") == 0:
         markdown_output += "✅ **All Clear:** This commit looks good! 👍\n\n"
     else:
         markdown_output += (
@@ -90,9 +90,15 @@ def create_pr_review_text(
 
 def create_stats_section(reviews: List[Dict]) -> str:
     total_issues = len(reviews)
-    critical_issues = sum(1 for review in reviews if review["impact"] == "critical")
-    important_issues = sum(1 for review in reviews if review["impact"] == "important")
-    minor_issues = sum(1 for review in reviews if review["impact"] in ["moderate"])
+    critical_issues = sum(
+        1 for review in reviews if review.get("impact", "") == "critical"
+    )
+    important_issues = sum(
+        1 for review in reviews if review.get("impact", "") == "important"
+    )
+    minor_issues = sum(
+        1 for review in reviews if review.get("impact", "") in ["moderate"]
+    )
     files_affected = len(set(review["file_path"] for review in reviews))
 
     output = "## 📊 Stats\n"
@@ -106,7 +112,7 @@ def create_stats_section(reviews: List[Dict]) -> str:
 
 def create_issues_section(issues: List[Dict]) -> str:
     output = "<details>\n"
-    output += f"<summary><strong>{issues[0]['topic']} ({len(issues)} issues)</strong></summary>\n\n"
+    output += f"<summary><strong>{issues[0]['category']} ({len(issues)} issues)</strong></summary>\n\n"
     for i, issue in enumerate(issues, 1):
         output += create_issue_section(issue, i)
     output += "</details>\n\n"
@@ -114,10 +120,10 @@ def create_issues_section(issues: List[Dict]) -> str:
 
 
 def create_issue_section(issue: Dict, index: int) -> str:
-    output = f"### {index}. {issue['comment']}\n"
-    output += f"📁 **File:** `{issue['file_name']}:{issue['start_line']}`\n"
-    output += f"⚖️ **Severity:** {issue['severity_level']}/10\n"
-    output += f"🔍 **Description:** {issue.get('reason', '')}\n"
+    output = f"### {index}. {issue['description']}\n"
+    output += f"📁 **File:** `{issue['file_path']}:{issue['start_line']}`\n"
+    output += f"⚖️ **Severity:** {issue['severity']}/10\n"
+    output += f"🔍 **Description:** {issue.get('description', '')}\n"
     output += f"💡 **Solution:** {issue.get('solution', '')}\n\n"
     if issue.get("current_code", None) or issue.get("fixed_code", ""):
         output += "**Current Code:**\n"
diff --git a/kaizen/llms/prompts/code_review_prompts.py b/kaizen/llms/prompts/code_review_prompts.py
index ff153a87..86cf82c1 100644
--- a/kaizen/llms/prompts/code_review_prompts.py
+++ b/kaizen/llms/prompts/code_review_prompts.py
@@ -8,7 +8,7 @@
   "code_quality_percentage": <0_TO_100>,
   "review": [
     {{
-      "topic": "<ISSUE_CATEGORY>",
+      "category": "<ISSUE_CATEGORY>",
       "description": "<CONCISE_ISSUE_DESCRIPTION>",
       "impact": "critical|high|medium|low|trivial",
       "rationale": "<DETAILED_EXPLANATION>",
@@ -18,77 +18,63 @@
       "file_path": "<FULL_FILE_PATH>",
       "start_line": <STARTING_LINE_NUMBER>,
       "end_line": <ENDING_LINE_NUMBER>,
-      "change_type": "addition|deletion|modification",
       "sentiment": "positive|negative|neutral",
       "severity": <1_TO_10>
     }}
   ]
 }}
 
-## Guidelines:
-- Provide specific feedback with file paths and line numbers
-- Use markdown for code snippets. Make sure all code is following the original indentations.
-- Merge duplicate feedback
-- Examine: syntax/logic errors, resource leaks, race conditions, security vulnerabilities
-- If no issues found: {{"review": []}}
-
-## Patch Data Format:
-- First column: Original file line numbers
-- Second column: New file line numbers (spaces for removed lines)
-- Third column: Change type
-  '-1:[-]': Line removed
-  '+1:[+]': Line added
-  '0:[.]': Unchanged line
-- Remaining columns: Code content
-
-Example:
-
-1    -    -1:[-] def old_function(x):
-2         +1:[+] def new_function(x, y):
-3    3     0:[.]     result = x * 2
-4         +1:[+]     result += y
-4    5     0:[.]     return result
-
-This snippet shows a diff (difference) between two versions of a function:
-
-1. The function name changed from 'old_function' to 'new_function'.
-2. A new parameter 'y' was added to the function.
-3. The line 'result = x * 2' remained unchanged.
-4. A new line 'result += y' was added, incorporating the new parameter.
-5. The return statement remained unchanged.
-
-## Review Focus:
-1. Removals (-1:[-]): Identify if removal causes problems in remaining code. Remember any line having -1:[-] is removed line from the new code.
-2. Additions (+1:[+]): Provide detailed feedback and suggest improvements. Remember any line having +1:[+] is added line.
-3. Consider impact of changes on overall code structure and functionality.
-4. Note unchanged lines (0:[.]) for context.
-5. For 'fixed_code' -> always suggest changes for Additions. 
-6. Evaluate code for adherence to best practices and potential optimizations.
-7. Consider performance implications of the changes.
-8. Check for proper error handling and exception management.
-9. Assess the quality and coverage of unit tests, if present.
-10. Evaluate the clarity and completeness of comments and documentation.
-11. Look for potential security vulnerabilities or privacy concerns.
-12. Consider the scalability and maintainability of the changes.
-
-## Field Guidelines:
-- "fixed_code": Corrected code for additions only, between start_line and end_line. make sure start_line you suggest does not has `0:[.]`.
-- "actual_code": Current Code line which you think has error. make sure it always done on `+1:[+]` lines. If not, keep it empty ''.
+## Guidelines
+1. Syntax and Logic Errors:
+   - Incorrect use of language constructs
+   - Off-by-one errors
+   - Null pointer dereferences
+   - Type mismatches
+   - Incorrect boolean logic
+
+2. Performance Issues:
+   - Inefficient algorithms or data structures
+   - Unnecessary computations or memory allocations
+   - Improper use of caching mechanisms
+   - N+1 query problems in database operations
+   - Unoptimized loops or recursions
+
+3. Resource Management:
+   - Memory leaks
+   - Unclosed resources (file handles, database connections)
+   - Improper disposal of heavy objects
+
+4. Concurrency and Threading:
+   - Race conditions
+   - Deadlocks or livelocks
+   - Improper use of synchronization primitives
+   - Thread safety issues
+
+5. Security Vulnerabilities:
+   - SQL injection
+   - Cross-site scripting (XSS)
+   - Insecure cryptographic practices
+   - Improper input validation
+   - Hard-coded credentials or sensitive information
+
+6. Code Structure and Design:
+   - Violations of SOLID principles
+   - Excessive complexity (high cyclomatic complexity)
+   - Code duplication
+   - Poor separation of concerns
+   - Inconsistent naming conventions or coding style
+
+7. Error Handling and Logging:
+   - Inadequate or overly broad exception handling
+   - Lack of proper logging or monitoring
+   - Swallowed exceptions without proper handling
+
+Field Guidelines:
+- "suggested_code": Corrected code for additions only, between start_line and end_line.
 - "start_line" and "end_line": Actual line numbers in the additions.
-- "severity_level": 1 (least severe) to 10 (most critical).
-
-## Suggestion Guidelines:
-- Prioritize issues based on their potential impact on code quality, functionality, and maintainability.
-- Provide concrete examples or code snippets when suggesting improvements.
-- Consider both short-term fixes and long-term architectural improvements.
-- Suggest automated tools or techniques that could help prevent similar issues in the future.
-- When appropriate, reference relevant design patterns or best practices in your suggestions.
+- "severity": 1 (least severe) to 10 (most critical).
 
-## Holistic Review:
-- Consider how the changes fit into the broader context of the project.
-- Assess whether the changes introduce or resolve technical debt.
-- Evaluate if the changes are consistent with the overall architecture and coding style of the project.
-- Consider potential side effects or unintended consequences of the changes.
+Prioritize issues based on their potential impact on code quality, functionality, and maintainability. Provide concrete examples or code snippets when suggesting improvements.
 
 ## PATCH DATA:
 ```{CODE_DIFF}```
@@ -100,67 +86,74 @@
   "code_quality_percentage": <0_TO_100>,
   "review": [
     {{
-      "topic": "<SECTION_TOPIC>",
-      "comment": "<CONCISE_ISSUE_DESCRIPTION>",
-      "confidence": "critical|important|moderate|low|trivial",
-      "reason": "<ISSUE_REASONING>",
-      "solution": "<HIGH_LEVEL_SOLUTION>",
-      "actual_code": "<PEICE_OF_CODE_WHICH_HAS_ISSUES>",
-      "fixed_code": "<CORRECTED_CODE>",
-      "file_name": "<FULL_FILE_PATH>",
-      "start_line": <START_LINE_NUMBER>,
-      "end_line": <END_LINE_NUMBER>,
-      "side": "LEFT|RIGHT",
+      "category": "<ISSUE_CATEGORY>",
+      "description": "<CONCISE_ISSUE_DESCRIPTION>",
+      "impact": "critical|high|medium|low|trivial",
+      "rationale": "<DETAILED_EXPLANATION>",
+      "recommendation": "<SPECIFIC_IMPROVEMENT_SUGGESTION>",
+      "current_code": "<PROBLEMATIC_CODE_SNIPPET>",
+      "suggested_code": "<IMPROVED_CODE_SNIPPET>",
+      "file_path": "<FULL_FILE_PATH>",
+      "start_line": <STARTING_LINE_NUMBER>,
+      "end_line": <ENDING_LINE_NUMBER>,
       "sentiment": "positive|negative|neutral",
-      "severity_level": <1_TO_10>
+      "severity": <1_TO_10>
     }}
   ]
 }}
 
-## Guidelines:
-- Provide specific feedback with file paths and line numbers
-- Use markdown for code snippets. Make sure all code is following the original indentations.
-- Merge duplicate feedback
-- Examine: syntax/logic errors, resource leaks, race conditions, security vulnerabilities
-- If no issues found: {{"review": []}}
-
-## Patch Data Format:
-- First column: Original file line numbers
-- Second column: New file line numbers (spaces for removed lines)
-- Third column: Change type
-  '-1:[-]': Line removed
-  '+1:[+]': Line added
-  '0:[.]': Unchanged line
-- Remaining columns: Code content
 
-Example:
-
-1    -    -1:[-] def old_function(x):
-2         +1:[+] def new_function(x, y):
-3    3     0:[.]     result = x * 2
-4         +1:[+]     result += y
-4    5     0:[.]     return result
-
-This snippet shows a diff (difference) between two versions of a function:
-
-1. The function name changed from 'old_function' to 'new_function'.
-2. A new parameter 'y' was added to the function.
-3. The line 'result = x * 2' remained unchanged.
-4. A new line 'result += y' was added, incorporating the new parameter.
-5. The return statement remained unchanged.
-
-## Review Focus:
-1. Removals (-1:[-]): Identify if removal causes problems in remaining code. Remember any line having -1:[-] is removed line from the new code.
-2. Additions (+1:[+]): Provide detailed feedback and suggest improvements. Remember any line having +1:[+] is added line.
-3. Consider impact of changes on overall code structure and functionality.
-4. Note unchanged lines (0:[.]) for context.
-5. For 'fixed_code' -> always suggest changes for Additions. 
-
-## Field Guidelines:
-- "fixed_code": Corrected code for additions only, between start_line and end_line. make sure start_line you suggest does not has `0:[.]`.
-- "actual_code": Current Code line which you think has error. make sure it always done on `+1:[+]` lines. If not, keep it empty ''.
+## Guidelines
+1. Syntax and Logic Errors:
+   - Incorrect use of language constructs
+   - Off-by-one errors
+   - Null pointer dereferences
+   - Type mismatches
+   - Incorrect boolean logic
+
+2. Performance Issues:
+   - Inefficient algorithms or data structures
+   - Unnecessary computations or memory allocations
+   - Improper use of caching mechanisms
+   - N+1 query problems in database operations
+   - Unoptimized loops or recursions
+
+3. Resource Management:
+   - Memory leaks
+   - Unclosed resources (file handles, database connections)
+   - Improper disposal of heavy objects
+
+4. Concurrency and Threading:
+   - Race conditions
+   - Deadlocks or livelocks
+   - Improper use of synchronization primitives
+   - Thread safety issues
+
+5. Security Vulnerabilities:
+   - SQL injection
+   - Cross-site scripting (XSS)
+   - Insecure cryptographic practices
+   - Improper input validation
+   - Hard-coded credentials or sensitive information
+
+6. Code Structure and Design:
+   - Violations of SOLID principles
+   - Excessive complexity (high cyclomatic complexity)
+   - Code duplication
+   - Poor separation of concerns
+   - Inconsistent naming conventions or coding style
+
+7. Error Handling and Logging:
+   - Inadequate or overly broad exception handling
+   - Lack of proper logging or monitoring
+   - Swallowed exceptions without proper handling
+
+Field Guidelines:
+- "suggested_code": Corrected code for additions only, between start_line and end_line.
 - "start_line" and "end_line": Actual line numbers in the additions.
-- "severity_level": 1 (least severe) to 10 (most critical).
+- "severity": 1 (least severe) to 10 (most critical).
+
+Prioritize issues based on their potential impact on code quality, functionality, and maintainability. Provide concrete examples or code snippets when suggesting improvements.
 
 ## File PATCH Data:
 ```{FILE_PATCH}```
diff --git a/kaizen/reviewer/code_review.py b/kaizen/reviewer/code_review.py
index 37902bac..e4408607 100644
--- a/kaizen/reviewer/code_review.py
+++ b/kaizen/reviewer/code_review.py
@@ -160,7 +160,7 @@ def review_pull_request(
 
         reviews.extend(self.check_sensitive_files(pull_request_files))
 
-        topics = self._merge_topics(reviews)
+        categories = self._merge_categories(reviews)
         prompt_cost, completion_cost = self.provider.get_usage_cost(
             total_usage=self.total_usage
         )
@@ -168,7 +168,7 @@ def review_pull_request(
         return ReviewOutput(
             usage=self.total_usage,
             model_name=self.provider.model,
-            topics=topics,
+            topics=categories,
             issues=reviews,
             code_quality=code_quality,
             cost={"prompt_cost": prompt_cost, "completion_cost": completion_cost},
@@ -304,11 +304,11 @@ def _reevaluate_response(self, prompt: str, resp: str, user: Optional[str]) -> s
         return resp
 
     @staticmethod
-    def _merge_topics(reviews: List[Dict]) -> Dict[str, List[Dict]]:
-        topics = {}
+    def _merge_categories(reviews: List[Dict]) -> Dict[str, List[Dict]]:
+        categories = {}
         for review in reviews:
-            topics.setdefault(review["topic"], []).append(review)
-        return topics
+            categories.setdefault(review["category"], []).append(review)
+        return categories
 
     def check_sensitive_files(self, pull_request_files: list):
         reviews = []
@@ -324,18 +324,18 @@ def check_sensitive_files(self, pull_request_files: list):
                             line = patch.split(" ")[2].split(",")[0][1:]
                         reviews.append(
                             {
-                                "topic": category,
-                                "comment": "Changes made to sensitive file",
-                                "confidence": "critical",
-                                "reason": f"Changes were made to {file_name}, which needs review",
-                                "solution": "NA",
+                                "category": category,
+                                "description": "Changes made to sensitive file",
+                                "impact": "critical",
+                                "recommendation": f"Changes were made to {file_name}, which needs review",
+                                "current_code": "NA",
                                 "fixed_code": "",
                                 "start_line": line,
                                 "end_line": line,
                                 "side": "RIGHT",
-                                "file_name": file_name,
+                                "file_path": file_name,
                                 "sentiment": "negative",
-                                "severity_level": 10,
+                                "severity": 10,
                             }
                         )
         return reviews
diff --git a/poetry.lock b/poetry.lock
index b11ee42b..bbc8bd9e 100644
--- a/poetry.lock
+++ b/poetry.lock
@@ -3398,6 +3398,93 @@ botocore = ">=1.33.2,<2.0a.0"
 [package.extras]
 crt = ["botocore[crt] (>=1.33.2,<2.0a.0)"]
 
+[[package]]
+name = "scikit-learn"
+version = "1.5.1"
+description = "A set of python modules for machine learning and data mining"
+optional = false
+python-versions = ">=3.9"
+files = [
+    {file = "scikit_learn-1.5.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:781586c414f8cc58e71da4f3d7af311e0505a683e112f2f62919e3019abd3745"},
+    {file = "scikit_learn-1.5.1-cp310-cp310-macosx_12_0_arm64.whl", hash = "sha256:f5b213bc29cc30a89a3130393b0e39c847a15d769d6e59539cd86b75d276b1a7"},
+    {file = "scikit_learn-1.5.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1ff4ba34c2abff5ec59c803ed1d97d61b036f659a17f55be102679e88f926fac"},
+    {file = "scikit_learn-1.5.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:161808750c267b77b4a9603cf9c93579c7a74ba8486b1336034c2f1579546d21"},
+    {file = "scikit_learn-1.5.1-cp310-cp310-win_amd64.whl", hash = "sha256:10e49170691514a94bb2e03787aa921b82dbc507a4ea1f20fd95557862c98dc1"},
+    {file = "scikit_learn-1.5.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:154297ee43c0b83af12464adeab378dee2d0a700ccd03979e2b821e7dd7cc1c2"},
+    {file = "scikit_learn-1.5.1-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:b5e865e9bd59396220de49cb4a57b17016256637c61b4c5cc81aaf16bc123bbe"},
+    {file = "scikit_learn-1.5.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:909144d50f367a513cee6090873ae582dba019cb3fca063b38054fa42704c3a4"},
+    {file = "scikit_learn-1.5.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:689b6f74b2c880276e365fe84fe4f1befd6a774f016339c65655eaff12e10cbf"},
+    {file = "scikit_learn-1.5.1-cp311-cp311-win_amd64.whl", hash = "sha256:9a07f90846313a7639af6a019d849ff72baadfa4c74c778821ae0fad07b7275b"},
+    {file = "scikit_learn-1.5.1-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:5944ce1faada31c55fb2ba20a5346b88e36811aab504ccafb9f0339e9f780395"},
+    {file = "scikit_learn-1.5.1-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:0828673c5b520e879f2af6a9e99eee0eefea69a2188be1ca68a6121b809055c1"},
+    {file = "scikit_learn-1.5.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:508907e5f81390e16d754e8815f7497e52139162fd69c4fdbd2dfa5d6cc88915"},
+    {file = "scikit_learn-1.5.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:97625f217c5c0c5d0505fa2af28ae424bd37949bb2f16ace3ff5f2f81fb4498b"},
+    {file = "scikit_learn-1.5.1-cp312-cp312-win_amd64.whl", hash = "sha256:da3f404e9e284d2b0a157e1b56b6566a34eb2798205cba35a211df3296ab7a74"},
+    {file = "scikit_learn-1.5.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:88e0672c7ac21eb149d409c74cc29f1d611d5158175846e7a9c2427bd12b3956"},
+    {file = "scikit_learn-1.5.1-cp39-cp39-macosx_12_0_arm64.whl", hash = "sha256:7b073a27797a283187a4ef4ee149959defc350b46cbf63a84d8514fe16b69855"},
+    {file = "scikit_learn-1.5.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b59e3e62d2be870e5c74af4e793293753565c7383ae82943b83383fdcf5cc5c1"},
+    {file = "scikit_learn-1.5.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1bd8d3a19d4bd6dc5a7d4f358c8c3a60934dc058f363c34c0ac1e9e12a31421d"},
+    {file = "scikit_learn-1.5.1-cp39-cp39-win_amd64.whl", hash = "sha256:5f57428de0c900a98389c4a433d4a3cf89de979b3aa24d1c1d251802aa15e44d"},
+    {file = "scikit_learn-1.5.1.tar.gz", hash = "sha256:0ea5d40c0e3951df445721927448755d3fe1d80833b0b7308ebff5d2a45e6414"},
+]
+
+[package.dependencies]
+joblib = ">=1.2.0"
+numpy = ">=1.19.5"
+scipy = ">=1.6.0"
+threadpoolctl = ">=3.1.0"
+
+[package.extras]
+benchmark = ["matplotlib (>=3.3.4)", "memory_profiler (>=0.57.0)", "pandas (>=1.1.5)"]
+build = ["cython (>=3.0.10)", "meson-python (>=0.16.0)", "numpy (>=1.19.5)", "scipy (>=1.6.0)"]
+docs = ["Pillow (>=7.1.2)", "matplotlib (>=3.3.4)", "memory_profiler (>=0.57.0)", "numpydoc (>=1.2.0)", "pandas (>=1.1.5)", "plotly (>=5.14.0)", "polars (>=0.20.23)", "pooch (>=1.6.0)", "pydata-sphinx-theme (>=0.15.3)", "scikit-image (>=0.17.2)", "seaborn (>=0.9.0)", "sphinx (>=7.3.7)", "sphinx-copybutton (>=0.5.2)", "sphinx-design (>=0.5.0)", "sphinx-gallery (>=0.16.0)", "sphinx-prompt (>=1.4.0)", "sphinx-remove-toctrees (>=1.0.0.post1)", "sphinxcontrib-sass (>=0.3.4)", "sphinxext-opengraph (>=0.9.1)"]
+examples = ["matplotlib (>=3.3.4)", "pandas (>=1.1.5)", "plotly (>=5.14.0)", "pooch (>=1.6.0)", "scikit-image (>=0.17.2)", "seaborn (>=0.9.0)"]
+install = ["joblib (>=1.2.0)", "numpy (>=1.19.5)", "scipy (>=1.6.0)", "threadpoolctl (>=3.1.0)"]
+maintenance = ["conda-lock (==2.5.6)"]
+tests = ["black (>=24.3.0)", "matplotlib (>=3.3.4)", "mypy (>=1.9)", "numpydoc (>=1.2.0)", "pandas (>=1.1.5)", "polars (>=0.20.23)", "pooch (>=1.6.0)", "pyamg (>=4.0.0)", "pyarrow (>=12.0.0)", "pytest (>=7.1.2)", "pytest-cov (>=2.9.0)", "ruff (>=0.2.1)", "scikit-image (>=0.17.2)"]
+
+[[package]]
+name = "scipy"
+version = "1.13.1"
+description = "Fundamental algorithms for scientific computing in Python"
+optional = false
+python-versions = ">=3.9"
+files = [
+    {file = "scipy-1.13.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:20335853b85e9a49ff7572ab453794298bcf0354d8068c5f6775a0eabf350aca"},
+    {file = "scipy-1.13.1-cp310-cp310-macosx_12_0_arm64.whl", hash = "sha256:d605e9c23906d1994f55ace80e0125c587f96c020037ea6aa98d01b4bd2e222f"},
+    {file = "scipy-1.13.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:cfa31f1def5c819b19ecc3a8b52d28ffdcc7ed52bb20c9a7589669dd3c250989"},
+    {file = "scipy-1.13.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f26264b282b9da0952a024ae34710c2aff7d27480ee91a2e82b7b7073c24722f"},
+    {file = "scipy-1.13.1-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:eccfa1906eacc02de42d70ef4aecea45415f5be17e72b61bafcfd329bdc52e94"},
+    {file = "scipy-1.13.1-cp310-cp310-win_amd64.whl", hash = "sha256:2831f0dc9c5ea9edd6e51e6e769b655f08ec6db6e2e10f86ef39bd32eb11da54"},
+    {file = "scipy-1.13.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:27e52b09c0d3a1d5b63e1105f24177e544a222b43611aaf5bc44d4a0979e32f9"},
+    {file = "scipy-1.13.1-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:54f430b00f0133e2224c3ba42b805bfd0086fe488835effa33fa291561932326"},
+    {file = "scipy-1.13.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e89369d27f9e7b0884ae559a3a956e77c02114cc60a6058b4e5011572eea9299"},
+    {file = "scipy-1.13.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a78b4b3345f1b6f68a763c6e25c0c9a23a9fd0f39f5f3d200efe8feda560a5fa"},
+    {file = "scipy-1.13.1-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:45484bee6d65633752c490404513b9ef02475b4284c4cfab0ef946def50b3f59"},
+    {file = "scipy-1.13.1-cp311-cp311-win_amd64.whl", hash = "sha256:5713f62f781eebd8d597eb3f88b8bf9274e79eeabf63afb4a737abc6c84ad37b"},
+    {file = "scipy-1.13.1-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:5d72782f39716b2b3509cd7c33cdc08c96f2f4d2b06d51e52fb45a19ca0c86a1"},
+    {file = "scipy-1.13.1-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:017367484ce5498445aade74b1d5ab377acdc65e27095155e448c88497755a5d"},
+    {file = "scipy-1.13.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:949ae67db5fa78a86e8fa644b9a6b07252f449dcf74247108c50e1d20d2b4627"},
+    {file = "scipy-1.13.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:de3ade0e53bc1f21358aa74ff4830235d716211d7d077e340c7349bc3542e884"},
+    {file = "scipy-1.13.1-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:2ac65fb503dad64218c228e2dc2d0a0193f7904747db43014645ae139c8fad16"},
+    {file = "scipy-1.13.1-cp312-cp312-win_amd64.whl", hash = "sha256:cdd7dacfb95fea358916410ec61bbc20440f7860333aee6d882bb8046264e949"},
+    {file = "scipy-1.13.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:436bbb42a94a8aeef855d755ce5a465479c721e9d684de76bf61a62e7c2b81d5"},
+    {file = "scipy-1.13.1-cp39-cp39-macosx_12_0_arm64.whl", hash = "sha256:8335549ebbca860c52bf3d02f80784e91a004b71b059e3eea9678ba994796a24"},
+    {file = "scipy-1.13.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d533654b7d221a6a97304ab63c41c96473ff04459e404b83275b60aa8f4b7004"},
+    {file = "scipy-1.13.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:637e98dcf185ba7f8e663e122ebf908c4702420477ae52a04f9908707456ba4d"},
+    {file = "scipy-1.13.1-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:a014c2b3697bde71724244f63de2476925596c24285c7a637364761f8710891c"},
+    {file = "scipy-1.13.1-cp39-cp39-win_amd64.whl", hash = "sha256:392e4ec766654852c25ebad4f64e4e584cf19820b980bc04960bca0b0cd6eaa2"},
+    {file = "scipy-1.13.1.tar.gz", hash = "sha256:095a87a0312b08dfd6a6155cbbd310a8c51800fc931b8c0b84003014b874ed3c"},
+]
+
+[package.dependencies]
+numpy = ">=1.22.4,<2.3"
+
+[package.extras]
+dev = ["cython-lint (>=0.12.2)", "doit (>=0.36.0)", "mypy", "pycodestyle", "pydevtool", "rich-click", "ruff", "types-psutil", "typing_extensions"]
+doc = ["jupyterlite-pyodide-kernel", "jupyterlite-sphinx (>=0.12.0)", "jupytext", "matplotlib (>=3.5)", "myst-nb", "numpydoc", "pooch", "pydata-sphinx-theme (>=0.15.2)", "sphinx (>=5.0.0)", "sphinx-design (>=0.4.0)"]
+test = ["array-api-strict", "asv", "gmpy2", "hypothesis (>=6.30)", "mpmath", "pooch", "pytest", "pytest-cov", "pytest-timeout", "pytest-xdist", "scikit-umfpack", "threadpoolctl"]
+
 [[package]]
 name = "setuptools"
 version = "73.0.1"
@@ -3589,6 +3676,17 @@ files = [
     {file = "text_unidecode-1.3-py2.py3-none-any.whl", hash = "sha256:1311f10e8b895935241623731c2ba64f4c455287888b18189350b67134a822e8"},
 ]
 
+[[package]]
+name = "threadpoolctl"
+version = "3.5.0"
+description = "threadpoolctl"
+optional = false
+python-versions = ">=3.8"
+files = [
+    {file = "threadpoolctl-3.5.0-py3-none-any.whl", hash = "sha256:56c1e26c150397e58c4926da8eeee87533b1e32bef131bd4bf6a2f45f3185467"},
+    {file = "threadpoolctl-3.5.0.tar.gz", hash = "sha256:082433502dd922bf738de0d8bcc4fdcbf0979ff44c42bd40f5af8a282f6fa107"},
+]
+
 [[package]]
 name = "tiktoken"
 version = "0.7.0"
@@ -4225,4 +4323,4 @@ typescript = ["tree-sitter-typescript"]
 [metadata]
 lock-version = "2.0"
 python-versions = "^3.9.0"
-content-hash = "4467d43095e2647060b964f8a8f13b4245d62ebe0f48b947220b2fc0dd115281"
+content-hash = "b4274a85a73c01164dc0249b6fae75f4ff6bb62765e7a16ec68766f75fbdfcdb"
diff --git a/pyproject.toml b/pyproject.toml
index a1b996a7..71f12053 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "kaizen-cloudcode"
-version = "0.4.5"
+version = "0.4.6"
 description = "An intelligent coding companion that accelerates your development workflow by providing efficient assistance, enabling you to craft high-quality code more rapidly."
 authors = ["Saurav Panda <saurav.panda@cloudcode.ai>"]
 license = "Apache2.0"
@@ -52,6 +52,9 @@ typescript = ["tree-sitter-typescript"]
 rust = ["tree-sitter-rust"]
 all = ["tree-sitter-python", "tree-sitter-javascript", "tree-sitter-typescript", "tree-sitter-rust"]
 
+[tool.poetry.group.dev.dependencies]
+scikit-learn = "^1.5.1"
+
 [build-system]
 requires = ["poetry-core"]
 build-backend = "poetry.core.masonry.api"

From d0cf2960e6ef8d344f7ccd8801fbca478f42a2c1 Mon Sep 17 00:00:00 2001
From: Saurav Panda <sgp65@cornell.edu>
Date: Tue, 27 Aug 2024 19:20:28 -0700
Subject: [PATCH 18/19] feat: updated the code review prompt

---
 .../no_eval/pr_222/comments.json              | 102 -----
 .../no_eval/pr_222/issues.json                | 372 ----------------
 .../no_eval/pr_222/review.md                  | 166 -------
 .../no_eval/pr_232/comments.json              |   1 -
 .../no_eval/pr_232/issues.json                | 227 ----------
 .../no_eval/pr_232/review.md                  |  41 --
 .../no_eval/pr_252/comments.json              |   1 -
 .../no_eval/pr_252/issues.json                |  77 ----
 .../no_eval/pr_252/review.md                  |  41 --
 .../no_eval/pr_335/comments.json              |   1 -
 .../no_eval/pr_335/issues.json                |  62 ---
 .../no_eval/pr_335/review.md                  |  41 --
 .../no_eval/pr_400/comments.json              |  16 -
 .../no_eval/pr_400/issues.json                | 421 ------------------
 .../no_eval/pr_400/review.md                  |  64 ---
 .../no_eval/pr_440/comments.json              |   1 -
 .../no_eval/pr_440/issues.json                |  47 --
 .../no_eval/pr_440/review.md                  |  41 --
 .../no_eval/pr_476/comments.json              |  16 -
 .../no_eval/pr_476/issues.json                | 121 -----
 .../no_eval/pr_476/review.md                  |  64 ---
 .../no_eval/pr_5/comments.json                |  17 -
 .../no_eval/pr_5/issues.json                  | 107 -----
 .../no_eval/pr_5/review.md                    |  64 ---
 .../no_eval/pr_222/comments.json              |  87 ----
 .../no_eval/pr_222/issues.json                | 342 --------------
 .../no_eval/pr_222/review.md                  | 136 ------
 .../no_eval/pr_232/comments.json              |   1 -
 .../no_eval/pr_232/issues.json                | 197 --------
 .../no_eval/pr_232/review.md                  |  33 --
 .../no_eval/pr_252/comments.json              |   1 -
 .../no_eval/pr_252/issues.json                |  62 ---
 .../no_eval/pr_252/review.md                  |  33 --
 .../no_eval/pr_335/comments.json              |   1 -
 .../no_eval/pr_335/issues.json                |  62 ---
 .../no_eval/pr_335/review.md                  |  33 --
 .../no_eval/pr_400/comments.json              |  16 -
 .../no_eval/pr_400/issues.json                | 346 --------------
 .../no_eval/pr_400/review.md                  |  56 ---
 .../no_eval/pr_440/comments.json              |   1 -
 .../no_eval/pr_440/issues.json                |  47 --
 .../no_eval/pr_440/review.md                  |  33 --
 .../no_eval/pr_476/comments.json              |  31 --
 .../no_eval/pr_476/issues.json                |  91 ----
 .../no_eval/pr_476/review.md                  |  72 ---
 .../no_eval/pr_5/comments.json                |  17 -
 .../no_eval/pr_5/issues.json                  | 107 -----
 .../no_eval/pr_5/review.md                    |  64 ---
 .../no_eval/pr_222/comments.json              | 100 -----
 .../no_eval/pr_222/issues.json                | 282 ------------
 .../no_eval/pr_222/review.md                  | 152 -------
 .../no_eval/pr_232/comments.json              |   1 -
 .../no_eval/pr_232/issues.json                | 198 --------
 .../no_eval/pr_232/review.md                  |  33 --
 .../no_eval/pr_252/comments.json              |   1 -
 .../no_eval/pr_252/issues.json                |  58 ---
 .../no_eval/pr_252/review.md                  |  33 --
 .../no_eval/pr_335/comments.json              |   1 -
 .../no_eval/pr_335/issues.json                |  72 ---
 .../no_eval/pr_335/review.md                  |  33 --
 .../no_eval/pr_400/comments.json              |  16 -
 .../no_eval/pr_400/issues.json                | 380 ----------------
 .../no_eval/pr_400/review.md                  |  56 ---
 .../no_eval/pr_440/comments.json              |   1 -
 .../no_eval/pr_440/issues.json                |  44 --
 .../no_eval/pr_440/review.md                  |  33 --
 .../no_eval/pr_476/comments.json              |  30 --
 .../no_eval/pr_476/issues.json                |  86 ----
 .../no_eval/pr_476/review.md                  |  72 ---
 .../no_eval/pr_5/comments.json                |   1 -
 .../no_eval/pr_5/issues.json                  |  86 ----
 .../no_eval/pr_5/review.md                    |  41 --
 .../no_eval/pr_222/comments.json              |  86 ----
 .../no_eval/pr_222/issues.json                | 282 ------------
 .../no_eval/pr_222/review.md                  | 136 ------
 .../no_eval/pr_232/comments.json              |   1 -
 .../no_eval/pr_232/issues.json                | 156 -------
 .../no_eval/pr_232/review.md                  |  33 --
 .../no_eval/pr_252/comments.json              |   1 -
 .../no_eval/pr_252/issues.json                |  58 ---
 .../no_eval/pr_252/review.md                  |  33 --
 .../no_eval/pr_335/comments.json              |   1 -
 .../no_eval/pr_335/issues.json                |  58 ---
 .../no_eval/pr_335/review.md                  |  33 --
 .../no_eval/pr_400/comments.json              |  30 --
 .../no_eval/pr_400/issues.json                | 338 --------------
 .../no_eval/pr_400/review.md                  |  81 ----
 .../no_eval/pr_440/comments.json              |   1 -
 .../no_eval/pr_440/issues.json                |  44 --
 .../no_eval/pr_440/review.md                  |  33 --
 .../no_eval/pr_476/comments.json              |  16 -
 .../no_eval/pr_476/issues.json                |  72 ---
 .../no_eval/pr_476/review.md                  |  64 ---
 .../no_eval/pr_5/comments.json                |  16 -
 .../no_eval/pr_5/issues.json                  |  72 ---
 .../no_eval/pr_5/review.md                    |  64 ---
 .../code_review/dataset/pr_222/issues.json    |  47 +-
 .../code_review/dataset/pr_252/issues.json    |  21 +-
 .../code_review/dataset/pr_335/issues.json    |   2 +-
 .../code_review/dataset/pr_400/issues.json    |  30 --
 .../code_review/dataset/pr_476/issues.json    |  12 +-
 .experiments/code_review/evaluate.py          |  13 +-
 kaizen/llms/prompts/code_review_prompts.py    | 185 ++++----
 103 files changed, 98 insertions(+), 7711 deletions(-)
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/review.md
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/comments.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/issues.json
 delete mode 100644 .experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/review.md

diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/comments.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/comments.json
deleted file mode 100644
index 259d5518..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/comments.json
+++ /dev/null
@@ -1,102 +0,0 @@
-[
-  {
-    "category": "Security",
-    "description": "Hardcoded API keys in config.json",
-    "impact": "critical",
-    "rationale": "Hardcoding API keys can lead to security vulnerabilities if the code is shared or exposed. It is a best practice to use environment variables for sensitive information.",
-    "recommendation": "Replace hardcoded API keys with environment variables.",
-    "current_code": "\"api_key\": \"os.environ/AZURE_API_KEY\"",
-    "suggested_code": "\"api_key\": \"${AZURE_API_KEY}\"",
-    "file_path": "config.json",
-    "start_line": 13,
-    "end_line": 23,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 9
-  },
-  {
-    "category": "Security",
-    "description": "Potential SQL injection vulnerability in `store_abstraction_and_embedding` method.",
-    "impact": "critical",
-    "rationale": "Directly inserting user input into SQL queries can lead to SQL injection attacks.",
-    "recommendation": "Use parameterized queries to prevent SQL injection.",
-    "current_code": "embedding_query = text(\n    \"\"\"\n    INSERT INTO function_embeddings (function_id, vector)\n    VALUES (:function_id, :vector)\n    ON CONFLICT (function_id) DO UPDATE SET vector = EXCLUDED.vector\n    \"\"\"\n)",
-    "suggested_code": "embedding_query = text(\n    \"\"\"\n    INSERT INTO function_embeddings (function_id, vector)\n    VALUES (:function_id, :vector)\n    ON CONFLICT (function_id) DO UPDATE SET vector = EXCLUDED.vector\n    \"\"\"\n).bindparams(function_id=function_id, vector=embedding)",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 158,
-    "end_line": 163,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 9
-  },
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Docker",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to Dockerfile, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_path": "Dockerfile",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Docker",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to docker-compose.yml, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "15",
-    "end_line": "15",
-    "side": "RIGHT",
-    "file_path": "docker-compose.yml",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Version Control",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to .gitignore, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "164",
-    "end_line": "164",
-    "side": "RIGHT",
-    "file_path": ".gitignore",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Database",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to db_setup/init.sql, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_path": "db_setup/init.sql",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/issues.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/issues.json
deleted file mode 100644
index 3c296611..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/issues.json
+++ /dev/null
@@ -1,372 +0,0 @@
-[
-  {
-    "category": "Security",
-    "description": "Hardcoded API keys in config.json",
-    "impact": "critical",
-    "rationale": "Hardcoding API keys can lead to security vulnerabilities if the code is shared or exposed. It is a best practice to use environment variables for sensitive information.",
-    "recommendation": "Replace hardcoded API keys with environment variables.",
-    "current_code": "\"api_key\": \"os.environ/AZURE_API_KEY\"",
-    "suggested_code": "\"api_key\": \"${AZURE_API_KEY}\"",
-    "file_path": "config.json",
-    "start_line": 13,
-    "end_line": 23,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 9
-  },
-  {
-    "category": "Performance",
-    "description": "Inefficient Dockerfile layer management",
-    "impact": "high",
-    "rationale": "Combining multiple RUN commands into a single command can reduce the number of layers in the Docker image, leading to smaller image sizes and faster build times.",
-    "recommendation": "Combine the RUN commands into a single command.",
-    "current_code": "RUN apt-get update && apt-get install -y \\\n    git \\\n    build-essential \\\n    && rm -rf /var/lib/apt/lists/*",
-    "suggested_code": "RUN apt-get update && apt-get install -y \\\n    git \\\n    build-essential \\\n    && rm -rf /var/lib/apt/lists/*",
-    "file_path": "Dockerfile",
-    "start_line": 8,
-    "end_line": 11,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 7
-  },
-  {
-    "category": "Error Handling",
-    "description": "Lack of error handling in install_tree_sitter_languages.sh",
-    "impact": "medium",
-    "rationale": "The script does not check if the git clone or tree-sitter generate commands succeed, which can lead to silent failures.",
-    "recommendation": "Add error handling to check the success of git clone and tree-sitter generate commands.",
-    "current_code": "git clone \"https://github.com/tree-sitter/tree-sitter-$lang\" \"$LANGUAGE_DIR/tree-sitter-$lang\"\ntree-sitter generate",
-    "suggested_code": "git clone \"https://github.com/tree-sitter/tree-sitter-$lang\" \"$LANGUAGE_DIR/tree-sitter-$lang\" ||{echo 'git clone failed'; exit 1;}\ntree-sitter generate ||{echo 'tree-sitter generate failed'; exit 1;}",
-    "file_path": "install_tree_sitter_languages.sh",
-    "start_line": 24,
-    "end_line": 34,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Readability",
-    "description": "Commented-out code in provider.py",
-    "impact": "low",
-    "rationale": "Commented-out code can clutter the codebase and make it harder to read and maintain.",
-    "recommendation": "Remove commented-out code if it is no longer needed.",
-    "current_code": "# for model in self.config[\"language_model\"][\"models\"]:\n#     if model[\"model_name\"] == \"embedding\":\n#         break",
-    "suggested_code": "",
-    "file_path": "kaizen/llms/provider.py",
-    "start_line": 239,
-    "end_line": 241,
-    "change_type": "deletion",
-    "sentiment": "neutral",
-    "severity": 3
-  },
-  {
-    "category": "Best Practices",
-    "description": "No newline at end of file",
-    "impact": "trivial",
-    "rationale": "A newline at the end of a file is a POSIX standard and helps with version control diffs.",
-    "recommendation": "Add a newline at the end of the file.",
-    "current_code": "",
-    "suggested_code": "\n",
-    "file_path": "install_tree_sitter_languages.sh",
-    "start_line": 48,
-    "end_line": 48,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 1
-  },
-  {
-    "category": "Error Handling",
-    "description": "Lack of error handling in database query execution.",
-    "impact": "high",
-    "rationale": "If the database query fails, the code does not handle the exception, which could lead to application crashes or unhandled exceptions.",
-    "recommendation": "Add try-except blocks to handle potential exceptions during the database query execution.",
-    "current_code": "cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))",
-    "suggested_code": "try:\n    cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))\nexcept Exception as e:\n    # Handle exception (e.g., log error, raise custom exception)\n    print(f\"Database query failed:{e}\")",
-    "file_path": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 41,
-    "end_line": 41,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 8
-  },
-  {
-    "category": "Performance",
-    "description": "Inefficient normalization of query embedding.",
-    "impact": "medium",
-    "rationale": "Using numpy for normalization is efficient, but it can be optimized further by avoiding redundant operations.",
-    "recommendation": "Normalize the query embedding in-place to avoid creating unnecessary copies.",
-    "current_code": "query_embedding_normalized = query_embedding_np / np.linalg.norm(query_embedding_np)",
-    "suggested_code": "query_embedding_np /= np.linalg.norm(query_embedding_np)",
-    "file_path": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 16,
-    "end_line": 16,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Code Readability",
-    "description": "SQL query string formatting using f-strings.",
-    "impact": "low",
-    "rationale": "Using f-strings for SQL queries can lead to SQL injection vulnerabilities if not handled properly.",
-    "recommendation": "Use parameterized queries to prevent SQL injection and improve readability.",
-    "current_code": "query = f\"\"\"\nSELECT \n    e.node_id,\n    e.text,\n    e.metadata,\n    1 - (e.embedding <=> %s::vector) as similarity\nFROM \n{self.table_name}e\nJOIN \n    function_abstractions fa ON e.node_id = fa.function_id::text\nJOIN \n    files f ON fa.file_id = f.file_id\nWHERE \n    f.repo_id = %s\nORDER BY \n    similarity DESC\nLIMIT \n    %s\n\"\"\"",
-    "suggested_code": "query = \"\"\"\nSELECT \n    e.node_id,\n    e.text,\n    e.metadata,\n    1 - (e.embedding <=> %s::vector) as similarity\nFROM \n    %s e\nJOIN \n    function_abstractions fa ON e.node_id = fa.function_id::text\nJOIN \n    files f ON fa.file_id = f.file_id\nWHERE \n    f.repo_id = %s\nORDER BY \n    similarity DESC\nLIMIT \n    %s\n\"\"\"",
-    "file_path": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 19,
-    "end_line": 36,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 4
-  },
-  {
-    "category": "Type Hinting",
-    "description": "Missing type hint for the `correction` parameter in `add_feedback` method.",
-    "impact": "low",
-    "rationale": "Providing type hints improves code readability and helps with static analysis.",
-    "recommendation": "Add type hint for the `correction` parameter.",
-    "current_code": "def add_feedback(\n    self, code_id: str, abstraction: str, rating: int, correction: str = None\n) -> None:",
-    "suggested_code": "def add_feedback(\n    self, code_id: str, abstraction: str, rating: int, correction: Optional[str] = None\n) -> None:",
-    "file_path": "kaizen/retriever/feedback_system.py",
-    "start_line": 9,
-    "end_line": 10,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 3
-  },
-  {
-    "category": "Documentation",
-    "description": "Lack of docstrings for methods.",
-    "impact": "low",
-    "rationale": "Docstrings provide essential information about the purpose and usage of methods, improving code maintainability.",
-    "recommendation": "Add docstrings to all methods to describe their functionality, parameters, and return values.",
-    "current_code": "def custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:",
-    "suggested_code": "\"\"\"\nExecutes a custom query to retrieve similar embeddings.\n\n:param query_embedding: The embedding to query against.\n:param repo_id: The repository ID to filter results.\n:param similarity_top_k: The number of top similar results to return.\n:return: A list of dictionaries containing the results.\n\"\"\"\ndef custom_query(self, query_embedding: List[float], repo_id: int, similarity_top_k: int) -> List[dict]:",
-    "file_path": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 13,
-    "end_line": 13,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 2
-  },
-  {
-    "category": "Error Handling",
-    "description": "Exception handling in `generate_abstraction` method is too generic.",
-    "impact": "medium",
-    "rationale": "Catching all exceptions with a generic `except Exception as e` can obscure the root cause of errors and make debugging more difficult.",
-    "recommendation": "Handle specific exceptions where possible and log meaningful error messages.",
-    "current_code": "except Exception as e:\n    raise e",
-    "suggested_code": "except SomeSpecificException as e:\n    logger.error(f\"Specific error occurred:{str(e)}\")\n    raise e\nexcept Exception as e:\n    logger.error(f\"Unexpected error:{str(e)}\")\n    raise e",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 218,
-    "end_line": 219,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Resource Management",
-    "description": "Potential resource leak in `store_abstraction_and_embedding` method.",
-    "impact": "high",
-    "rationale": "The database connection is not explicitly closed, which can lead to resource leaks.",
-    "recommendation": "Use a context manager to ensure the connection is properly closed.",
-    "current_code": "with self.engine.begin() as connection:\n    connection.execute(...)",
-    "suggested_code": "with self.engine.connect() as connection:\n    with connection.begin():\n        connection.execute(...)",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 157,
-    "end_line": 171,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 7
-  },
-  {
-    "category": "Performance",
-    "description": "Inefficient string concatenation in `generate_abstraction` method.",
-    "impact": "low",
-    "rationale": "Using f-strings for multi-line strings can be inefficient and harder to read.",
-    "recommendation": "Use triple-quoted strings for multi-line text.",
-    "current_code": "prompt = f\"\"\"Generate a concise yet comprehensive abstract description of the following{language}code block. \nInclude information about:\n1. The purpose or functionality of the code\n2. Input parameters and return values (if applicable)\n3. Any important algorithms or data structures used\n4. Key dependencies or external libraries used\n5. Any notable design patterns or architectural choices\n6. Potential edge cases or error handling\n\nCode:\n```{language}\n{code_block}\n```\n\"\"\"",
-    "suggested_code": "prompt = f\"\"\"\nGenerate a concise yet comprehensive abstract description of the following{language}code block. \nInclude information about:\n1. The purpose or functionality of the code\n2. Input parameters and return values (if applicable)\n3. Any important algorithms or data structures used\n4. Key dependencies or external libraries used\n5. Any notable design patterns or architectural choices\n6. Potential edge cases or error handling\n\nCode:\n```{language}\n{code_block}\n```\n\"\"\"",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 186,
-    "end_line": 198,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 3
-  },
-  {
-    "category": "Security",
-    "description": "Potential SQL injection vulnerability in `store_abstraction_and_embedding` method.",
-    "impact": "critical",
-    "rationale": "Directly inserting user input into SQL queries can lead to SQL injection attacks.",
-    "recommendation": "Use parameterized queries to prevent SQL injection.",
-    "current_code": "embedding_query = text(\n    \"\"\"\n    INSERT INTO function_embeddings (function_id, vector)\n    VALUES (:function_id, :vector)\n    ON CONFLICT (function_id) DO UPDATE SET vector = EXCLUDED.vector\n    \"\"\"\n)",
-    "suggested_code": "embedding_query = text(\n    \"\"\"\n    INSERT INTO function_embeddings (function_id, vector)\n    VALUES (:function_id, :vector)\n    ON CONFLICT (function_id) DO UPDATE SET vector = EXCLUDED.vector\n    \"\"\"\n).bindparams(function_id=function_id, vector=embedding)",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 158,
-    "end_line": 163,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 9
-  },
-  {
-    "category": "Readability",
-    "description": "Logging configuration is hardcoded.",
-    "impact": "low",
-    "rationale": "Hardcoding logging configuration makes it difficult to change log settings without modifying the code.",
-    "recommendation": "Move logging configuration to a configuration file or environment variables.",
-    "current_code": "logging.basicConfig(\n    level=logging.INFO, format=\"%(asctime)s - %(levelname)s - %(message)s\"\n)",
-    "suggested_code": "logging.basicConfig(\n    level=os.getenv('LOG_LEVEL', logging.INFO),\n    format=os.getenv('LOG_FORMAT', \"%(asctime)s - %(levelname)s - %(message)s\")\n)",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 23,
-    "end_line": 25,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 2
-  },
-  {
-    "category": "Error Handling",
-    "description": "General exception caught without specific handling",
-    "impact": "high",
-    "rationale": "Catching a general exception can obscure the root cause of the error and make debugging difficult. It is better to catch specific exceptions.",
-    "recommendation": "Catch specific exceptions where possible and handle them appropriately.",
-    "current_code": "except Exception as e:",
-    "suggested_code": "except ImportError as e:\n    logger.error(f\"ImportError:{str(e)}\")\n    raise\nexcept ValueError as e:\n    logger.error(f\"ValueError:{str(e)}\")\n    raise",
-    "file_path": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 28,
-    "end_line": 30,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 7
-  },
-  {
-    "category": "Logging",
-    "description": "Logging configuration should be placed in a main guard",
-    "impact": "medium",
-    "rationale": "Setting up logging configuration at the module level can lead to unexpected behavior when the module is imported.",
-    "recommendation": "Move logging configuration inside a main guard.",
-    "current_code": "logging.basicConfig(level=logging.INFO)",
-    "suggested_code": "if __name__ == \"__main__\":\n    logging.basicConfig(level=logging.INFO)",
-    "file_path": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 8,
-    "end_line": 8,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Performance",
-    "description": "Inefficient string concatenation in logging",
-    "impact": "low",
-    "rationale": "Using f-strings in logging can lead to unnecessary string concatenation even if the log level is not enabled.",
-    "recommendation": "Use lazy formatting in logging.",
-    "current_code": "logger.error(f\"Failed to load language{language}:{str(e)}\")",
-    "suggested_code": "logger.error(\"Failed to load language %s: %s\", language, str(e))",
-    "file_path": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 29,
-    "end_line": 29,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 3
-  },
-  {
-    "category": "Code Duplication",
-    "description": "Duplicate code for printing chunks",
-    "impact": "medium",
-    "rationale": "Duplicated code can lead to maintenance issues and inconsistencies.",
-    "recommendation": "Refactor the code to avoid duplication.",
-    "current_code": "print_chunks(\"JavaScript\", chunk_code(javascript_code, \"javascript\"))",
-    "suggested_code": "print_chunks(\"JavaScript\", chunk_code(javascript_code, \"javascript\"))",
-    "file_path": "tests/retriever/test_chunker.py",
-    "start_line": 99,
-    "end_line": 100,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 4
-  },
-  {
-    "category": "Code Style",
-    "description": "Unused import statement",
-    "impact": "trivial",
-    "rationale": "Unused imports can clutter the code and affect readability.",
-    "recommendation": "Remove the unused import statement.",
-    "current_code": "import json",
-    "suggested_code": "",
-    "file_path": "tests/retriever/test_chunker.py",
-    "start_line": 2,
-    "end_line": 2,
-    "change_type": "deletion",
-    "sentiment": "neutral",
-    "severity": 1
-  },
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Docker",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to Dockerfile, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_path": "Dockerfile",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Docker",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to docker-compose.yml, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "15",
-    "end_line": "15",
-    "side": "RIGHT",
-    "file_path": "docker-compose.yml",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Version Control",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to .gitignore, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "164",
-    "end_line": "164",
-    "side": "RIGHT",
-    "file_path": ".gitignore",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Database",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to db_setup/init.sql, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_path": "db_setup/init.sql",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/review.md b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/review.md
deleted file mode 100644
index 06ec2b07..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_222/review.md
+++ /dev/null
@@ -1,166 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/222
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 25
-- Critical: 7
-- Important: 0
-- Minor: 0
-- Files Affected: 12
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Security (7 issues)</strong></summary>
-
-### 1. Hardcoded API keys in config.json
-📁 **File:** `config.json:13`
-⚖️ **Severity:** 9/10
-🔍 **Description:** Hardcoded API keys in config.json
-💡 **Solution:** 
-
-**Current Code:**
-```python
-"api_key": "os.environ/AZURE_API_KEY"
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 2. Potential SQL injection vulnerability in `store_abstraction_and_embedding` method.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:158`
-⚖️ **Severity:** 9/10
-🔍 **Description:** Potential SQL injection vulnerability in `store_abstraction_and_embedding` method.
-💡 **Solution:** 
-
-**Current Code:**
-```python
-embedding_query = text(
-    """
-    INSERT INTO function_embeddings (function_id, vector)
-    VALUES (:function_id, :vector)
-    ON CONFLICT (function_id) DO UPDATE SET vector = EXCLUDED.vector
-    """
-)
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 3. Changes made to sensitive file
-📁 **File:** `config.json:4`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 4. Changes made to sensitive file
-📁 **File:** `Dockerfile:4`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 5. Changes made to sensitive file
-📁 **File:** `docker-compose.yml:15`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 6. Changes made to sensitive file
-📁 **File:** `.gitignore:164`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 7. Changes made to sensitive file
-📁 **File:** `db_setup/init.sql:1`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (9 issues)</strong></summary>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-2024-05-13)
-{"prompt_tokens": 22541, "completion_tokens": 4406, "total_tokens": 26947}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/comments.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/comments.json
deleted file mode 100644
index 0637a088..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/comments.json
+++ /dev/null
@@ -1 +0,0 @@
-[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/issues.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/issues.json
deleted file mode 100644
index d2e57187..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/issues.json
+++ /dev/null
@@ -1,227 +0,0 @@
-[
-  {
-    "category": "readability",
-    "description": "Unnecessary comment removed",
-    "impact": "low",
-    "rationale": "The comment explaining the merging of lists was removed. While the code is self-explanatory, comments can help future developers understand the intent.",
-    "recommendation": "Consider adding a concise comment to explain the merging process.",
-    "current_code": "// merge spaces & memories to{item: \"memory\" | \"space\", date: Date, data: Content | StoredSpace}",
-    "suggested_code": "// Merge spaces and memories into a unified list with item type and date",
-    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 98,
-    "end_line": 98,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 2
-  },
-  {
-    "category": "maintainability",
-    "description": "Type alias introduced for props",
-    "impact": "medium",
-    "rationale": "Introducing a type alias for component props improves maintainability by centralizing the type definition.",
-    "recommendation": "Ensure that the type alias is updated if the props change.",
-    "current_code": "}: TMemoriesPage)",
-    "suggested_code": "}: TMemoriesPage)",
-    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 51,
-    "end_line": 51,
-    "change_type": "addition",
-    "sentiment": "positive",
-    "severity": 4
-  },
-  {
-    "category": "error_handling",
-    "description": "Error message improved for clarity",
-    "impact": "low",
-    "rationale": "The error message was slightly modified for better readability.",
-    "recommendation": "No further changes needed.",
-    "current_code": "toast.error(\"Failed to delete space\")",
-    "suggested_code": "toast.error(\"Failed to delete the space\")",
-    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 73,
-    "end_line": 73,
-    "change_type": "addition",
-    "sentiment": "positive",
-    "severity": 1
-  },
-  {
-    "category": "performance",
-    "description": "Removed unused imports",
-    "impact": "low",
-    "rationale": "Removing unused imports can slightly improve performance and reduce bundle size.",
-    "recommendation": "Continue to ensure that imports are only added when necessary.",
-    "current_code": "import{MemoriesIcon, NextIcon, SearchIcon, UrlIcon}from \"@repo/ui/icons\";",
-    "suggested_code": "import{MemoriesIcon, NextIcon, UrlIcon}from \"@repo/ui/icons\";",
-    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 4,
-    "end_line": 4,
-    "change_type": "modification",
-    "sentiment": "positive",
-    "severity": 2
-  },
-  {
-    "category": "usability",
-    "description": "Button styling improved",
-    "impact": "medium",
-    "rationale": "Improving button styling enhances the user interface and user experience.",
-    "recommendation": "Ensure consistent styling across all buttons in the application.",
-    "current_code": "className={`transition px-6 py-2 rounded-xl hover:text-[#369DFD]\" text-[#B3BCC5] bg-secondary hover:bg-secondary hover:text-[#76a3cc]`}",
-    "suggested_code": "className={`transition px-4 py-2 rounded-lg text-[#B3BCC5] bg-secondary hover:bg-secondary hover:text-[#76a3cc]`}",
-    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 214,
-    "end_line": 214,
-    "change_type": "addition",
-    "sentiment": "positive",
-    "severity": 3
-  },
-  {
-    "category": "code_style",
-    "description": "Consistent spacing in JSX",
-    "impact": "trivial",
-    "rationale": "Ensuring consistent spacing in JSX improves code readability.",
-    "recommendation": "Adopt a consistent code style guide and use linters to enforce it.",
-    "current_code": "{title.slice(0, 2).toUpperCase()}{id}",
-    "suggested_code": "{title.slice(0, 2).toUpperCase()}{id}",
-    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 292,
-    "end_line": 292,
-    "change_type": "addition",
-    "sentiment": "positive",
-    "severity": 1
-  },
-  {
-    "category": "performance",
-    "description": "Removed unnecessary useEffect import",
-    "impact": "low",
-    "rationale": "Removing unused imports can slightly improve performance and reduce bundle size.",
-    "recommendation": "Continue to ensure that imports are only added when necessary.",
-    "current_code": "import React,{useEffect, useState}from \"react\";",
-    "suggested_code": "import React,{useState}from \"react\";",
-    "file_path": "apps/web/app/(dash)/home/queryinput.tsx",
-    "start_line": 3,
-    "end_line": 3,
-    "change_type": "modification",
-    "sentiment": "positive",
-    "severity": 2
-  },
-  {
-    "category": "Code Duplication",
-    "description": "Duplicated logic for fetching spaces and handling form submission.",
-    "impact": "high",
-    "rationale": "Duplicated code increases maintenance overhead and the risk of inconsistencies. Any future changes will need to be made in multiple places, increasing the chance of errors.",
-    "recommendation": "Extract the duplicated logic into reusable hooks or utility functions.",
-    "current_code": "useEffect(() =>{(async () =>{let spaces = await getSpaces(); if (!spaces.success || !spaces.data){toast.warning('Unable to get spaces',{richColors: true}); setSpaces([]); return;}setSpaces(spaces.data);})();},[]);",
-    "suggested_code": "const useFetchSpaces = () =>{const[spaces, setSpaces] = useState<StoredSpace[]>([]); useEffect(() =>{(async () =>{let spaces = await getSpaces(); if (!spaces.success || !spaces.data){toast.warning('Unable to get spaces',{richColors: true}); setSpaces([]); return;}setSpaces(spaces.data);})();},[]); return spaces;};",
-    "file_path": "apps/web/app/(dash)/menu.tsx",
-    "start_line": 173,
-    "end_line": 186,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 7
-  },
-  {
-    "category": "Error Handling",
-    "description": "Improper error handling in form submission.",
-    "impact": "medium",
-    "rationale": "Throwing errors without proper handling can lead to unhandled promise rejections and a poor user experience.",
-    "recommendation": "Use try-catch blocks to handle errors gracefully and provide feedback to the user.",
-    "current_code": "const handleSubmit = async (content?: string, spaces?: number[]) =>{setDialogClose(); if (!content || content.length === 0){throw new Error('Content is required');}const cont = await createMemory({content: content, spaces: spaces ?? undefined}); setContent(''); setSelectedSpaces([]); if (cont.success){toast.success('Memory queued',{richColors: true});}else{toast.error(`Memory creation failed: ${cont.error}`); throw new Error(`Memory creation failed: ${cont.error}`); return cont;}};",
-    "suggested_code": "const handleSubmit = async (content?: string, spaces?: number[]) =>{setDialogClose(); if (!content || content.length === 0){toast.error('Content is required'); return;}try{const cont = await createMemory({content: content, spaces: spaces ?? undefined}); setContent(''); setSelectedSpaces([]); if (cont.success){toast.success('Memory queued',{richColors: true});}else{toast.error(`Memory creation failed: ${cont.error}`);}}catch (error){toast.error(`An error occurred: ${error.message}`);}};",
-    "file_path": "apps/web/app/(dash)/menu.tsx",
-    "start_line": 213,
-    "end_line": 233,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Code Readability",
-    "description": "Long and complex inline functions.",
-    "impact": "low",
-    "rationale": "Long inline functions can make the code harder to read and understand.",
-    "recommendation": "Extract complex inline functions into named functions.",
-    "current_code": "action={async (e: FormData) =>{const content = e.get('content')?.toString(); toast.promise(handleSubmit(content, selectedSpaces),{loading: (<span><PlusCircleIcon className='w-4 h-4 inline mr-2 text-white animate-spin' /> Creating memory...</span>), success: (data) => 'Memory queued', error: (error) => error.message, richColors: true,});}}",
-    "suggested_code": "const handleFormSubmit = async (e: FormData) =>{const content = e.get('content')?.toString(); toast.promise(handleSubmit(content, selectedSpaces),{loading: (<span><PlusCircleIcon className='w-4 h-4 inline mr-2 text-white animate-spin' /> Creating memory...</span>), success: (data) => 'Memory queued', error: (error) => error.message, richColors: true,});};",
-    "file_path": "apps/web/app/(dash)/menu.tsx",
-    "start_line": 237,
-    "end_line": 249,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 4
-  },
-  {
-    "category": "Performance",
-    "description": "Inefficient use of useMemo for options.",
-    "impact": "low",
-    "rationale": "Using useMemo for simple transformations can be unnecessary and might not provide a performance benefit.",
-    "recommendation": "Consider removing useMemo if the performance benefit is negligible.",
-    "current_code": "const options = useMemo(() => spaces.map((x) => ({label: x.name, value: x.id.toString()})),[spaces]);",
-    "suggested_code": "const options = spaces.map((x) => ({label: x.name, value: x.id.toString()}));",
-    "file_path": "apps/web/app/(dash)/menu.tsx",
-    "start_line": 208,
-    "end_line": 211,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 3
-  },
-  {
-    "category": "Error Handling",
-    "description": "Missing error handling for onSubmit function.",
-    "impact": "high",
-    "rationale": "The onSubmit function lacks error handling, which could lead to unhandled exceptions if the function fails.",
-    "recommendation": "Wrap the onSubmit call in a try-catch block to handle potential errors gracefully.",
-    "current_code": "onClick={() => onSubmit(inputValue)}",
-    "suggested_code": "onClick={async () =>{try{await onSubmit(inputValue);}catch (error){console.error('Submission failed', error);}}}",
-    "file_path": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 99,
-    "end_line": 99,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 8
-  },
-  {
-    "category": "Performance",
-    "description": "Inefficient filtering of options.",
-    "impact": "medium",
-    "rationale": "The filtering of options inside the render method can be optimized to avoid unnecessary computations on each render.",
-    "recommendation": "Memoize the filteredOptions using useMemo to avoid recalculating on each render.",
-    "current_code": "const filteredOptions = options.filter((option) => !selectedSpaces.includes(parseInt(option.value)));",
-    "suggested_code": "const filteredOptions = useMemo(() => options.filter((option) => !selectedSpaces.includes(parseInt(option.value))),[options, selectedSpaces]);",
-    "file_path": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 55,
-    "end_line": 57,
-    "change_type": "modification",
-    "sentiment": "positive",
-    "severity": 5
-  },
-  {
-    "category": "Code Readability",
-    "description": "Inline styles and class names are hard to read.",
-    "impact": "low",
-    "rationale": "Inline styles and complex class names reduce readability and maintainability of the code.",
-    "recommendation": "Extract class names into a separate variable or use a CSS-in-JS solution for better readability.",
-    "current_code": "className={`group flex bg-[#2F353C] h-min rounded-md ${selectedSpaces.length > 0 && \"p-2\"}transition-all mt-4 mb-4`}",
-    "suggested_code": "const commandClassNames = `group flex bg-[#2F353C] h-min rounded-md ${selectedSpaces.length > 0 && \"p-2\"}transition-all mt-4 mb-4`;\n...\n<Command className={commandClassNames}>",
-    "file_path": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 61,
-    "end_line": 61,
-    "change_type": "modification",
-    "sentiment": "positive",
-    "severity": 3
-  },
-  {
-    "category": "Code Consistency",
-    "description": "Inconsistent use of single and double quotes.",
-    "impact": "trivial",
-    "rationale": "Inconsistent use of quotes can lead to confusion and reduce code readability.",
-    "recommendation": "Standardize the use of quotes throughout the file, preferably using single quotes for consistency.",
-    "current_code": "className={`group flex bg-[#2F353C] h-min rounded-md ${selectedSpaces.length > 0 && \"p-2\"}transition-all mt-4 mb-4`}",
-    "suggested_code": "className={`group flex bg-[#2F353C] h-min rounded-md ${selectedSpaces.length > 0 && 'p-2'}transition-all mt-4 mb-4`}",
-    "file_path": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 61,
-    "end_line": 61,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 2
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/review.md b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/review.md
deleted file mode 100644
index 8cd757ad..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_232/review.md
+++ /dev/null
@@ -1,41 +0,0 @@
-PR URL: https://github.com/supermemoryai/supermemory/pull/232
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 15
-- Critical: 0
-- Important: 0
-- Minor: 0
-- Files Affected: 4
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (9 issues)</strong></summary>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-2024-05-13)
-{"prompt_tokens": 16642, "completion_tokens": 3356, "total_tokens": 19998}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/comments.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/comments.json
deleted file mode 100644
index 0637a088..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/comments.json
+++ /dev/null
@@ -1 +0,0 @@
-[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/issues.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/issues.json
deleted file mode 100644
index 12fe862d..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/issues.json
+++ /dev/null
@@ -1,77 +0,0 @@
-[
-  {
-    "category": "error_handling",
-    "description": "Lack of error handling for JSON parsing",
-    "impact": "high",
-    "rationale": "If the JSON data is malformed or not as expected, the `json.loads` call will raise an exception, which is not currently handled.",
-    "recommendation": "Add a try-except block to handle JSON parsing errors gracefully.",
-    "current_code": "parsed_data = json.loads(json_data)",
-    "suggested_code": "try:\n    parsed_data = json.loads(json_data)\nexcept json.JSONDecodeError as e:\n    logging.error(f\"Failed to parse JSON:{e}\")\n    return None",
-    "file_path": "kaizen/helpers/parser.py",
-    "start_line": 47,
-    "end_line": 47,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 8
-  },
-  {
-    "category": "performance",
-    "description": "Unnecessary multiple calls to `generate_twitter_post` and `generate_linkedin_post`",
-    "impact": "medium",
-    "rationale": "Generating posts multiple times with the same summary and user is redundant and can impact performance.",
-    "recommendation": "Store the generated posts in variables and reuse them.",
-    "current_code": "twitter_post = work_summary_generator.generate_twitter_post(summary, user=\"oss_example\")\nlinkedin_post = work_summary_generator.generate_linkedin_post(summary, user=\"oss_example\")",
-    "suggested_code": "twitter_post = work_summary_generator.generate_twitter_post(summary, user=\"oss_example\")\nlinkedin_post = work_summary_generator.generate_linkedin_post(summary, user=\"oss_example\")",
-    "file_path": "examples/work_summarizer/main.py",
-    "start_line": 59,
-    "end_line": 60,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "readability",
-    "description": "Long lines exceeding 80 characters",
-    "impact": "low",
-    "rationale": "Long lines can affect readability and maintainability of the code.",
-    "recommendation": "Break long lines into multiple lines for better readability.",
-    "current_code": "linkedin_post = work_summary_generator.generate_linkedin_post(summary, user=\"oss_example\")",
-    "suggested_code": "linkedin_post = work_summary_generator.generate_linkedin_post(\n    summary, user=\"oss_example\"\n)",
-    "file_path": "examples/work_summarizer/main.py",
-    "start_line": 60,
-    "end_line": 62,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 3
-  },
-  {
-    "category": "documentation",
-    "description": "Missing docstrings for new functions",
-    "impact": "medium",
-    "rationale": "Docstrings are important for understanding the purpose and usage of functions.",
-    "recommendation": "Add docstrings to the `generate_twitter_post` and `generate_linkedin_post` functions.",
-    "current_code": "",
-    "suggested_code": "def generate_twitter_post(\n    self,\n    summary: Dict,\n    user: Optional[str] = None,\n) -> str:\n    \"\"\"Generate a Twitter post based on the summary.\"\"\"\n    prompt = TWITTER_POST_PROMPT.format(SUMMARY=summary)\n    response, _ = self.provider.chat_completion(prompt, user=user)\n    return parser.extract_markdown_content(response)\n\n\ndef generate_linkedin_post(\n    self,\n    summary: Dict,\n    user: Optional[str] = None,\n) -> str:\n    \"\"\"Generate a LinkedIn post based on the summary.\"\"\"\n    prompt = LINKEDIN_POST_PROMPT.format(SUMMARY=summary)\n    response, _ = self.provider.chat_completion(prompt, user=user)\n    return parser.extract_markdown_content(response)",
-    "file_path": "kaizen/reviewer/work_summarizer.py",
-    "start_line": 58,
-    "end_line": 74,
-    "change_type": "addition",
-    "sentiment": "positive",
-    "severity": 6
-  },
-  {
-    "category": "syntax",
-    "description": "Missing trailing comma in import statement",
-    "impact": "low",
-    "rationale": "Trailing commas in import statements help to avoid syntax errors when adding new imports.",
-    "recommendation": "Add a trailing comma to the import statement.",
-    "current_code": "PR_REVIEW_EVALUATION_PROMPT",
-    "suggested_code": "PR_REVIEW_EVALUATION_PROMPT,",
-    "file_path": "kaizen/reviewer/code_review.py",
-    "start_line": 12,
-    "end_line": 12,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 2
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/review.md b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/review.md
deleted file mode 100644
index 13e4a3e7..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_252/review.md
+++ /dev/null
@@ -1,41 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/252
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 5
-- Critical: 0
-- Important: 0
-- Minor: 0
-- Files Affected: 4
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (2 issues)</strong></summary>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-2024-05-13)
-{"prompt_tokens": 4135, "completion_tokens": 1094, "total_tokens": 5229}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/comments.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/comments.json
deleted file mode 100644
index 0637a088..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/comments.json
+++ /dev/null
@@ -1 +0,0 @@
-[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/issues.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/issues.json
deleted file mode 100644
index 30e16352..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/issues.json
+++ /dev/null
@@ -1,62 +0,0 @@
-[
-  {
-    "category": "Code Readability",
-    "description": "Replaced `reeval_response` parameter removal",
-    "impact": "medium",
-    "rationale": "The removal of the `reeval_response` parameter reduces the flexibility of the method. This parameter might be necessary for certain use cases, and removing it could limit the functionality.",
-    "recommendation": "Reintroduce the `reeval_response` parameter to maintain the method's flexibility and ensure it can handle all anticipated use cases.",
-    "current_code": "def _process_full_diff(self, prompt, user):",
-    "suggested_code": "def _process_full_diff(self, prompt, user, reeval_response):",
-    "file_path": "kaizen/generator/pr_description.py",
-    "start_line": 77,
-    "end_line": 80,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling",
-    "description": "Removed re-evaluation logic",
-    "impact": "high",
-    "rationale": "Removing the re-evaluation logic could lead to less accurate or incomplete responses. The re-evaluation step ensures that the response is thoroughly checked and improved if necessary.",
-    "recommendation": "Reintroduce the re-evaluation logic to ensure that the responses are accurate and complete.",
-    "current_code": "return desc",
-    "suggested_code": "if reeval_response:\n    resp = self._reevaluate_response(prompt, resp, user)\nreturn resp['desc']",
-    "file_path": "kaizen/generator/pr_description.py",
-    "start_line": 83,
-    "end_line": 88,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Code Consistency",
-    "description": "Inconsistent prompt naming",
-    "impact": "low",
-    "rationale": "The naming of the prompts should be consistent to avoid confusion and maintain code readability.",
-    "recommendation": "Ensure that all prompt names follow a consistent naming convention.",
-    "current_code": "PR_DESCRIPTION_SYSTEM_PROMPT",
-    "suggested_code": "CODE_REVIEW_SYSTEM_PROMPT",
-    "file_path": "kaizen/generator/pr_description.py",
-    "start_line": 28,
-    "end_line": 29,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 3
-  },
-  {
-    "category": "Performance",
-    "description": "Removed detailed re-evaluation prompt",
-    "impact": "medium",
-    "rationale": "The detailed re-evaluation prompt ensures that the system can provide more accurate and refined feedback. Removing it could degrade the quality of the feedback.",
-    "recommendation": "Reintroduce the detailed re-evaluation prompt to maintain high-quality feedback.",
-    "current_code": "return desc",
-    "suggested_code": "return resp['desc']",
-    "file_path": "kaizen/generator/pr_description.py",
-    "start_line": 176,
-    "end_line": 201,
-    "change_type": "deletion",
-    "sentiment": "negative",
-    "severity": 6
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/review.md b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/review.md
deleted file mode 100644
index e6170bdc..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_335/review.md
+++ /dev/null
@@ -1,41 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/335
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 4
-- Critical: 0
-- Important: 0
-- Minor: 0
-- Files Affected: 1
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (1 issues)</strong></summary>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-2024-05-13)
-{"prompt_tokens": 7000, "completion_tokens": 724, "total_tokens": 7724}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/comments.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/comments.json
deleted file mode 100644
index 1ceabeef..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/comments.json
+++ /dev/null
@@ -1,16 +0,0 @@
-[
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "11",
-    "end_line": "11",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/issues.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/issues.json
deleted file mode 100644
index cca018d1..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/issues.json
+++ /dev/null
@@ -1,421 +0,0 @@
-[
-  {
-    "category": "Code Readability",
-    "description": "Inconsistent formatting in DESC_COLLAPSIBLE_TEMPLATE.",
-    "impact": "medium",
-    "rationale": "Inconsistent formatting can lead to confusion and maintenance issues. Ensuring consistent formatting improves readability and maintainability.",
-    "recommendation": "Ensure consistent formatting for multi-line strings.",
-    "current_code": "DESC_COLLAPSIBLE_TEMPLATE = \"<details><summary>Original Description</summary>{desc}</details>\"",
-    "suggested_code": "DESC_COLLAPSIBLE_TEMPLATE = \"<details><summary>Original Description</summary>\\n\\n{desc}\\n\\n</details>\"",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
-    "start_line": 5,
-    "end_line": 5,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 4
-  },
-  {
-    "category": "Code Readability",
-    "description": "Redundant comments in the test file.",
-    "impact": "low",
-    "rationale": "Comments should add value and not state the obvious. Redundant comments can clutter the code and reduce readability.",
-    "recommendation": "Remove redundant comments that do not add value.",
-    "current_code": "# Mock logger",
-    "suggested_code": "",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_folder.py",
-    "start_line": 6,
-    "end_line": 6,
-    "change_type": "deletion",
-    "sentiment": "neutral",
-    "severity": 2
-  },
-  {
-    "category": "Error Handling",
-    "description": "Lack of error handling for invalid folder paths.",
-    "impact": "high",
-    "rationale": "Proper error handling ensures that the application can gracefully handle unexpected inputs and scenarios.",
-    "recommendation": "Add error handling for invalid folder paths.",
-    "current_code": "create_folder(folder_path)",
-    "suggested_code": "if not folder_path: raise ValueError('Folder path cannot be empty')\ncreate_folder(folder_path)",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_folder.py",
-    "start_line": 62,
-    "end_line": 62,
-    "change_type": "modification",
-    "sentiment": "positive",
-    "severity": 7
-  },
-  {
-    "category": "Test Coverage",
-    "description": "Test cases for boundary conditions are missing execution time assertions.",
-    "impact": "medium",
-    "rationale": "Ensuring that boundary conditions are tested for performance can help identify potential performance bottlenecks.",
-    "recommendation": "Add assertions for execution time in boundary condition tests.",
-    "current_code": "print(f\"Execution time:{execution_time}seconds\")",
-    "suggested_code": "assert execution_time < 1, f\"Execution time exceeded:{execution_time}seconds\"",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
-    "start_line": 61,
-    "end_line": 61,
-    "change_type": "modification",
-    "sentiment": "positive",
-    "severity": 5
-  },
-  {
-    "category": "Code Maintainability",
-    "description": "Repeated code for mocking os.path.exists and os.makedirs.",
-    "impact": "medium",
-    "rationale": "Repeated code can lead to maintenance challenges and increase the risk of bugs. Using fixtures can help reduce redundancy.",
-    "recommendation": "Use pytest fixtures to mock os.path.exists and os.makedirs.",
-    "current_code": "@mock.patch(\"kaizen.helpers.output.os.makedirs\")\n@mock.patch(\"kaizen.helpers.output.os.path.exists\")",
-    "suggested_code": "@pytest.fixture\ndef mock_os_path_exists():\n    with mock.patch('os.path.exists') as mock_exists:\n        yield mock_exists\n\n@pytest.fixture\ndef mock_os_makedirs():\n    with mock.patch('os.makedirs') as mock_makedirs:\n        yield mock_makedirs",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_folder.py",
-    "start_line": 9,
-    "end_line": 17,
-    "change_type": "modification",
-    "sentiment": "positive",
-    "severity": 6
-  },
-  {
-    "category": "Code Readability",
-    "description": "Inconsistent string formatting style.",
-    "impact": "medium",
-    "rationale": "The code uses both triple quotes and parentheses for multi-line strings, which can be confusing and reduce readability.",
-    "recommendation": "Use a consistent string formatting style throughout the code. Prefer triple quotes for multi-line strings for better readability.",
-    "current_code": "PR_COLLAPSIBLE_TEMPLATE = (\n    \"<details>\\n\"\n    \"<summary>Review</summary>\\n\"\n    \"<p>\\n\"\n    \"<b>Comment:</b>{comment}<br>\\n\"\n    \"<b>Reason:</b>{reason}<br>\\n\"\n    \"<b>Solution:</b>{solution}<br>\\n\"\n    \"<b>Confidence:</b>{confidence}<br>\\n\"\n    \"<b>Start Line:</b>{start_line}<br>\\n\"\n    \"<b>End Line:</b>{end_line}<br>\\n\"\n    \"<b>File Name:</b>{file_name}<br>\\n\"\n    \"<b>Severity:</b>{severity}<br>\\n\"\n    \"</p>\\n\"\n    \"</details>\"\n)",
-    "suggested_code": "PR_COLLAPSIBLE_TEMPLATE = \"\"\"\n<details>\n<summary>Review</summary>\n<p>{comment}</p>\n<p><strong>Reason:</strong>{reason}</p>\n<p><strong>Solution:</strong>{solution}</p>\n<p><strong>Confidence:</strong>{confidence}</p>\n<p><strong>Start Line:</strong>{start_line}</p>\n<p><strong>End Line:</strong>{end_line}</p>\n<p><strong>File Name:</strong>{file_name}</p>\n<p><strong>Severity:</strong>{severity}</p>\n</details>\n\"\"\"",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 4,
-    "end_line": 16,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Test Coverage",
-    "description": "Missing test for edge cases.",
-    "impact": "high",
-    "rationale": "The tests do not cover edge cases such as empty strings or null values for the review fields.",
-    "recommendation": "Add tests to cover edge cases like empty strings or null values for the review fields to ensure robustness.",
-    "current_code": "",
-    "suggested_code": "@pytest.fixture\n def setup_edge_case_reviews():\n     return{\n         \"topic1\":[\n{\n                 \"comment\": \"\",\n                 \"reason\": None,\n                 \"solution\": \"\",\n                 \"confidence\": \"\",\n                 \"start_line\": 0,\n                 \"end_line\": 0,\n                 \"file_name\": \"\",\n                 \"severity_level\": 0,\n}\n         ]\n}\n\n def test_edge_case_reviews(setup_edge_case_reviews):\n     topics = setup_edge_case_reviews\n     expected_output = \"## Code Review\\n\\n\u2757 **Attention Required:** This PR has potential issues. \ud83d\udea8\\n\\n### topic1\\n\\n\" + PR_COLLAPSIBLE_TEMPLATE.format(\n         comment=\"\",\n         reason=\"NA\",\n         solution=\"\",\n         confidence=\"NA\",\n         start_line=0,\n         end_line=0,\n         file_name=\"\",\n         severity=0,\n     ) + \"\\n\"\n     assert create_pr_review_text(topics) == expected_output",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 271,
-    "end_line": 276,
-    "change_type": "addition",
-    "sentiment": "positive",
-    "severity": 7
-  },
-  {
-    "category": "Error Handling",
-    "description": "Lack of error handling for missing fields in review data.",
-    "impact": "high",
-    "rationale": "If any field in the review data is missing, it could cause runtime errors or incorrect output.",
-    "recommendation": "Add error handling to check for missing fields in the review data and provide default values if necessary.",
-    "current_code": "",
-    "suggested_code": "def create_pr_review_text(topics):\n     output = \"## Code Review\\n\\n\"\n     if not topics:\n         return output + \"\u2705 **All Clear:** This PR is ready to merge! \ud83d\udc4d\\n\\n\"\n\n     output += \"\u2757 **Attention Required:** This PR has potential issues. \ud83d\udea8\\n\\n\"\n     for topic, reviews in topics.items():\n         output += f\"###{topic}\\n\\n\"\n         for review in reviews:\n             comment = review.get(\"comment\", \"NA\")\n             reason = review.get(\"reason\", \"NA\")\n             solution = review.get(\"solution\", \"NA\")\n             confidence = review.get(\"confidence\", \"NA\")\n             start_line = review.get(\"start_line\", 0)\n             end_line = review.get(\"end_line\", 0)\n             file_name = review.get(\"file_name\", \"\")\n             severity = review.get(\"severity_level\", 0)\n             output += PR_COLLAPSIBLE_TEMPLATE.format(\n                 comment=comment,\n                 reason=reason,\n                 solution=solution,\n                 confidence=confidence,\n                 start_line=start_line,\n                 end_line=end_line,\n                 file_name=file_name,\n                 severity=severity,\n             ) + \"\\n\"\n     return output",
-    "file_path": "kaizen/helpers/output.py",
-    "start_line": 1,
-    "end_line": 20,
-    "change_type": "modification",
-    "sentiment": "positive",
-    "severity": 8
-  },
-  {
-    "category": "Best Practices",
-    "description": "Use of hardcoded values for folder paths and module names.",
-    "impact": "medium",
-    "rationale": "Hardcoding values can make the code less flexible and harder to maintain. Using constants or configuration files can improve maintainability.",
-    "recommendation": "Define constants for folder paths and module names, or use a configuration file to manage these values.",
-    "current_code": "folder_path = \"test_folder\"",
-    "suggested_code": "folder_path = CONFIG['TEST_FOLDER_PATH']",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 27,
-    "end_line": 27,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling",
-    "description": "Lack of error handling for file operations.",
-    "impact": "high",
-    "rationale": "File operations can fail due to various reasons such as permission issues or missing files. Proper error handling ensures the program can handle such situations gracefully.",
-    "recommendation": "Add try-except blocks around file operations to handle potential errors.",
-    "current_code": "with open(file_path, 'r') as f:\n    return f.read()",
-    "suggested_code": "try:\n    with open(file_path, 'r') as f:\n        return f.read()\nexcept IOError as e:\n    logger.error(f\"Error reading file{file_path}:{e}\")\n    return None",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 17,
-    "end_line": 18,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 7
-  },
-  {
-    "category": "Code Readability",
-    "description": "Long lines of code that reduce readability.",
-    "impact": "low",
-    "rationale": "Long lines can make the code harder to read and understand. Breaking them into multiple lines can improve readability.",
-    "recommendation": "Break long lines into multiple lines for better readability.",
-    "current_code": "mock.patch('kaizen.helpers.output.create_folder') as mock_create_folder, mock.patch('kaizen.helpers.output.general.clean_python_code') as mock_clean_python_code, mock.patch('kaizen.helpers.output.logger') as mock_logger:",
-    "suggested_code": "mock.patch('kaizen.helpers.output.create_folder') as mock_create_folder, \nmock.patch('kaizen.helpers.output.general.clean_python_code') as mock_clean_python_code, \nmock.patch('kaizen.helpers.output.logger') as mock_logger:",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 10,
-    "end_line": 12,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 3
-  },
-  {
-    "category": "Security",
-    "description": "Potential security vulnerability with file names containing special characters.",
-    "impact": "high",
-    "rationale": "Special characters in file names can lead to security vulnerabilities such as path traversal attacks. Sanitizing file names can mitigate this risk.",
-    "recommendation": "Ensure that file names are sanitized to remove special characters.",
-    "current_code": "def sanitize_filename(filename):\n    return \"\".join(c if c.isalnum() or c in (' ', '.', '_') else '_' for c in filename)",
-    "suggested_code": "def sanitize_filename(filename):\n    return \"\".join(c if c.isalnum() or c in (' ', '.', '_') else '_' for c in filename).replace(' ', '_')",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 21,
-    "end_line": 22,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 8
-  },
-  {
-    "category": "Testing",
-    "description": "Missing test cases for edge scenarios.",
-    "impact": "medium",
-    "rationale": "Edge scenarios such as empty input or invalid data should be tested to ensure the robustness of the code.",
-    "recommendation": "Add test cases for edge scenarios such as empty input or invalid data.",
-    "current_code": "",
-    "suggested_code": "def test_empty_input(tmp_path, mock_dependencies):\n    mock_create_folder, mock_clean_python_code, mock_logger = mock_dependencies\n    json_tests =[]\n    create_test_files(json_tests, tmp_path)\n    assert os.path.exists(os.path.join(tmp_path, \"tests.json\"))\n    assert os.path.getsize(os.path.join(tmp_path, \"tests.json\")) == 0",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 92,
-    "end_line": 102,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "error_handling",
-    "description": "Inconsistent error messages in test_get_parent_folder_error_handling",
-    "impact": "medium",
-    "rationale": "The error messages in the mock.patch side_effect and pytest.raises do not match, which can lead to confusion and potential test failures.",
-    "recommendation": "Ensure that the error messages in mock.patch side_effect and pytest.raises match exactly.",
-    "current_code": "with mock.patch('os.getcwd', side_effect=OSError(\"Unable to determine current working directory\")):\n    with pytest.raises(OSError, match=\"Unable to determine current working directory\"):\n        get_parent_folder()",
-    "suggested_code": "with mock.patch('os.getcwd', side_effect=OSError(\"Failed to get current working directory\")):\n    with pytest.raises(OSError, match=\"Failed to get current working directory\"):\n        get_parent_folder()",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py",
-    "start_line": 20,
-    "end_line": 22,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "performance",
-    "description": "Redundant import of nest_asyncio",
-    "impact": "low",
-    "rationale": "The import of nest_asyncio is redundant as it is already being patched in the test cases. This can lead to unnecessary imports and slightly increased memory usage.",
-    "recommendation": "Remove the redundant import of nest_asyncio.",
-    "current_code": "import nest_asyncio",
-    "suggested_code": "",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 5,
-    "end_line": 5,
-    "change_type": "deletion",
-    "sentiment": "neutral",
-    "severity": 2
-  },
-  {
-    "category": "readability",
-    "description": "Improper indentation in test_get_web_html_large_content",
-    "impact": "low",
-    "rationale": "Improper indentation can make the code harder to read and maintain.",
-    "recommendation": "Ensure proper indentation for better readability.",
-    "current_code": "expected_output = \"<html>\\n <body>\\n\" + \"  <p>\\n   Test\\n  </p>\\n\" * 10000 + \" </body>\\n</html>\"",
-    "suggested_code": "expected_output = (\n    \"<html>\\n <body>\\n\" + \"  <p>\\n   Test\\n  </p>\\n\" * 10000 + \" </body>\\n</html>\"\n)",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 95,
-    "end_line": 95,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 3
-  },
-  {
-    "category": "best_practices",
-    "description": "Hardcoded API key in config.json",
-    "impact": "high",
-    "rationale": "Hardcoding sensitive information like API keys can lead to security vulnerabilities.",
-    "recommendation": "Use environment variables to store sensitive information.",
-    "current_code": "\"api_key\": \"os.environ/AZURE_API_KEY\"",
-    "suggested_code": "\"api_key\": os.getenv('AZURE_API_KEY')",
-    "file_path": "config.json",
-    "start_line": 13,
-    "end_line": 13,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "maintainability",
-    "description": "Unnecessary comments in main.py",
-    "impact": "trivial",
-    "rationale": "Comments that do not add value or are redundant can clutter the code and make it harder to maintain.",
-    "recommendation": "Remove unnecessary comments to improve code clarity.",
-    "current_code": "# generator.generate_tests(file_path=\"sample.py\", content=code)  # Replace with the actual file path",
-    "suggested_code": "",
-    "file_path": "examples/unittest/main.py",
-    "start_line": 34,
-    "end_line": 34,
-    "change_type": "deletion",
-    "sentiment": "neutral",
-    "severity": 1
-  },
-  {
-    "category": "Import Management",
-    "description": "Redundant import of `tqdm` removed and re-added.",
-    "impact": "low",
-    "rationale": "The import of `tqdm` was removed and then re-added later, which is unnecessary and can cause confusion.",
-    "recommendation": "Remove the redundant import statement.",
-    "current_code": "from tqdm import tqdm",
-    "suggested_code": "",
-    "file_path": "kaizen/generator/unit_test.py",
-    "start_line": 14,
-    "end_line": 14,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 2
-  },
-  {
-    "category": "Code Duplication",
-    "description": "Duplicated dictionary for supported languages.",
-    "impact": "medium",
-    "rationale": "The `SUPPORTED_LANGUAGES` dictionary was duplicated, which can lead to maintenance issues if the list needs to be updated.",
-    "recommendation": "Remove the duplicated dictionary and ensure it's defined in one place.",
-    "current_code": "SUPPORTED_LANGUAGES ={\"py\": \"PythonParser\", \"js\": \"JavaScriptParser\", \"ts\": \"TypeScriptParser\", \"jsx\": \"ReactParser\", \"tsx\": \"ReactTSParser\", \"rs\": \"RustParser\"}",
-    "suggested_code": "",
-    "file_path": "kaizen/generator/unit_test.py",
-    "start_line": 24,
-    "end_line": 31,
-    "change_type": "addition",
-    "sentiment": "negative",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling",
-    "description": "Lack of error handling in file operations.",
-    "impact": "high",
-    "rationale": "File operations such as reading and writing files should include error handling to manage potential I/O errors.",
-    "recommendation": "Add try-except blocks around file operations to handle potential I/O errors.",
-    "current_code": "with open(file_path, \"r\") as file: return file.read()",
-    "suggested_code": "try: with open(file_path, \"r\") as file: return file.read() except IOError as e: self.logger.error(f\"Failed to read file{file_path}:{e}\") return None",
-    "file_path": "kaizen/generator/unit_test.py",
-    "start_line": 109,
-    "end_line": 110,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 7
-  },
-  {
-    "category": "Logging",
-    "description": "Logging should be consistent and meaningful.",
-    "impact": "medium",
-    "rationale": "Logging messages should provide meaningful information and be consistent throughout the codebase.",
-    "recommendation": "Ensure that all logging messages are consistent and provide useful information.",
-    "current_code": "print(f\"  \u2713 File will be saved as:{test_file_path}\")",
-    "suggested_code": "self.logger.info(f\"  \u2713 File will be saved as:{test_file_path}\")",
-    "file_path": "kaizen/generator/unit_test.py",
-    "start_line": 140,
-    "end_line": 140,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 4
-  },
-  {
-    "category": "Code Structure",
-    "description": "Refactor long methods into smaller, more manageable methods.",
-    "impact": "high",
-    "rationale": "Long methods can be difficult to read and maintain. Refactoring them into smaller methods improves readability and maintainability.",
-    "recommendation": "Refactor long methods into smaller, more manageable methods.",
-    "current_code": "def generate_test_files(self, parsed_data, file_extension, file_path): ...",
-    "suggested_code": "def generate_test_files(self, parsed_data, file_extension, file_path): for item in tqdm(parsed_data, desc=\"Processing Items\", unit=\"item\"): self._process_item(item, file_extension, file_path)",
-    "file_path": "kaizen/generator/unit_test.py",
-    "start_line": 112,
-    "end_line": 117,
-    "change_type": "addition",
-    "sentiment": "positive",
-    "severity": 6
-  },
-  {
-    "category": "Performance",
-    "description": "Repeated calls to `os.makedirs`.",
-    "impact": "medium",
-    "rationale": "Repeatedly calling `os.makedirs` can be inefficient. It's better to ensure the directory exists once.",
-    "recommendation": "Ensure the directory exists once at the beginning of the process.",
-    "current_code": "os.makedirs(self.log_dir, exist_ok=True)",
-    "suggested_code": "self._setup_directories()",
-    "file_path": "kaizen/generator/unit_test.py",
-    "start_line": 58,
-    "end_line": 58,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "error_handling",
-    "description": "Potential issue with setting log levels",
-    "impact": "high",
-    "rationale": "The function `set_all_loggers_to_ERROR` sets all loggers to ERROR level, which might override specific loggers' configurations that are necessary for debugging or monitoring. This could lead to a loss of important log information.",
-    "recommendation": "Add a parameter to the function to allow excluding certain loggers from being set to ERROR level.",
-    "current_code": "def set_all_loggers_to_ERROR():\n    print(\"All Loggers and their levels:\")\n    for name, logger in logging.Logger.manager.loggerDict.items():\n        if isinstance(logger, logging.Logger):\n            print(f\"Logger:{name}, Level:{logging.getLevelName(logger.level)}\")\n            logging.getLogger(name).setLevel(logging.ERROR)\n        else:\n            print(f\"PlaceHolder:{name}\")",
-    "suggested_code": "def set_all_loggers_to_ERROR(exclude_loggers=None):\n    if exclude_loggers is None:\n        exclude_loggers =[]\n    print(\"All Loggers and their levels:\")\n    for name, logger in logging.Logger.manager.loggerDict.items():\n        if isinstance(logger, logging.Logger) and name not in exclude_loggers:\n            print(f\"Logger:{name}, Level:{logging.getLevelName(logger.level)}\")\n            logging.getLogger(name).setLevel(logging.ERROR)\n        else:\n            print(f\"PlaceHolder:{name}\")",
-    "file_path": "kaizen/llms/provider.py",
-    "start_line": 13,
-    "end_line": 21,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 7
-  },
-  {
-    "category": "performance",
-    "description": "Unnecessary repeated calls to `logging.getLogger`",
-    "impact": "medium",
-    "rationale": "Calling `logging.getLogger(name)` multiple times for the same logger is redundant and can be optimized.",
-    "recommendation": "Store the logger instance in a variable and reuse it.",
-    "current_code": "for name, logger in logging.Logger.manager.loggerDict.items():\n    if isinstance(logger, logging.Logger):\n        print(f\"Logger:{name}, Level:{logging.getLevelName(logger.level)}\")\n        logging.getLogger(name).setLevel(logging.ERROR)",
-    "suggested_code": "for name, logger in logging.Logger.manager.loggerDict.items():\n    if isinstance(logger, logging.Logger):\n        print(f\"Logger:{name}, Level:{logging.getLevelName(logger.level)}\")\n        logger.setLevel(logging.ERROR)",
-    "file_path": "kaizen/llms/provider.py",
-    "start_line": 15,
-    "end_line": 18,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "readability",
-    "description": "Inline comments for setting specific loggers to ERROR",
-    "impact": "low",
-    "rationale": "The inline comments provide context but could be more descriptive for maintainability.",
-    "recommendation": "Expand the comments to explain why these specific loggers are being set to ERROR level.",
-    "current_code": "# Set litellm log level to ERROR\nlogging.getLogger(\"LiteLLM\").setLevel(logging.ERROR)\nlogging.getLogger(\"LiteLLM Router\").setLevel(logging.ERROR)\nlogging.getLogger(\"LiteLLM Proxy\").setLevel(logging.ERROR)",
-    "suggested_code": "# Set specific LiteLLM loggers to ERROR level to minimize log noise and focus on critical issues\nlogging.getLogger(\"LiteLLM\").setLevel(logging.ERROR)\nlogging.getLogger(\"LiteLLM Router\").setLevel(logging.ERROR)\nlogging.getLogger(\"LiteLLM Proxy\").setLevel(logging.ERROR)",
-    "file_path": "kaizen/llms/provider.py",
-    "start_line": 25,
-    "end_line": 27,
-    "change_type": "modification",
-    "sentiment": "positive",
-    "severity": 3
-  },
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "11",
-    "end_line": "11",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/review.md b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/review.md
deleted file mode 100644
index 0644c0bf..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_400/review.md
+++ /dev/null
@@ -1,64 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/400
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 28
-- Critical: 1
-- Important: 0
-- Minor: 0
-- Files Affected: 11
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Configuration (1 issues)</strong></summary>
-
-### 1. Changes made to sensitive file
-📁 **File:** `config.json:11`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (7 issues)</strong></summary>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-2024-05-13)
-{"prompt_tokens": 38388, "completion_tokens": 6513, "total_tokens": 44901}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/comments.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/comments.json
deleted file mode 100644
index 0637a088..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/comments.json
+++ /dev/null
@@ -1 +0,0 @@
-[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/issues.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/issues.json
deleted file mode 100644
index 162fbd69..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/issues.json
+++ /dev/null
@@ -1,47 +0,0 @@
-[
-  {
-    "category": "Code Clarity",
-    "description": "Uncommented setup_repository call",
-    "impact": "medium",
-    "rationale": "Uncommenting the setup_repository call without context or explanation can lead to confusion for future maintainers. It's important to provide context or comments explaining why this line is uncommented.",
-    "recommendation": "Add a comment explaining why the setup_repository call is uncommented and when it should be used.",
-    "current_code": "analyzer.setup_repository(\"./github_app/\")",
-    "suggested_code": "# Uncommented for initial setup or updates\nanalyzer.setup_repository(\"./github_app/\")",
-    "file_path": "examples/ragify_codebase/main.py",
-    "start_line": 7,
-    "end_line": 7,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 4
-  },
-  {
-    "category": "Code Commenting",
-    "description": "Incomplete TODO comment",
-    "impact": "low",
-    "rationale": "The TODO comment 'DONT PUSH DUPLICATE' is too vague and does not provide enough information on what needs to be done. This can lead to confusion and potential errors in the future.",
-    "recommendation": "Provide a more detailed comment explaining what needs to be done to avoid pushing duplicates.",
-    "current_code": "# TODO: DONT PUSH DUPLICATE",
-    "suggested_code": "# TODO: Implement logic to check for and avoid pushing duplicate embeddings",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 157,
-    "end_line": 157,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 3
-  },
-  {
-    "category": "Dependency Management",
-    "description": "Updated dependency versions",
-    "impact": "medium",
-    "rationale": "Updating dependency versions without proper testing can introduce new bugs or incompatibilities. It's important to ensure that all tests pass and that the new versions are compatible with the rest of the codebase.",
-    "recommendation": "Run all tests and perform thorough testing to ensure that the updated dependencies do not introduce any issues.",
-    "current_code": "llama-index-core = \"0.10.65\"",
-    "suggested_code": "",
-    "file_path": "pyproject.toml",
-    "start_line": 27,
-    "end_line": 27,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/review.md b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/review.md
deleted file mode 100644
index e9f81733..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_440/review.md
+++ /dev/null
@@ -1,41 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/440
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 3
-- Critical: 0
-- Important: 0
-- Minor: 0
-- Files Affected: 3
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (1 issues)</strong></summary>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-2024-05-13)
-{"prompt_tokens": 1613, "completion_tokens": 581, "total_tokens": 2194}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/comments.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/comments.json
deleted file mode 100644
index 9cbf8d49..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/comments.json
+++ /dev/null
@@ -1,16 +0,0 @@
-[
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/issues.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/issues.json
deleted file mode 100644
index b84036f9..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/issues.json
+++ /dev/null
@@ -1,121 +0,0 @@
-[
-  {
-    "category": "Error Handling",
-    "description": "Broad exception handling",
-    "impact": "high",
-    "rationale": "Using a broad `except Exception` can mask other errors and make debugging difficult. It is better to catch specific exceptions.",
-    "recommendation": "Replace `except Exception` with `except KeyError` to handle the specific error.",
-    "current_code": "except Exception:\n    print(\"Error\")",
-    "suggested_code": "except KeyError:\n    raise ValueError(f\"Invalid confidence level:{min_confidence}\")",
-    "file_path": "github_app/github_helper/pull_requests.py",
-    "start_line": 140,
-    "end_line": 141,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Code Quality",
-    "description": "Unnecessary print statements",
-    "impact": "medium",
-    "rationale": "Print statements can clutter the output and are not suitable for production code. Use logging instead.",
-    "recommendation": "Replace `print` statements with appropriate logging calls.",
-    "current_code": "print(\"issues: \", issues)",
-    "suggested_code": "logger.info(\"issues: %s\", issues)",
-    "file_path": "github_app/github_helper/pull_requests.py",
-    "start_line": 157,
-    "end_line": 157,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Code Quality",
-    "description": "Unnecessary print statements",
-    "impact": "medium",
-    "rationale": "Print statements can clutter the output and are not suitable for production code. Use logging instead.",
-    "recommendation": "Replace `print` statements with appropriate logging calls.",
-    "current_code": "print(\"diff: \", diff_text)\nprint(\"pr_files\", pr_files)",
-    "suggested_code": "logger.info(\"diff: %s\", diff_text)\nlogger.info(\"pr_files: %s\", pr_files)",
-    "file_path": "examples/code_review/main.py",
-    "start_line": 21,
-    "end_line": 22,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Code Quality",
-    "description": "Unnecessary print statements",
-    "impact": "medium",
-    "rationale": "Print statements can clutter the output and are not suitable for production code. Use logging instead.",
-    "recommendation": "Replace `print` statements with appropriate logging calls.",
-    "current_code": "print(f\"Raw Topics: \\n{json.dumps(review_data.issues, indent=2)}\\n\")",
-    "suggested_code": "logger.info(\"Raw Topics: \\n %s\\n\", json.dumps(review_data.issues, indent=2))",
-    "file_path": "examples/code_review/main.py",
-    "start_line": 39,
-    "end_line": 39,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Functionality",
-    "description": "Potential issue with test generation",
-    "impact": "medium",
-    "rationale": "The `generate_tests` function is added but not properly integrated. This could lead to unexpected behavior.",
-    "recommendation": "Ensure `generate_tests` is called correctly and its output is used appropriately.",
-    "current_code": "tests = generate_tests(pr_files)",
-    "suggested_code": "tests = generate_tests(sorted_files)",
-    "file_path": "github_app/github_helper/pull_requests.py",
-    "start_line": 58,
-    "end_line": 58,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Functionality",
-    "description": "Unnecessary return statement",
-    "impact": "low",
-    "rationale": "The return statement at the end of the function is unnecessary and can be removed.",
-    "recommendation": "Remove the return statement.",
-    "current_code": "return True, review_desc.decode(\"utf-8\")",
-    "suggested_code": "",
-    "file_path": "github_app/github_helper/pull_requests.py",
-    "start_line": 67,
-    "end_line": 67,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 3
-  },
-  {
-    "category": "Code Quality",
-    "description": "Unnecessary print statements",
-    "impact": "medium",
-    "rationale": "Print statements can clutter the output and are not suitable for production code. Use logging instead.",
-    "recommendation": "Replace `print` statements with appropriate logging calls.",
-    "current_code": "print(f\"Raw Topics: \\n{json.dumps(review_data.issues, indent=2)}\\n\")",
-    "suggested_code": "logger.info(\"Raw Topics: \\n %s\\n\", json.dumps(review_data.issues, indent=2))",
-    "file_path": "examples/code_review/main.py",
-    "start_line": 39,
-    "end_line": 39,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/review.md b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/review.md
deleted file mode 100644
index 71a2795c..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_476/review.md
+++ /dev/null
@@ -1,64 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/476
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 8
-- Critical: 1
-- Important: 0
-- Minor: 0
-- Files Affected: 3
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Configuration (1 issues)</strong></summary>
-
-### 1. Changes made to sensitive file
-📁 **File:** `config.json:1`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (1 issues)</strong></summary>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-2024-05-13)
-{"prompt_tokens": 4256, "completion_tokens": 1196, "total_tokens": 5452}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/comments.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/comments.json
deleted file mode 100644
index d23463e6..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/comments.json
+++ /dev/null
@@ -1,17 +0,0 @@
-[
-  {
-    "category": "Error Handling",
-    "description": "Division by zero potential if total_tokens is zero.",
-    "impact": "critical",
-    "rationale": "Division by zero will cause the program to crash. This needs to be handled to ensure robustness.",
-    "recommendation": "Add a check to handle the case when total_tokens is zero.",
-    "current_code": "print(f\"Total tokens used:{total_tokens:,}\")",
-    "suggested_code": "if total_tokens == 0:\n    print(\"Total tokens used: 0\")\nelse:\n    print(f\"Total tokens used:{total_tokens:,}\")",
-    "file_path": "main.py",
-    "start_line": 161,
-    "end_line": 161,
-    "change_type": "addition",
-    "sentiment": "positive",
-    "severity": 9
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/issues.json b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/issues.json
deleted file mode 100644
index 68ddc5bb..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/issues.json
+++ /dev/null
@@ -1,107 +0,0 @@
-[
-  {
-    "category": "Unused Import",
-    "description": "The 'random' module is imported but not used anywhere in the code.",
-    "impact": "trivial",
-    "rationale": "Unused imports can clutter the code and may lead to confusion. They also slightly increase the memory footprint.",
-    "recommendation": "Remove the unused import statement.",
-    "current_code": "import random  # Unused import",
-    "suggested_code": "",
-    "file_path": "main.py",
-    "start_line": 8,
-    "end_line": 8,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 1
-  },
-  {
-    "category": "Error Handling",
-    "description": "Potential for API call to fail without retry mechanism.",
-    "impact": "high",
-    "rationale": "If the API call fails, the function will not handle the failure gracefully, which may lead to incomplete processing of applicant data.",
-    "recommendation": "Implement a retry mechanism with exponential backoff for the API call.",
-    "current_code": "response = completion(\n    model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n)",
-    "suggested_code": "import time\n\nfor attempt in range(3):\n    try:\n        response = completion(\n            model=os.environ.get(\"model\", \"anyscale/mistralai/Mixtral-8x22B-Instruct-v0.1\"), messages=messages\n        )\n        break\n    except Exception as e:\n        if attempt < 2:\n            time.sleep(2 ** attempt)\n        else:\n            raise e",
-    "file_path": "main.py",
-    "start_line": 66,
-    "end_line": 68,
-    "change_type": "addition",
-    "sentiment": "positive",
-    "severity": 8
-  },
-  {
-    "category": "Error Handling",
-    "description": "Silent failure without logging in JSONDecodeError exception.",
-    "impact": "medium",
-    "rationale": "Silent failures can make debugging difficult. Logging the error provides visibility into what went wrong.",
-    "recommendation": "Add logging for the JSONDecodeError exception.",
-    "current_code": "except json.JSONDecodeError:\n    result ={key: \"\" for key in[\"feedback\", \"review\", \"should_interview\", \"rating\", \"input_tokens\", \"output_tokens\"]}",
-    "suggested_code": "except json.JSONDecodeError as e:\n    print(f\"Failed to parse content for applicant:{e}\")\n    result ={key: \"\" for key in[\"feedback\", \"review\", \"should_interview\", \"rating\", \"input_tokens\", \"output_tokens\"]}",
-    "file_path": "main.py",
-    "start_line": 82,
-    "end_line": 94,
-    "change_type": "addition",
-    "sentiment": "positive",
-    "severity": 6
-  },
-  {
-    "category": "Performance",
-    "description": "Inefficient way to print progress.",
-    "impact": "low",
-    "rationale": "Printing progress in this manner can be inefficient and slow down the processing, especially for large datasets.",
-    "recommendation": "Consider using a more efficient method for progress reporting, such as updating the progress bar less frequently.",
-    "current_code": "print(f\"\\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}\", end=\"\", flush=True)",
-    "suggested_code": "if index % 10 == 0 or index == total - 1:\n    print(f\"\\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}\", end=\"\", flush=True)",
-    "file_path": "main.py",
-    "start_line": 121,
-    "end_line": 121,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 3
-  },
-  {
-    "category": "Redundant Code",
-    "description": "The check for empty DataFrame is unnecessary.",
-    "impact": "trivial",
-    "rationale": "The DataFrame length check before processing is redundant as the subsequent code will handle an empty DataFrame gracefully.",
-    "recommendation": "Remove the redundant check for an empty DataFrame.",
-    "current_code": "if len(df) == 0:\n    return",
-    "suggested_code": "",
-    "file_path": "main.py",
-    "start_line": 141,
-    "end_line": 143,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 1
-  },
-  {
-    "category": "Error Handling",
-    "description": "No error handling for file not found.",
-    "impact": "high",
-    "rationale": "If the file does not exist, the program will crash without providing a meaningful error message.",
-    "recommendation": "Add error handling for file not found.",
-    "current_code": "main(input_file)",
-    "suggested_code": "if not os.path.exists(input_file):\n    print(f\"File not found:{input_file}\")\nelse:\n    main(input_file)",
-    "file_path": "main.py",
-    "start_line": 174,
-    "end_line": 175,
-    "change_type": "addition",
-    "sentiment": "positive",
-    "severity": 7
-  },
-  {
-    "category": "Error Handling",
-    "description": "Division by zero potential if total_tokens is zero.",
-    "impact": "critical",
-    "rationale": "Division by zero will cause the program to crash. This needs to be handled to ensure robustness.",
-    "recommendation": "Add a check to handle the case when total_tokens is zero.",
-    "current_code": "print(f\"Total tokens used:{total_tokens:,}\")",
-    "suggested_code": "if total_tokens == 0:\n    print(\"Total tokens used: 0\")\nelse:\n    print(f\"Total tokens used:{total_tokens:,}\")",
-    "file_path": "main.py",
-    "start_line": 161,
-    "end_line": 161,
-    "change_type": "addition",
-    "sentiment": "positive",
-    "severity": 9
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/review.md b/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/review.md
deleted file mode 100644
index eee73c57..00000000
--- a/.experiments/code_review/code_reviews_20240827_161838/no_eval/pr_5/review.md
+++ /dev/null
@@ -1,64 +0,0 @@
-PR URL: https://github.com/sauravpanda/applicant-screening/pull/5
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 7
-- Critical: 1
-- Important: 0
-- Minor: 0
-- Files Affected: 1
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Error Handling (1 issues)</strong></summary>
-
-### 1. Division by zero potential if total_tokens is zero.
-📁 **File:** `main.py:161`
-⚖️ **Severity:** 9/10
-🔍 **Description:** Division by zero potential if total_tokens is zero.
-💡 **Solution:** 
-
-**Current Code:**
-```python
-print(f"Total tokens used:{total_tokens:,}")
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (3 issues)</strong></summary>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-2024-05-13)
-{"prompt_tokens": 6403, "completion_tokens": 1420, "total_tokens": 7823}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/comments.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/comments.json
deleted file mode 100644
index ef33088e..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/comments.json
+++ /dev/null
@@ -1,87 +0,0 @@
-[
-  {
-    "category": "Security",
-    "description": "Hardcoded API keys in config.json.",
-    "impact": "critical",
-    "rationale": "Storing sensitive information like API keys in the source code can lead to security vulnerabilities if the code is exposed.",
-    "recommendation": "Use environment variables or a secure vault service to manage sensitive information instead of hardcoding them in the configuration file.",
-    "current_code": "\"api_key\": \"os.environ/AZURE_API_KEY\"",
-    "suggested_code": "\"api_key\": \"${AZURE_API_KEY}\"",
-    "file_path": "config.json",
-    "start_line": 13,
-    "end_line": 13,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 9
-  },
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Docker",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to Dockerfile, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_path": "Dockerfile",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Docker",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to docker-compose.yml, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "15",
-    "end_line": "15",
-    "side": "RIGHT",
-    "file_path": "docker-compose.yml",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Version Control",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to .gitignore, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "164",
-    "end_line": "164",
-    "side": "RIGHT",
-    "file_path": ".gitignore",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Database",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to db_setup/init.sql, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_path": "db_setup/init.sql",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/issues.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/issues.json
deleted file mode 100644
index 9ce8a512..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/issues.json
+++ /dev/null
@@ -1,342 +0,0 @@
-[
-  {
-    "category": "Documentation",
-    "description": "Lack of comments in critical sections.",
-    "impact": "high",
-    "rationale": "While the code is mostly self-explanatory, certain sections, especially in the Dockerfile and SQL scripts, would benefit from additional comments to clarify their purpose and usage.",
-    "recommendation": "Add comments to explain the purpose of each section, especially in the Dockerfile and SQL setup scripts.",
-    "current_code": "RUN apt-get update && apt-get install -y \\\n    git \\\n    build-essential \\\n    && rm -rf /var/lib/apt/lists/*",
-    "suggested_code": "# Install system dependencies\nRUN apt-get update && apt-get install -y \\\n    git \\\n    build-essential \\\n    && rm -rf /var/lib/apt/lists/*",
-    "file_path": "Dockerfile",
-    "start_line": 7,
-    "end_line": 12,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Security",
-    "description": "Hardcoded API keys in config.json.",
-    "impact": "critical",
-    "rationale": "Storing sensitive information like API keys in the source code can lead to security vulnerabilities if the code is exposed.",
-    "recommendation": "Use environment variables or a secure vault service to manage sensitive information instead of hardcoding them in the configuration file.",
-    "current_code": "\"api_key\": \"os.environ/AZURE_API_KEY\"",
-    "suggested_code": "\"api_key\": \"${AZURE_API_KEY}\"",
-    "file_path": "config.json",
-    "start_line": 13,
-    "end_line": 13,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 9
-  },
-  {
-    "category": "Error Handling",
-    "description": "Missing error handling in the install script.",
-    "impact": "medium",
-    "rationale": "The install_tree_sitter_languages.sh script lacks error handling, which could lead to silent failures if a language fails to install.",
-    "recommendation": "Add error handling to the installation process to ensure that failures are logged and handled appropriately.",
-    "current_code": "git clone \"https://github.com/tree-sitter/tree-sitter-$lang\" \"$LANGUAGE_DIR/tree-sitter-$lang\"",
-    "suggested_code": "git clone \"https://github.com/tree-sitter/tree-sitter-$lang\" \"$LANGUAGE_DIR/tree-sitter-$lang\" ||{echo \"Failed to clone $lang\"; exit 1;}",
-    "file_path": "install_tree_sitter_languages.sh",
-    "start_line": 24,
-    "end_line": 24,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 7
-  },
-  {
-    "category": "Performance",
-    "description": "Potential inefficiency in the chunk_code function.",
-    "impact": "medium",
-    "rationale": "The function processes nodes recursively, which could lead to performance issues with larger codebases. Consider optimizing the traversal method.",
-    "recommendation": "Implement an iterative approach or optimize the recursive function to handle larger codebases more efficiently.",
-    "current_code": "def process_node(node):\n    result = parse_code(code, language)",
-    "suggested_code": "def process_node(node):\n    result = parse_code(code, language)\n    if not result:\n        return",
-    "file_path": "kaizen/retriever/code_chunker.py",
-    "start_line": 20,
-    "end_line": 22,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Readability",
-    "description": "Lack of docstrings for classes and methods.",
-    "impact": "medium",
-    "rationale": "Docstrings provide essential context for understanding the purpose and usage of classes and methods, which enhances maintainability and usability.",
-    "recommendation": "Add docstrings to all classes and methods to describe their functionality and parameters.",
-    "current_code": "",
-    "suggested_code": "\"\"\"\nClass to handle feedback abstraction.\n\nAttributes:\n    feedback_store (Dict[str, Dict[str, Any]]): A dictionary to store feedback.\n\"\"\"",
-    "file_path": "kaizen/retriever/feedback_system.py",
-    "start_line": 4,
-    "end_line": 5,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling",
-    "description": "Potential lack of error handling in database operations.",
-    "impact": "high",
-    "rationale": "Database operations can fail for various reasons (e.g., connection issues, SQL errors). Not handling these exceptions can lead to crashes or unhandled errors in production.",
-    "recommendation": "Wrap database operations in try-except blocks to handle potential exceptions gracefully.",
-    "current_code": "cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))",
-    "suggested_code": "try:\n    cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))\nexcept Exception as e:\n    # Handle exception (log it, re-raise, etc.)\n    raise RuntimeError(\"Database query failed\") from e",
-    "file_path": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 41,
-    "end_line": 41,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Performance",
-    "description": "Normalization of query embedding could be optimized.",
-    "impact": "medium",
-    "rationale": "Using numpy for normalization is efficient, but additional checks can be added to handle edge cases (e.g., zero-length vectors) to avoid potential runtime errors.",
-    "recommendation": "Add a check to ensure the query embedding is not zero-length before normalization.",
-    "current_code": "query_embedding_normalized = query_embedding_np / np.linalg.norm(query_embedding_np)",
-    "suggested_code": "norm = np.linalg.norm(query_embedding_np)\nif norm == 0:\n    raise ValueError(\"Query embedding cannot be zero-length\")\nquery_embedding_normalized = query_embedding_np / norm",
-    "file_path": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 16,
-    "end_line": 16,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Error Handling",
-    "description": "Lack of specific error handling in critical sections.",
-    "impact": "high",
-    "rationale": "While there are general exception handlers in place, specific exceptions should be caught to provide more meaningful error messages and to handle different failure scenarios appropriately.",
-    "recommendation": "Implement more granular exception handling to catch specific exceptions where they may occur, such as FileNotFoundError or SQLAlchemy specific exceptions.",
-    "current_code": "except Exception as e:",
-    "suggested_code": "except FileNotFoundError as e:\n    logger.error(f'File not found:{file_path}')\nexcept SQLAlchemyError as e:\n    logger.error(f'Database error:{str(e)}')",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 108,
-    "end_line": 110,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 7
-  },
-  {
-    "category": "Logging",
-    "description": "Inconsistent logging levels used.",
-    "impact": "medium",
-    "rationale": "The code uses different logging levels (INFO, DEBUG, ERROR) inconsistently. This can lead to confusion regarding the importance of messages and may cause critical issues to be overlooked.",
-    "recommendation": "Standardize logging levels across the codebase to ensure that critical information is logged at the appropriate level.",
-    "current_code": "logger.debug(f'Parsing file:{file_path}')",
-    "suggested_code": "logger.info(f'Parsing file:{file_path}')",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 90,
-    "end_line": 90,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Code Readability",
-    "description": "Long methods with multiple responsibilities.",
-    "impact": "high",
-    "rationale": "The methods are quite long and handle multiple responsibilities, which makes them harder to read and maintain. This violates the Single Responsibility Principle.",
-    "recommendation": "Refactor long methods into smaller, more focused methods to improve readability and maintainability.",
-    "current_code": "def parse_file(self, file_path: str):\n    ...",
-    "suggested_code": "def parse_file(self, file_path: str):\n    self.read_file(file_path)\n    self.process_file_content(content)\n    ...",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 89,
-    "end_line": 110,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Performance",
-    "description": "Potential inefficiency in database operations.",
-    "impact": "high",
-    "rationale": "The code performs multiple database operations in a loop which can lead to performance issues, especially with a large number of records.",
-    "recommendation": "Batch database operations where possible to reduce the number of transactions and improve performance.",
-    "current_code": "connection.execute(query,{...})",
-    "suggested_code": "with connection.begin():\n    connection.execute(batch_query, batch_data)",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 298,
-    "end_line": 312,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 8
-  },
-  {
-    "category": "Documentation",
-    "description": "Insufficient comments and documentation.",
-    "impact": "medium",
-    "rationale": "While the code is largely self-explanatory, additional comments and documentation would enhance understanding, especially for complex methods.",
-    "recommendation": "Add docstrings to all methods and inline comments where necessary to explain complex logic.",
-    "current_code": "def store_code_in_db(self, code: str, abstraction: str, file_path: str, section: str, name: str):",
-    "suggested_code": "def store_code_in_db(self, code: str, abstraction: str, file_path: str, section: str, name: str):\n    \"\"\"\n    Stores the provided code and its abstraction in the database.\n    Args:\n        code (str): The code to store.\n        abstraction (str): The abstract description of the code.\n        file_path (str): The path of the file containing the code.\n        section (str): The section of the code (e.g., functions).\n        name (str): The name of the function or code block.\n    \"\"\"",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 246,
-    "end_line": 248,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 4
-  },
-  {
-    "category": "Error Handling",
-    "description": "General exception handling in language loading and parsing.",
-    "impact": "high",
-    "rationale": "Using a broad exception catch can obscure specific errors, making debugging difficult. It's better to catch specific exceptions where possible.",
-    "recommendation": "Refine exception handling to catch specific exceptions instead of using a generic Exception.",
-    "current_code": "except Exception as e:",
-    "suggested_code": "except (ImportError, ValueError) as e:",
-    "file_path": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 28,
-    "end_line": 30,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 7
-  },
-  {
-    "category": "Logging",
-    "description": "Inconsistent logging levels for errors and warnings.",
-    "impact": "medium",
-    "rationale": "Using different logging levels can lead to confusion about the severity of issues. It's important to use logging levels consistently.",
-    "recommendation": "Use logger.warning for recoverable errors and logger.error for critical failures.",
-    "current_code": "logger.error(f\"Failed to load language{language}:{str(e)}\")",
-    "suggested_code": "logger.warning(f\"Failed to load language{language}:{str(e)}\")",
-    "file_path": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 108,
-    "end_line": 108,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Code Readability",
-    "description": "Lack of comments in complex sections.",
-    "impact": "medium",
-    "rationale": "While the code is mostly clear, adding comments in complex sections (like dynamic imports and tree traversal) would enhance readability for future developers.",
-    "recommendation": "Add comments explaining the purpose and functionality of complex code blocks.",
-    "current_code": "",
-    "suggested_code": "# Dynamically import the language module\n# Traverse the tree structure to extract relevant information",
-    "file_path": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 20,
-    "end_line": 21,
-    "change_type": "addition",
-    "sentiment": "positive",
-    "severity": 4
-  },
-  {
-    "category": "Performance",
-    "description": "Caching of language loading is not explicitly documented.",
-    "impact": "medium",
-    "rationale": "The use of lru_cache is a good performance optimization, but it's important to document its purpose and implications for maintainability.",
-    "recommendation": "Add a docstring to the load_language method explaining the caching behavior.",
-    "current_code": "",
-    "suggested_code": "\"\"\"Load and cache the specified language module using LRU caching.\"\"\"",
-    "file_path": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 15,
-    "end_line": 15,
-    "change_type": "addition",
-    "sentiment": "positive",
-    "severity": 3
-  },
-  {
-    "category": "Dependency Management",
-    "description": "Updating Python version in pyproject.toml.",
-    "impact": "medium",
-    "rationale": "Updating the Python version to a more recent one can improve compatibility with newer libraries and features.",
-    "recommendation": "Ensure all dependencies are compatible with the updated Python version.",
-    "current_code": "python = \"^3.8.1\"",
-    "suggested_code": "python = \"^3.9.0\"",
-    "file_path": "pyproject.toml",
-    "start_line": 13,
-    "end_line": 13,
-    "change_type": "modification",
-    "sentiment": "positive",
-    "severity": 5
-  },
-  {
-    "category": "Testing",
-    "description": "Lack of unit tests for new functionality.",
-    "impact": "high",
-    "rationale": "New features should have corresponding unit tests to ensure functionality and prevent regressions.",
-    "recommendation": "Add unit tests for the new language loading and parsing functionalities.",
-    "current_code": "",
-    "suggested_code": "# Add unit tests for LanguageLoader and ParserFactory",
-    "file_path": "tests/retriever/test_chunker.py",
-    "start_line": 1,
-    "end_line": 1,
-    "change_type": "addition",
-    "sentiment": "positive",
-    "severity": 8
-  },
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Docker",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to Dockerfile, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_path": "Dockerfile",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Docker",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to docker-compose.yml, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "15",
-    "end_line": "15",
-    "side": "RIGHT",
-    "file_path": "docker-compose.yml",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Version Control",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to .gitignore, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "164",
-    "end_line": "164",
-    "side": "RIGHT",
-    "file_path": ".gitignore",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Database",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to db_setup/init.sql, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_path": "db_setup/init.sql",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/review.md b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/review.md
deleted file mode 100644
index 907b5595..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_222/review.md
+++ /dev/null
@@ -1,136 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/222
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 23
-- Critical: 6
-- Important: 0
-- Minor: 0
-- Files Affected: 13
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Security (6 issues)</strong></summary>
-
-### 1. Hardcoded API keys in config.json.
-📁 **File:** `config.json:13`
-⚖️ **Severity:** 9/10
-🔍 **Description:** Hardcoded API keys in config.json.
-💡 **Solution:** 
-
-**Current Code:**
-```python
-"api_key": "os.environ/AZURE_API_KEY"
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 2. Changes made to sensitive file
-📁 **File:** `config.json:4`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 3. Changes made to sensitive file
-📁 **File:** `Dockerfile:4`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 4. Changes made to sensitive file
-📁 **File:** `docker-compose.yml:15`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 5. Changes made to sensitive file
-📁 **File:** `.gitignore:164`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 6. Changes made to sensitive file
-📁 **File:** `db_setup/init.sql:1`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 22541, "completion_tokens": 3669, "total_tokens": 26210}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/comments.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/comments.json
deleted file mode 100644
index 0637a088..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/comments.json
+++ /dev/null
@@ -1 +0,0 @@
-[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/issues.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/issues.json
deleted file mode 100644
index 6c136153..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/issues.json
+++ /dev/null
@@ -1,197 +0,0 @@
-[
-  {
-    "category": "Import Optimization",
-    "description": "Redundant imports removed.",
-    "impact": "medium",
-    "rationale": "Removing unused imports helps in reducing bundle size and improving readability.",
-    "recommendation": "Ensure that all imports are necessary and remove any that are not used in the code.",
-    "current_code": "import{MemoriesIcon, NextIcon, SearchIcon, UrlIcon}from \"@repo/ui/icons\";",
-    "suggested_code": "import{MemoriesIcon, NextIcon, UrlIcon}from \"@repo/ui/icons\";",
-    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 4,
-    "end_line": 4,
-    "change_type": "modification",
-    "sentiment": "positive",
-    "severity": 5
-  },
-  {
-    "category": "Type Definition",
-    "description": "Type definition for TMemoriesPage is added.",
-    "impact": "high",
-    "rationale": "Providing a type definition improves type safety and makes the code easier to understand and maintain.",
-    "recommendation": "Consider adding more detailed comments to explain the purpose of each property in the type definition.",
-    "current_code": "",
-    "suggested_code": "type TMemoriesPage ={\n  memoriesAndSpaces:{memories: Content[]; spaces: StoredSpace[]};\n  title?: string;\n  currentSpace?: StoredSpace;\n  usersWithAccess?: string[];\n};",
-    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 40,
-    "end_line": 44,
-    "change_type": "addition",
-    "sentiment": "positive",
-    "severity": 7
-  },
-  {
-    "category": "Error Handling",
-    "description": "Improved error message for space deletion.",
-    "impact": "medium",
-    "rationale": "Providing a more specific error message can help users understand the issue better.",
-    "recommendation": "Consider logging the error or providing more context in the error message.",
-    "current_code": "toast.error(\"Failed to delete space\");",
-    "suggested_code": "toast.error(\"Failed to delete the space\");",
-    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 75,
-    "end_line": 75,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 4
-  },
-  {
-    "category": "Code Readability",
-    "description": "Improved readability of JSX structure.",
-    "impact": "medium",
-    "rationale": "Improving the structure of JSX can enhance readability and maintainability.",
-    "recommendation": "Consider using more descriptive class names and breaking down complex components into smaller ones.",
-    "current_code": "<div className=\"px-2 md:px-32 py-36 h-full flex mx-auto w-full flex-col gap-6\">",
-    "suggested_code": "<div className=\"px-2 md:px-32 py-36 h-full flex mx-auto w-full flex-col gap-6\">",
-    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 140,
-    "end_line": 140,
-    "change_type": "modification",
-    "sentiment": "positive",
-    "severity": 3
-  },
-  {
-    "category": "Component Naming",
-    "description": "Renamed TabComponent to SpaceComponent for clarity.",
-    "impact": "medium",
-    "rationale": "Using descriptive names for components improves code readability and understanding.",
-    "recommendation": "Ensure that component names reflect their functionality.",
-    "current_code": "function TabComponent({",
-    "suggested_code": "function SpaceComponent({",
-    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 275,
-    "end_line": 275,
-    "change_type": "modification",
-    "sentiment": "positive",
-    "severity": 5
-  },
-  {
-    "category": "Code Removal",
-    "description": "Removal of key state variables and imports without clear justification.",
-    "impact": "high",
-    "rationale": "The removal of state variables like 'spaces', 'content', and 'selectedSpaces' can lead to loss of functionality in the Menu component. Additionally, removing imports like 'useMeasure' may indicate missing functionality that could be required for layout measurements.",
-    "recommendation": "Reassess the necessity of removing these variables and imports. If they are not needed, ensure that the functionality relying on them is also removed or refactored appropriately.",
-    "current_code": "import useMeasure from \"react-use-measure\";\nconst[spaces, setSpaces] = useState<StoredSpace[]>([]);\nconst[content, setContent] = useState(\"\");\nconst[selectedSpaces, setSelectedSpaces] = useState<number[]>([]);",
-    "suggested_code": "import useMeasure from \"react-use-measure\";\nconst[spaces, setSpaces] = useState<StoredSpace[]>([]);\nconst[content, setContent] = useState(\"\");\nconst[selectedSpaces, setSelectedSpaces] = useState<number[]>([]);",
-    "file_path": "apps/web/app/(dash)/menu.tsx",
-    "start_line": 32,
-    "end_line": 80,
-    "change_type": "deletion",
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Error Handling",
-    "description": "Inconsistent error handling in async functions.",
-    "impact": "high",
-    "rationale": "The error handling in the 'handleSubmit' function is not consistent. It throws an error after displaying a toast, which could lead to unhandled promise rejections. This can cause the application to crash or behave unexpectedly.",
-    "recommendation": "Ensure that all async functions have consistent error handling. Consider using try-catch blocks to manage errors gracefully and avoid throwing errors after displaying user notifications.",
-    "current_code": "if (cont.success){\n    toast.success(\"Memory queued\",{\n        richColors: true,\n});\n}else{\n    toast.error(`Memory creation failed: ${cont.error}`);\n    throw new Error(`Memory creation failed: ${cont.error}`);\n}",
-    "suggested_code": "try{\n    const cont = await createMemory({\n        content: content,\n        spaces: spaces ?? undefined,\n});\n    if (cont.success){\n        toast.success(\"Memory queued\",{\n            richColors: true,\n});\n}else{\n        throw new Error(`Memory creation failed: ${cont.error}`);\n}\n}catch (error){\n    toast.error(error.message);\n}",
-    "file_path": "apps/web/app/(dash)/menu.tsx",
-    "start_line": 112,
-    "end_line": 130,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Readability",
-    "description": "Complexity in JSX structure can hinder readability.",
-    "impact": "medium",
-    "rationale": "The JSX structure in the 'DialogContentContainer' function is quite complex and may be difficult for future developers to read and maintain. Breaking it down into smaller components can improve readability and maintainability.",
-    "recommendation": "Consider breaking down the JSX structure into smaller, reusable components. This will enhance readability and make it easier to manage the code in the future.",
-    "current_code": "<DialogContent className=\"sm:max-w-[575px] text-[#F2F3F5] rounded-2xl bg-background z-[39]\">\n    <form action={async (e: FormData) =>{...}}>\n{/* Form Fields */}\n    </form>\n</DialogContent>",
-    "suggested_code": "<DialogContent className=\"sm:max-w-[575px] text-[#F2F3F5] rounded-2xl bg-background z-[39]\">\n    <FormComponent onSubmit={handleSubmit}/>\n</DialogContent>",
-    "file_path": "apps/web/app/(dash)/menu.tsx",
-    "start_line": 135,
-    "end_line": 346,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Import Optimization",
-    "description": "Unused import from React.",
-    "impact": "medium",
-    "rationale": "The import statement for useEffect is removed but not used in the code, which can lead to confusion about the necessity of the import.",
-    "recommendation": "Remove unused imports to enhance code clarity and maintainability.",
-    "current_code": "import{useState, useEffect}from \"react\";",
-    "suggested_code": "import{useState}from \"react\";",
-    "file_path": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 3,
-    "end_line": 3,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Interface Definition",
-    "description": "New interface 'StoredSpace' added.",
-    "impact": "high",
-    "rationale": "The addition of the 'StoredSpace' interface improves type safety and clarity in the code, allowing for better maintainability and understanding of the data structure.",
-    "recommendation": "Ensure that all interfaces are documented to provide context for future developers.",
-    "current_code": "",
-    "suggested_code": "interface StoredSpace{\n\tid: number;\n\tname: string;\n}",
-    "file_path": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 19,
-    "end_line": 22,
-    "change_type": "addition",
-    "sentiment": "positive",
-    "severity": 7
-  },
-  {
-    "category": "Props Structure",
-    "description": "Props structure modified to include selectedSpaces and setSelectedSpaces.",
-    "impact": "high",
-    "rationale": "This change enhances the component's functionality by allowing it to manage selected spaces, which is crucial for the component's purpose.",
-    "recommendation": "Consider adding PropTypes or TypeScript documentation for better clarity on prop types.",
-    "current_code": "placeholder?: string;\nemptyMessage?: string;\ncreateNewMessage?: string;\nclassName?: string;",
-    "suggested_code": "selectedSpaces: number[];\nsetSelectedSpaces: React.Dispatch<React.SetStateAction<number[]>>;",
-    "file_path": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 23,
-    "end_line": 28,
-    "change_type": "modification",
-    "sentiment": "positive",
-    "severity": 6
-  },
-  {
-    "category": "Event Handling",
-    "description": "Event handlers for input change and key down events added.",
-    "impact": "high",
-    "rationale": "These handlers improve user interaction by providing feedback and functionality when users type or press keys, enhancing the overall user experience.",
-    "recommendation": "Ensure that event handlers are well-tested to handle edge cases, such as rapid key presses.",
-    "current_code": "",
-    "suggested_code": "const handleInputChange = (e: React.ChangeEvent<HTMLInputElement>) =>{\n\tsetInputValue(e.target.value);\n};\n\nconst handleKeyDown = (e: React.KeyboardEvent<HTMLInputElement>) =>{\n\tif (\n\t\te.key === \"Backspace\" &&\n\t\tinputValue === \"\" &&\n\t\tselectedSpaces.length > 0\n\t\t){\n\t\tsetSelectedSpaces((prev) => prev.slice(0, -1));\n\t}\n};",
-    "file_path": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 41,
-    "end_line": 53,
-    "change_type": "addition",
-    "sentiment": "positive",
-    "severity": 7
-  },
-  {
-    "category": "Filtering Logic",
-    "description": "Filtering logic for options based on selectedSpaces added.",
-    "impact": "high",
-    "rationale": "This logic ensures that the user only sees options that are not already selected, improving usability and preventing confusion.",
-    "recommendation": "Consider adding unit tests to verify that the filtering logic works as expected under various scenarios.",
-    "current_code": "",
-    "suggested_code": "const filteredOptions = options.filter(\n\t(option) => !selectedSpaces.includes(parseInt(option.value)),\n);",
-    "file_path": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 55,
-    "end_line": 57,
-    "change_type": "addition",
-    "sentiment": "positive",
-    "severity": 6
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/review.md b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/review.md
deleted file mode 100644
index ea09b5db..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_232/review.md
+++ /dev/null
@@ -1,33 +0,0 @@
-PR URL: https://github.com/supermemoryai/supermemory/pull/232
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 13
-- Critical: 0
-- Important: 0
-- Minor: 0
-- Files Affected: 3
-## 🏆 Code Quality
-[███████████████░░░░░] 75% (Fair)
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 16642, "completion_tokens": 2919, "total_tokens": 19561}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/comments.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/comments.json
deleted file mode 100644
index 0637a088..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/comments.json
+++ /dev/null
@@ -1 +0,0 @@
-[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/issues.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/issues.json
deleted file mode 100644
index 0f16989d..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/issues.json
+++ /dev/null
@@ -1,62 +0,0 @@
-[
-  {
-    "category": "Code Structure",
-    "description": "Inconsistent parameter formatting in function calls.",
-    "impact": "medium",
-    "rationale": "The change to the `generate_linkedin_post` function call introduces a multi-line format, which could reduce readability and consistency in function calls across the codebase.",
-    "recommendation": "Consider keeping function calls in a single line unless they exceed a reasonable length. This will enhance readability and maintainability.",
-    "current_code": "linkedin_post = work_summary_generator.generate_linkedin_post(\n     summary, user=\"oss_example\"\n )",
-    "suggested_code": "linkedin_post = work_summary_generator.generate_linkedin_post(summary, user=\"oss_example\")",
-    "file_path": "examples/work_summarizer/main.py",
-    "start_line": 60,
-    "end_line": 62,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling",
-    "description": "Missing error handling for JSON parsing.",
-    "impact": "high",
-    "rationale": "The `json.loads` function in `kaizen/helpers/parser.py` does not handle potential exceptions that may arise from malformed JSON, which could lead to runtime crashes.",
-    "recommendation": "Wrap the JSON parsing in a try-except block to handle possible exceptions and provide meaningful error messages.",
-    "current_code": "parsed_data = json.loads(json_data)",
-    "suggested_code": "try:\n    parsed_data = json.loads(json_data)\nexcept json.JSONDecodeError as e:\n    logging.error(f\"Failed to parse JSON:{e}\")\n    return None",
-    "file_path": "kaizen/helpers/parser.py",
-    "start_line": 47,
-    "end_line": 48,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Documentation",
-    "description": "Lack of docstrings for new functions.",
-    "impact": "medium",
-    "rationale": "New functions like `generate_twitter_post` and `generate_linkedin_post` lack docstrings, making it difficult for future developers to understand their purpose and usage.",
-    "recommendation": "Add docstrings to all new functions to describe their parameters, return types, and functionality.",
-    "current_code": "",
-    "suggested_code": "\"\"\"\nGenerates a Twitter post based on the provided summary.\n\nArgs:\n    summary (Dict): The summary data.\n    user (Optional[str]): The user requesting the post.\n\nReturns:\n    str: The generated Twitter post.\n\"\"\"",
-    "file_path": "kaizen/reviewer/work_summarizer.py",
-    "start_line": 58,
-    "end_line": 66,
-    "change_type": "addition",
-    "sentiment": "positive",
-    "severity": 6
-  },
-  {
-    "category": "Code Consistency",
-    "description": "Inconsistent naming conventions for constants.",
-    "impact": "medium",
-    "rationale": "The constants `TWITTER_POST_PROMPT` and `LINKEDIN_POST_PROMPT` are defined without a consistent naming style compared to other constants in the codebase.",
-    "recommendation": "Ensure all constants follow a consistent naming convention, such as using uppercase letters with underscores.",
-    "current_code": "TWITTER_POST_PROMPT = \"\"\"\nGiven the following work summary, create a concise and engaging Twitter post (max 280 characters) that highlights the key changes or improvements. Format the post as markdown, enclosed in triple backticks:\n\nSummary:\n{SUMMARY}\n\nTwitter Post:\n```\n<Your Twitter post here>\n```\n\"\"\"",
-    "suggested_code": "TWITTER_POST_PROMPT = \"\"\"\nGiven the following work summary, create a concise and engaging Twitter post (max 280 characters) that highlights the key changes or improvements. Format the post as markdown, enclosed in triple backticks:\n\nSummary:\n{SUMMARY}\n\nTwitter Post:\n```\n<Your Twitter post here>\n```\n\"\"\"",
-    "file_path": "kaizen/llms/prompts/work_summary_prompts.py",
-    "start_line": 44,
-    "end_line": 54,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 4
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/review.md b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/review.md
deleted file mode 100644
index 3de0640e..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_252/review.md
+++ /dev/null
@@ -1,33 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/252
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 4
-- Critical: 0
-- Important: 0
-- Minor: 0
-- Files Affected: 4
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 4135, "completion_tokens": 989, "total_tokens": 5124}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/comments.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/comments.json
deleted file mode 100644
index 0637a088..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/comments.json
+++ /dev/null
@@ -1 +0,0 @@
-[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/issues.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/issues.json
deleted file mode 100644
index b6b4d441..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/issues.json
+++ /dev/null
@@ -1,62 +0,0 @@
-[
-  {
-    "category": "Import Changes",
-    "description": "Updated import paths for prompts.",
-    "impact": "medium",
-    "rationale": "Changing import paths can lead to confusion if not documented properly, especially if the previous imports were widely used.",
-    "recommendation": "Ensure that the new import paths are well-documented and tested to avoid runtime errors.",
-    "current_code": "from kaizen.llms.prompts.code_review_prompts import (",
-    "suggested_code": "from kaizen.llms.prompts.pr_desc_prompts import (",
-    "file_path": "kaizen/generator/pr_description.py",
-    "start_line": 8,
-    "end_line": 8,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Parameter Removal",
-    "description": "Removed 'reeval_response' parameter from multiple functions.",
-    "impact": "high",
-    "rationale": "Removing parameters can break existing functionality if the functions are called elsewhere with those parameters.",
-    "recommendation": "Review all usages of these functions to ensure that removing 'reeval_response' does not cause issues. If it's unnecessary, consider marking it as deprecated instead of removing it outright.",
-    "current_code": "reeval_response: bool = False,",
-    "suggested_code": "",
-    "file_path": "kaizen/generator/pr_description.py",
-    "start_line": 43,
-    "end_line": 43,
-    "change_type": "deletion",
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Error Handling",
-    "description": "Potential lack of error handling after removing 'reeval_response'.",
-    "impact": "high",
-    "rationale": "The removal of this parameter could lead to unhandled exceptions if the logic that depended on it is not properly adjusted.",
-    "recommendation": "Add error handling or checks to ensure that the new logic remains robust and handles edge cases appropriately.",
-    "current_code": "desc = self._process_full_diff(prompt, user)",
-    "suggested_code": "if not prompt or not user: raise ValueError('Invalid parameters')\ndesc = self._process_full_diff(prompt, user)",
-    "file_path": "kaizen/generator/pr_description.py",
-    "start_line": 54,
-    "end_line": 54,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Code Clarity",
-    "description": "Improved clarity in prompt descriptions.",
-    "impact": "medium",
-    "rationale": "The new prompts are clearer and more structured, which enhances readability and maintainability.",
-    "recommendation": "Continue to follow this pattern for future prompt definitions to maintain consistency.",
-    "current_code": "PR_DESCRIPTION_PROMPT = \"\"\"",
-    "suggested_code": "PR_DESCRIPTION_SYSTEM_PROMPT = \"\"\"",
-    "file_path": "kaizen/llms/prompts/pr_desc_prompts.py",
-    "start_line": 1,
-    "end_line": 1,
-    "change_type": "addition",
-    "sentiment": "positive",
-    "severity": 3
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/review.md b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/review.md
deleted file mode 100644
index 4919cde8..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_335/review.md
+++ /dev/null
@@ -1,33 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/335
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 4
-- Critical: 0
-- Important: 0
-- Minor: 0
-- Files Affected: 2
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 7000, "completion_tokens": 730, "total_tokens": 7730}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/comments.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/comments.json
deleted file mode 100644
index 1ceabeef..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/comments.json
+++ /dev/null
@@ -1,16 +0,0 @@
-[
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "11",
-    "end_line": "11",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/issues.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/issues.json
deleted file mode 100644
index d371af1f..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/issues.json
+++ /dev/null
@@ -1,346 +0,0 @@
-[
-  {
-    "category": "Code Clarity",
-    "description": "Improper formatting of multi-line strings.",
-    "impact": "medium",
-    "rationale": "The formatting of the `DESC_COLLAPSIBLE_TEMPLATE` string could lead to confusion regarding its structure and readability. Multi-line strings should be formatted consistently to enhance clarity.",
-    "recommendation": "Use triple quotes for multi-line strings to maintain readability and avoid confusion.",
-    "current_code": "DESC_COLLAPSIBLE_TEMPLATE = \"<details><summary>Original Description</summary>{desc}</details>\"",
-    "suggested_code": "DESC_COLLAPSIBLE_TEMPLATE = \"\"\"\n<details><summary>Original Description</summary>{desc}</details>\n\"\"\"",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
-    "start_line": 5,
-    "end_line": 6,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling",
-    "description": "Lack of type validation for function parameters.",
-    "impact": "high",
-    "rationale": "The function `create_pr_description` should validate input types to prevent runtime errors. Currently, it assumes that inputs are always strings, which may lead to unexpected behavior.",
-    "recommendation": "Implement type checks for the parameters and raise appropriate exceptions if the types are incorrect.",
-    "current_code": "def create_pr_description(desc, original_desc):",
-    "suggested_code": "def create_pr_description(desc: str, original_desc: str):\n    if not isinstance(desc, str):\n        raise TypeError('desc must be a string')\n    if not isinstance(original_desc, str):\n        raise TypeError('original_desc must be a string')",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/output.py",
-    "start_line": 1,
-    "end_line": 1,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Test Coverage",
-    "description": "Insufficient coverage for edge cases in tests.",
-    "impact": "high",
-    "rationale": "While there are tests for normal cases, edge cases such as extremely long strings or special characters are not adequately covered, which could lead to failures in production.",
-    "recommendation": "Add more test cases to cover edge scenarios, including empty strings, very long strings, and special characters.",
-    "current_code": "def test_create_pr_description_normal_and_edge_cases(desc, original_desc, expected):",
-    "suggested_code": "def test_create_pr_description_edge_cases():\n    # Add tests for edge cases here",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
-    "start_line": 28,
-    "end_line": 28,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 7
-  },
-  {
-    "category": "Performance",
-    "description": "Potential performance issue with repeated string concatenation.",
-    "impact": "medium",
-    "rationale": "Using repeated string concatenation can lead to performance degradation, especially with large strings. It is more efficient to use `str.join()` for concatenation.",
-    "recommendation": "Refactor string concatenation to use `str.join()` for better performance.",
-    "current_code": "result = desc + original_desc",
-    "suggested_code": "result = ''.join([desc, original_desc])",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/output.py",
-    "start_line": 10,
-    "end_line": 10,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Code Structure",
-    "description": "Redundant code removal and template improvement.",
-    "impact": "high",
-    "rationale": "The original PR_COLLAPSIBLE_TEMPLATE was overly verbose and could lead to maintenance challenges. The new format is cleaner and easier to read.",
-    "recommendation": "Ensure that all templates follow a consistent structure to improve readability and maintainability.",
-    "current_code": "PR_COLLAPSIBLE_TEMPLATE = (\n    \"<details>\\n\"\n    \"<summary>Review</summary>\\n\"\n    \"<p>\\n\"\n    \"<b>Comment:</b>{comment}<br>\\n\"\n    \"<b>Reason:</b>{reason}<br>\\n\"\n    \"<b>Solution:</b>{solution}<br>\\n\"\n    \"<b>Confidence:</b>{confidence}<br>\\n\"\n    \"<b>Start Line:</b>{start_line}<br>\\n\"\n    \"<b>End Line:</b>{end_line}<br>\\n\"\n    \"<b>File Name:</b>{file_name}<br>\\n\"\n    \"<b>Severity:</b>{severity}<br>\\n\"\n    \"</p>\\n\"\n    \"</details>\"\n)",
-    "suggested_code": "PR_COLLAPSIBLE_TEMPLATE = \"\"\"\n<details>\n<summary>Review Comment</summary>\n<p>{comment}</p>\n<p><strong>Reason:</strong>{reason}</p>\n<p><strong>Solution:</strong>{solution}</p>\n<p><strong>Confidence:</strong>{confidence}</p>\n<p><strong>Start Line:</strong>{start_line}</p>\n<p><strong>End Line:</strong>{end_line}</p>\n<p><strong>File Name:</strong>{file_name}</p>\n<p><strong>Severity:</strong>{severity}</p>\n</details>\n\"\"\"",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 4,
-    "end_line": 16,
-    "change_type": "modification",
-    "sentiment": "positive",
-    "severity": 7
-  },
-  {
-    "category": "Error Handling",
-    "description": "Missing validation for required fields in review comments.",
-    "impact": "high",
-    "rationale": "The tests should validate that all required fields are provided to avoid runtime errors when generating review texts.",
-    "recommendation": "Implement checks to ensure that all required fields (comment, reason, solution, confidence, start_line, end_line, file_name, severity_level) are present before processing.",
-    "current_code": "topics ={\n    \"topic1\":[\n{\n            \"comment\": \"This is a test comment.\",\n            \"reason\": \"This is a test reason.\",\n            \"solution\": \"This is a test solution.\",\n            \"confidence\": \"critical\",\n            \"start_line\": 10,\n            \"end_line\": 20,\n            \"file_name\": \"test_file.py\",\n            \"severity_level\": 9,\n}\n    ]\n}",
-    "suggested_code": "def validate_review(review):\n    required_fields =[\"comment\", \"reason\", \"solution\", \"confidence\", \"start_line\", \"end_line\", \"file_name\", \"severity_level\"]\n    for field in required_fields:\n        if field not in review:\n            raise ValueError(f'Missing required field:{field}')",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 18,
-    "end_line": 18,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 8
-  },
-  {
-    "category": "Test Coverage",
-    "description": "Insufficient tests for edge cases.",
-    "impact": "medium",
-    "rationale": "While there are tests for normal cases, edge cases such as missing fields or empty topics are not adequately covered.",
-    "recommendation": "Add unit tests that specifically check for edge cases, such as missing fields or empty lists, to ensure robustness.",
-    "current_code": "def test_empty_topics():\n    topics ={}\n    expected_output = \"## Code Review\\n\\n\u2705 **All Clear:** This PR is ready to merge! \ud83d\udc4d\\n\\n\"\n    assert create_pr_review_text(topics) == expected_output",
-    "suggested_code": "def test_reviews_with_missing_fields():\n    topics ={\n        \"topic1\":[{\n            \"comment\": \"This is a test comment.\",\n            \"confidence\": \"critical\",\n            \"start_line\": 10,\n            \"end_line\": 20,\n            \"file_name\": \"test_file.py\",\n            \"severity_level\": 9,\n}]\n}\n    with pytest.raises(ValueError):\n        create_pr_review_text(topics)",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 238,
-    "end_line": 238,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Readability",
-    "description": "Inconsistent naming conventions for importance values.",
-    "impact": "medium",
-    "rationale": "The importance values are sometimes written in lowercase ('high') and sometimes in title case ('High'). This inconsistency can lead to confusion and errors when comparing or processing these values.",
-    "recommendation": "Standardize the naming convention for importance values across the codebase. Choose either all lowercase or title case and apply it consistently.",
-    "current_code": "            \"importance\": \"high\",",
-    "suggested_code": "            \"importance\": \"High\",",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 56,
-    "end_line": 56,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling",
-    "description": "Lack of error handling for file operations.",
-    "impact": "high",
-    "rationale": "The code does not handle potential exceptions that may arise from file operations, such as file not found or permission errors. This could lead to unhandled exceptions during runtime.",
-    "recommendation": "Wrap file operations in try-except blocks to handle potential exceptions gracefully and provide meaningful error messages.",
-    "current_code": "    with open(file_path, 'r') as f:",
-    "suggested_code": "    try:\n        with open(file_path, 'r') as f:\n            return f.read()\n    except FileNotFoundError:\n        raise Exception(f\"File not found:{file_path}\")\n    except PermissionError:\n        raise Exception(f\"Permission denied:{file_path}\")",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 17,
-    "end_line": 17,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Performance",
-    "description": "Inefficient handling of long test names.",
-    "impact": "medium",
-    "rationale": "The code currently constructs long test names by concatenating strings, which can be inefficient and hard to read. This could be optimized for better performance and clarity.",
-    "recommendation": "Use f-strings for constructing long test names to improve readability and performance.",
-    "current_code": "     long_test_name = \"Test \" + \"Example \" * 50",
-    "suggested_code": "     long_test_name = f\"Test{' '.join(['Example'] * 50)}\"",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 131,
-    "end_line": 131,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 4
-  },
-  {
-    "category": "Maintainability",
-    "description": "Redundant mocking of dependencies in multiple tests.",
-    "impact": "medium",
-    "rationale": "Mocking dependencies is repeated across multiple test functions, which can lead to code duplication and make maintenance harder.",
-    "recommendation": "Consider using a fixture for mocking dependencies to reduce code duplication and improve maintainability.",
-    "current_code": "@mock.patch(\"kaizen.helpers.output.create_folder\")\n@mock.patch(\"kaizen.helpers.output.general.clean_python_code\")\n@mock.patch(\"kaizen.helpers.output.logger\")\ndef test_create_test_files_normal_case(",
-    "suggested_code": "@pytest.fixture\ndef mock_dependencies():\n    with mock.patch('kaizen.helpers.output.create_folder') as mock_create_folder,\n         mock.patch('kaizen.helpers.output.general.clean_python_code') as mock_clean_python_code,\n         mock.patch('kaizen.helpers.output.logger') as mock_logger:\n        yield mock_create_folder, mock_clean_python_code, mock_logger",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 7,
-    "end_line": 13,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Test Coverage",
-    "description": "Missing tests for edge cases and error handling.",
-    "impact": "high",
-    "rationale": "While the tests cover normal cases, edge cases and error handling scenarios are crucial for ensuring the robustness of the code. Without these tests, potential bugs may go unnoticed.",
-    "recommendation": "Add tests for edge cases such as invalid inputs and unexpected behaviors.",
-    "current_code": "def test_get_web_html_normal_case(mock_nest_asyncio_apply, mock_get_html):",
-    "suggested_code": "def test_get_web_html_edge_cases(mock_nest_asyncio_apply, mock_get_html):",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 14,
-    "end_line": 14,
-    "change_type": "addition",
-    "sentiment": "neutral",
-    "severity": 7
-  },
-  {
-    "category": "Error Handling",
-    "description": "Inconsistent error messages in exception handling.",
-    "impact": "medium",
-    "rationale": "The error messages in the exception handling do not provide clear context for debugging. Consistent and descriptive error messages are important for understanding failures.",
-    "recommendation": "Use consistent error messages that include context about the operation being performed.",
-    "current_code": "with pytest.raises(OSError, match=\"Unable to determine current working directory\"):",
-    "suggested_code": "with pytest.raises(OSError, match=\"Failed to get current working directory during get_parent_folder call\"):",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py",
-    "start_line": 21,
-    "end_line": 21,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Code Readability",
-    "description": "Lack of comments and documentation for new functions.",
-    "impact": "medium",
-    "rationale": "New functions introduced in the tests lack comments explaining their purpose and usage, which can hinder understanding for future developers.",
-    "recommendation": "Add docstrings to new functions to describe their purpose and expected behavior.",
-    "current_code": "def test_get_web_html_normal():",
-    "suggested_code": "def test_get_web_html_normal():  # Test case for normal HTML input",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 13,
-    "end_line": 13,
-    "change_type": "modification",
-    "sentiment": "positive",
-    "severity": 4
-  },
-  {
-    "category": "Performance",
-    "description": "Potential performance issues with large HTML content.",
-    "impact": "medium",
-    "rationale": "The test for large HTML content may lead to performance issues if the content size grows significantly. This could slow down test execution.",
-    "recommendation": "Consider limiting the size of the HTML content in tests or using mocks to simulate large content.",
-    "current_code": "large_html_content = \"<html><body>\" + \"<p>Test</p>\" * 10000 + \"</body></html>\"",
-    "suggested_code": "large_html_content = \"<html><body><p>Test</p></body></html>\"  # Simplified for performance",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 93,
-    "end_line": 93,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Code Duplication",
-    "description": "Redundant imports and definitions of supported languages.",
-    "impact": "high",
-    "rationale": "The original code had multiple definitions of supported languages and unnecessary imports, which can lead to confusion and maintenance challenges.",
-    "recommendation": "Consolidate the supported languages into a single definition and ensure imports are only made once.",
-    "current_code": "self.supported_languages ={\n    \"py\": \"PythonParser\",\n    \"js\": \"JavaScriptParser\",\n    \"ts\": \"TypeScriptParser\",\n    \"jsx\": \"ReactParser\",\n    \"tsx\": \"ReactTSParser\",\n    \"rs\": \"RustParser\",\n}",
-    "suggested_code": "SUPPORTED_LANGUAGES ={\n    \"py\": \"PythonParser\",\n    \"js\": \"JavaScriptParser\",\n    \"ts\": \"TypeScriptParser\",\n    \"jsx\": \"ReactParser\",\n    \"tsx\": \"ReactTSParser\",\n    \"rs\": \"RustParser\",\n}",
-    "file_path": "kaizen/generator/unit_test.py",
-    "start_line": 24,
-    "end_line": 30,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 7
-  },
-  {
-    "category": "Error Handling",
-    "description": "Lack of error handling in file operations.",
-    "impact": "high",
-    "rationale": "The code currently does not handle potential exceptions that may arise from file operations, such as file not found or permission errors, which could lead to runtime failures.",
-    "recommendation": "Implement try-except blocks around file operations to gracefully handle errors.",
-    "current_code": "with open(file_path, \"r\") as file:\n    return file.read()",
-    "suggested_code": "try:\n    with open(file_path, \"r\") as file:\n        return file.read()\nexcept (FileNotFoundError, IOError) as e:\n    print(f\"Error reading file{file_path}:{e}\")",
-    "file_path": "kaizen/generator/unit_test.py",
-    "start_line": 108,
-    "end_line": 110,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Code Readability",
-    "description": "Inconsistent use of comments and docstrings.",
-    "impact": "medium",
-    "rationale": "Some methods lack docstrings, making it harder to understand their purpose and usage. Consistent documentation improves maintainability and usability.",
-    "recommendation": "Add docstrings to all methods to describe their purpose, parameters, and return values.",
-    "current_code": "def generate_tests(self, file_path: str, content: str = None, output_path: str = None):",
-    "suggested_code": "def generate_tests(self, file_path: str, content: str = None, output_path: str = None):\n    \"\"\"Generate tests based on the provided file path and content.\"\"\"",
-    "file_path": "kaizen/generator/unit_test.py",
-    "start_line": 75,
-    "end_line": 75,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Performance",
-    "description": "Potential inefficiency in reading file content.",
-    "impact": "medium",
-    "rationale": "The method `_read_file_content` reads the entire file into memory, which may not be efficient for large files.",
-    "recommendation": "Consider processing the file in chunks if large files are expected, or document the expected file sizes.",
-    "current_code": "return file.read()",
-    "suggested_code": "content =[]\nfor line in file:\n    content.append(line)\nreturn ''.join(content)",
-    "file_path": "kaizen/generator/unit_test.py",
-    "start_line": 109,
-    "end_line": 109,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Code Structure",
-    "description": "Function definition should be encapsulated properly.",
-    "impact": "medium",
-    "rationale": "The function `set_all_loggers_to_ERROR` is defined and called at the module level, which can lead to unexpected behavior if the module is imported elsewhere. This could also make unit testing more difficult.",
-    "recommendation": "Encapsulate the logger configuration within a main guard to prevent it from executing on import.",
-    "current_code": "set_all_loggers_to_ERROR()",
-    "suggested_code": "if __name__ == '__main__':\n    set_all_loggers_to_ERROR()",
-    "file_path": "kaizen/llms/provider.py",
-    "start_line": 23,
-    "end_line": 23,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Error Handling",
-    "description": "Lack of error handling when setting logger levels.",
-    "impact": "high",
-    "rationale": "If there are issues with setting the logger levels (e.g., if a logger does not exist), the current implementation will fail silently, which can make debugging difficult.",
-    "recommendation": "Add error handling to log any exceptions that occur when setting logger levels.",
-    "current_code": "logging.getLogger(name).setLevel(logging.ERROR)",
-    "suggested_code": "try:\n    logging.getLogger(name).setLevel(logging.ERROR)\nexcept Exception as e:\n    print(f'Error setting level for logger{name}:{e}')",
-    "file_path": "kaizen/llms/provider.py",
-    "start_line": 18,
-    "end_line": 18,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Readability",
-    "description": "Inconsistent naming conventions for logger names.",
-    "impact": "medium",
-    "rationale": "The logger names such as 'LiteLLM', 'LiteLLM Router', and 'LiteLLM Proxy' should follow a consistent naming convention for better readability and maintainability.",
-    "recommendation": "Standardize the naming convention for loggers, possibly using underscores or camel case consistently.",
-    "current_code": "logging.getLogger(\"LiteLLM\").setLevel(logging.ERROR)",
-    "suggested_code": "logging.getLogger(\"lite_llm\").setLevel(logging.ERROR)",
-    "file_path": "kaizen/llms/provider.py",
-    "start_line": 26,
-    "end_line": 28,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "11",
-    "end_line": "11",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/review.md b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/review.md
deleted file mode 100644
index 2e01a332..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_400/review.md
+++ /dev/null
@@ -1,56 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/400
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 23
-- Critical: 1
-- Important: 0
-- Minor: 0
-- Files Affected: 9
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Configuration (1 issues)</strong></summary>
-
-### 1. Changes made to sensitive file
-📁 **File:** `config.json:11`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 38388, "completion_tokens": 5366, "total_tokens": 43754}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/comments.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/comments.json
deleted file mode 100644
index 0637a088..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/comments.json
+++ /dev/null
@@ -1 +0,0 @@
-[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/issues.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/issues.json
deleted file mode 100644
index d8a0bb37..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/issues.json
+++ /dev/null
@@ -1,47 +0,0 @@
-[
-  {
-    "category": "Code Modification",
-    "description": "Redundant comment and potential for confusion with TODO.",
-    "impact": "medium",
-    "rationale": "The comment 'TODO: DONT PUSH DUPLICATE' lacks clarity and may lead to misunderstandings about its purpose. Comments should provide clear guidance or context.",
-    "recommendation": "Clarify the comment to specify what should not be duplicated and why. Consider removing it if it does not add value.",
-    "current_code": "# TODO: DONT PUSH DUPLICATE",
-    "suggested_code": "# TODO: Ensure unique embeddings are stored to avoid duplicates in the database.",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 157,
-    "end_line": 157,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Code Consistency",
-    "description": "Inconsistent use of comments regarding repository setup.",
-    "impact": "medium",
-    "rationale": "The comment above the line where `setup_repository` is called suggests that it should only be done during initial analysis or updates, but the line itself is always executed. This can lead to confusion.",
-    "recommendation": "Update the comment to reflect the actual behavior of the code, or consider removing it if it does not provide significant context.",
-    "current_code": "# Set up the repository (do this when you first analyze a repo or when you want to update it)",
-    "suggested_code": "# Set up the repository to ensure it is ready for analysis.",
-    "file_path": "examples/ragify_codebase/main.py",
-    "start_line": 6,
-    "end_line": 6,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 4
-  },
-  {
-    "category": "Dependency Update",
-    "description": "Updating dependencies without clear documentation.",
-    "impact": "high",
-    "rationale": "The removal of specific versions in favor of a newer version can lead to compatibility issues if the new version introduces breaking changes. It is crucial to document such changes clearly.",
-    "recommendation": "Add a note in the changelog or comments explaining why the dependency versions were updated and any potential impacts.",
-    "current_code": "llama-index-core = \"^0.10.47\"",
-    "suggested_code": "llama-index-core = \"0.10.65\"  # Updated for performance improvements and bug fixes.",
-    "file_path": "pyproject.toml",
-    "start_line": 27,
-    "end_line": 27,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 6
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/review.md b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/review.md
deleted file mode 100644
index d14252ac..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_440/review.md
+++ /dev/null
@@ -1,33 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/440
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 3
-- Critical: 0
-- Important: 0
-- Minor: 0
-- Files Affected: 3
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 1613, "completion_tokens": 626, "total_tokens": 2239}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/comments.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/comments.json
deleted file mode 100644
index 507ce250..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/comments.json
+++ /dev/null
@@ -1,31 +0,0 @@
-[
-  {
-    "category": "Security Vulnerability",
-    "description": "Potential exposure of sensitive data in logs.",
-    "impact": "critical",
-    "rationale": "Logging sensitive information, such as access tokens or personal data, can lead to security vulnerabilities. It's essential to sanitize logs to avoid exposing sensitive data.",
-    "recommendation": "Review logging statements to ensure sensitive data is not logged. Implement logging best practices.",
-    "current_code": "logger.debug(f\"Post Review comment response:{response.text}\")",
-    "suggested_code": "logger.debug(\"Post Review comment response received.\")",
-    "file_path": "github_app/github_helper/pull_requests.py",
-    "start_line": 181,
-    "end_line": 182,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 9
-  },
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/issues.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/issues.json
deleted file mode 100644
index 1364b6ee..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/issues.json
+++ /dev/null
@@ -1,91 +0,0 @@
-[
-  {
-    "category": "Error Handling",
-    "description": "Broad exception handling without specific error management.",
-    "impact": "high",
-    "rationale": "Using a generic 'except Exception' clause can mask underlying issues and make debugging difficult. It is better to handle specific exceptions to provide clearer error messages and control flow.",
-    "recommendation": "Refine the exception handling to catch specific exceptions and log meaningful error messages.",
-    "current_code": "except Exception:\n    print(\"Error\")",
-    "suggested_code": "except KeyError:\n    print(f\"Key error occurred with min_confidence:{min_confidence}\")",
-    "file_path": "github_app/github_helper/pull_requests.py",
-    "start_line": 140,
-    "end_line": 141,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Code Readability",
-    "description": "Inconsistent naming conventions for variables.",
-    "impact": "medium",
-    "rationale": "Using inconsistent naming conventions can lead to confusion and reduce code readability. It's important to maintain a consistent style throughout the codebase.",
-    "recommendation": "Adopt a consistent naming convention (e.g., snake_case for variables) and ensure all variables follow this convention.",
-    "current_code": "repo_name, pull_request_files=pr_files",
-    "suggested_code": "repo_name, pull_request_files=pr_files",
-    "file_path": "github_app/github_helper/pull_requests.py",
-    "start_line": 55,
-    "end_line": 56,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Performance Optimization",
-    "description": "Inefficient sorting algorithm in sort_files function.",
-    "impact": "high",
-    "rationale": "The current sorting method is O(n^2) due to nested loops. Using Python's built-in sorting functions can significantly improve performance.",
-    "recommendation": "Utilize Python's built-in sorting capabilities to improve performance.",
-    "current_code": "for i, sorted_file in enumerate(sorted_files):\n    if file_name < sorted_file[\"filename\"]:\n        min_index = i\n        break",
-    "suggested_code": "sorted_files = sorted(files, key=lambda x: x['filename'])",
-    "file_path": "github_app/github_helper/pull_requests.py",
-    "start_line": 184,
-    "end_line": 194,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Documentation",
-    "description": "Lack of comments and documentation for newly added functions.",
-    "impact": "medium",
-    "rationale": "New functions like 'generate_tests' lack docstrings, making it difficult for other developers to understand their purpose and usage.",
-    "recommendation": "Add docstrings to all new functions to explain their purpose, parameters, and return values.",
-    "current_code": "def generate_tests(pr_files):\n    return[f[\"filename\"] for f in pr_files]",
-    "suggested_code": "def generate_tests(pr_files):\n    \"\"\"Generate a list of test filenames from PR files.\"\n    return[f[\"filename\"] for f in pr_files]",
-    "file_path": "github_app/github_helper/pull_requests.py",
-    "start_line": 199,
-    "end_line": 200,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 4
-  },
-  {
-    "category": "Security Vulnerability",
-    "description": "Potential exposure of sensitive data in logs.",
-    "impact": "critical",
-    "rationale": "Logging sensitive information, such as access tokens or personal data, can lead to security vulnerabilities. It's essential to sanitize logs to avoid exposing sensitive data.",
-    "recommendation": "Review logging statements to ensure sensitive data is not logged. Implement logging best practices.",
-    "current_code": "logger.debug(f\"Post Review comment response:{response.text}\")",
-    "suggested_code": "logger.debug(\"Post Review comment response received.\")",
-    "file_path": "github_app/github_helper/pull_requests.py",
-    "start_line": 181,
-    "end_line": 182,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 9
-  },
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/review.md b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/review.md
deleted file mode 100644
index d25b577b..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_476/review.md
+++ /dev/null
@@ -1,72 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/476
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 6
-- Critical: 2
-- Important: 0
-- Minor: 0
-- Files Affected: 2
-## 🏆 Code Quality
-[███████████████░░░░░] 75% (Fair)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Security Vulnerability (2 issues)</strong></summary>
-
-### 1. Potential exposure of sensitive data in logs.
-📁 **File:** `github_app/github_helper/pull_requests.py:181`
-⚖️ **Severity:** 9/10
-🔍 **Description:** Potential exposure of sensitive data in logs.
-💡 **Solution:** 
-
-**Current Code:**
-```python
-logger.debug(f"Post Review comment response:{response.text}")
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 2. Changes made to sensitive file
-📁 **File:** `config.json:1`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 4256, "completion_tokens": 998, "total_tokens": 5254}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/comments.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/comments.json
deleted file mode 100644
index b3f4d9f3..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/comments.json
+++ /dev/null
@@ -1,17 +0,0 @@
-[
-  {
-    "category": "Division by Zero Risk",
-    "description": "Potential division by zero if total_tokens is zero.",
-    "impact": "critical",
-    "rationale": "Dividing by zero will cause a runtime error, leading to application crashes.",
-    "recommendation": "Add a check to ensure total_tokens is not zero before performing division.",
-    "current_code": "print(f\"Total tokens used:{total_tokens:,}\")",
-    "suggested_code": "if total_tokens > 0: print(f\"Total tokens used:{total_tokens:,}\")",
-    "file_path": "main.py",
-    "start_line": 159,
-    "end_line": 159,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 9
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/issues.json b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/issues.json
deleted file mode 100644
index 3f2cd28f..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/issues.json
+++ /dev/null
@@ -1,107 +0,0 @@
-[
-  {
-    "category": "Unused Import",
-    "description": "Unused import statement for 'random'.",
-    "impact": "low",
-    "rationale": "Having unused imports can lead to confusion and clutter in the codebase.",
-    "recommendation": "Remove the unused import statement.",
-    "current_code": "import random  # Unused import",
-    "suggested_code": "",
-    "file_path": "main.py",
-    "start_line": 8,
-    "end_line": 8,
-    "change_type": "deletion",
-    "sentiment": "neutral",
-    "severity": 2
-  },
-  {
-    "category": "Error Handling",
-    "description": "Potential failure of API call without retry mechanism.",
-    "impact": "high",
-    "rationale": "If the API call fails, the application may crash or return unexpected results without a retry mechanism.",
-    "recommendation": "Implement a retry mechanism for the API call.",
-    "current_code": "response = completion(",
-    "suggested_code": "# Implement retry logic here",
-    "file_path": "main.py",
-    "start_line": 66,
-    "end_line": 66,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Silent Failure",
-    "description": "Silent failure in JSON parsing without logging.",
-    "impact": "high",
-    "rationale": "Failing to log errors can make debugging difficult and lead to unnoticed issues.",
-    "recommendation": "Add logging for the exception to capture the error details.",
-    "current_code": "print(f\"Failed to parse content for applicant\")",
-    "suggested_code": "print(f\"Failed to parse content for applicant:{e}\")",
-    "file_path": "main.py",
-    "start_line": 86,
-    "end_line": 86,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Inefficient Progress Reporting",
-    "description": "Inefficient way to print progress during applicant processing.",
-    "impact": "medium",
-    "rationale": "Using print statements in a loop can lead to performance issues and cluttered output.",
-    "recommendation": "Consider using a progress bar or logging framework for better performance and readability.",
-    "current_code": "print(f\"\\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}\", end='', flush=True)",
-    "suggested_code": "# Use a logging framework or a proper progress bar",
-    "file_path": "main.py",
-    "start_line": 121,
-    "end_line": 121,
-    "change_type": "modification",
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Redundant Code",
-    "description": "Redundant check for empty DataFrame before processing.",
-    "impact": "low",
-    "rationale": "The check for an empty DataFrame is unnecessary since processing will naturally handle it.",
-    "recommendation": "Remove the redundant check for empty DataFrame.",
-    "current_code": "if len(df) == 0: return",
-    "suggested_code": "",
-    "file_path": "main.py",
-    "start_line": 142,
-    "end_line": 143,
-    "change_type": "deletion",
-    "sentiment": "neutral",
-    "severity": 3
-  },
-  {
-    "category": "Division by Zero Risk",
-    "description": "Potential division by zero if total_tokens is zero.",
-    "impact": "critical",
-    "rationale": "Dividing by zero will cause a runtime error, leading to application crashes.",
-    "recommendation": "Add a check to ensure total_tokens is not zero before performing division.",
-    "current_code": "print(f\"Total tokens used:{total_tokens:,}\")",
-    "suggested_code": "if total_tokens > 0: print(f\"Total tokens used:{total_tokens:,}\")",
-    "file_path": "main.py",
-    "start_line": 159,
-    "end_line": 159,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 9
-  },
-  {
-    "category": "File Handling",
-    "description": "No error handling for file not found when running main function.",
-    "impact": "high",
-    "rationale": "If the file does not exist, the application will crash without providing user feedback.",
-    "recommendation": "Add error handling to manage file not found exceptions.",
-    "current_code": "main(input_file)",
-    "suggested_code": "try: main(input_file) except FileNotFoundError: print('File not found. Please check the file name.')",
-    "file_path": "main.py",
-    "start_line": 175,
-    "end_line": 175,
-    "change_type": "modification",
-    "sentiment": "negative",
-    "severity": 8
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/review.md b/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/review.md
deleted file mode 100644
index 6bd79ffb..00000000
--- a/.experiments/code_review/code_reviews_20240827_163259/no_eval/pr_5/review.md
+++ /dev/null
@@ -1,64 +0,0 @@
-PR URL: https://github.com/sauravpanda/applicant-screening/pull/5
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 7
-- Critical: 1
-- Important: 0
-- Minor: 0
-- Files Affected: 1
-## 🏆 Code Quality
-[███████████████░░░░░] 75% (Fair)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Division by Zero Risk (1 issues)</strong></summary>
-
-### 1. Potential division by zero if total_tokens is zero.
-📁 **File:** `main.py:159`
-⚖️ **Severity:** 9/10
-🔍 **Description:** Potential division by zero if total_tokens is zero.
-💡 **Solution:** 
-
-**Current Code:**
-```python
-print(f"Total tokens used:{total_tokens:,}")
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (2 issues)</strong></summary>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 6403, "completion_tokens": 1126, "total_tokens": 7529}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/comments.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/comments.json
deleted file mode 100644
index 6f3e44d6..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/comments.json
+++ /dev/null
@@ -1,100 +0,0 @@
-[
-  {
-    "category": "Security",
-    "description": "Potential security risk with hardcoded API keys.",
-    "impact": "critical",
-    "rationale": "Hardcoding sensitive information like API keys can lead to security vulnerabilities if the code is exposed. It is essential to use environment variables or secure vaults.",
-    "recommendation": "Replace hardcoded API keys with environment variables and ensure they are not included in version control.",
-    "current_code": "\"api_key\": \"os.environ/AZURE_API_KEY\",",
-    "suggested_code": "\"api_key\": os.getenv('AZURE_API_KEY'),",
-    "file_path": "config.json",
-    "start_line": 13,
-    "end_line": 13,
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Security",
-    "description": "Potential SQL injection risk in dynamic queries.",
-    "impact": "critical",
-    "rationale": "The use of string interpolation in SQL queries can expose the application to SQL injection attacks. Although parameterized queries are used in some places, there are instances where string interpolation is still present.",
-    "recommendation": "Always use parameterized queries instead of string interpolation to prevent SQL injection vulnerabilities.",
-    "current_code": "query = text(f\"SELECT ... WHERE node_content LIKE '%{caller}%' \")",
-    "suggested_code": "query = text(\"SELECT ... WHERE node_content LIKE :caller\").bindparams(caller=f'%{caller}%')",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 303,
-    "end_line": 303,
-    "sentiment": "negative",
-    "severity": 9
-  },
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Docker",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to Dockerfile, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_path": "Dockerfile",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Docker",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to docker-compose.yml, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "15",
-    "end_line": "15",
-    "side": "RIGHT",
-    "file_path": "docker-compose.yml",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Version Control",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to .gitignore, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "164",
-    "end_line": "164",
-    "side": "RIGHT",
-    "file_path": ".gitignore",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Database",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to db_setup/init.sql, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_path": "db_setup/init.sql",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/issues.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/issues.json
deleted file mode 100644
index 1a938a3d..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/issues.json
+++ /dev/null
@@ -1,282 +0,0 @@
-[
-  {
-    "category": "Documentation",
-    "description": "Lack of comments and documentation in several files.",
-    "impact": "high",
-    "rationale": "While the code is functional, it lacks sufficient comments and documentation, which can make it difficult for other developers to understand the purpose and functionality of various components.",
-    "recommendation": "Add comments to explain the purpose of functions, classes, and complex logic. Consider using docstrings for public methods and classes.",
-    "current_code": "def chunk_code(code: str, language: str) -> ParsedBody:",
-    "suggested_code": "def chunk_code(code: str, language: str) -> ParsedBody:\n    \"\"\"\n    Chunks the provided code into functions, classes, and other blocks based on the specified language.\n    \"\"\"",
-    "file_path": "kaizen/retriever/code_chunker.py",
-    "start_line": 7,
-    "end_line": 8,
-    "sentiment": "neutral",
-    "severity": 7
-  },
-  {
-    "category": "Error Handling",
-    "description": "Insufficient error handling in database interactions.",
-    "impact": "high",
-    "rationale": "The code interacts with the database without any error handling, which could lead to unhandled exceptions and application crashes.",
-    "recommendation": "Implement try-except blocks around database operations to handle potential exceptions gracefully.",
-    "current_code": "cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))",
-    "suggested_code": "try:\n    cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))\nexcept Exception as e:\n    # Log the error and handle it appropriately\n    print(f\"Database error:{e}\")",
-    "file_path": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 41,
-    "end_line": 41,
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Security",
-    "description": "Potential security risk with hardcoded API keys.",
-    "impact": "critical",
-    "rationale": "Hardcoding sensitive information like API keys can lead to security vulnerabilities if the code is exposed. It is essential to use environment variables or secure vaults.",
-    "recommendation": "Replace hardcoded API keys with environment variables and ensure they are not included in version control.",
-    "current_code": "\"api_key\": \"os.environ/AZURE_API_KEY\",",
-    "suggested_code": "\"api_key\": os.getenv('AZURE_API_KEY'),",
-    "file_path": "config.json",
-    "start_line": 13,
-    "end_line": 13,
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Performance",
-    "description": "Inefficient handling of large code chunks.",
-    "impact": "medium",
-    "rationale": "The current implementation may not efficiently handle large code inputs, leading to performance degradation.",
-    "recommendation": "Consider implementing a streaming approach or chunking the input code to process it in smaller parts.",
-    "current_code": "def chunk_code(code: str, language: str) -> ParsedBody:",
-    "suggested_code": "# Consider breaking down the input code into smaller chunks before processing.",
-    "file_path": "kaizen/retriever/code_chunker.py",
-    "start_line": 7,
-    "end_line": 7,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Code Structure",
-    "description": "Inconsistent naming conventions.",
-    "impact": "medium",
-    "rationale": "Inconsistent naming can lead to confusion and reduce code readability. Following a consistent naming convention improves maintainability.",
-    "recommendation": "Standardize naming conventions across the codebase (e.g., snake_case for variables and functions).",
-    "current_code": "def is_react_hook(name: str) -> bool:",
-    "suggested_code": "def is_react_hook(name: str) -> bool:",
-    "file_path": "kaizen/retriever/code_chunker.py",
-    "start_line": 65,
-    "end_line": 66,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling",
-    "description": "Lack of specific error handling in critical database operations.",
-    "impact": "high",
-    "rationale": "The code currently raises a generic exception in the `generate_abstraction` method without providing context or handling specific error types. This can lead to unhandled exceptions that crash the application.",
-    "recommendation": "Implement specific exception handling to provide more informative error messages and handle different types of exceptions appropriately.",
-    "current_code": "except Exception as e:\n    raise e",
-    "suggested_code": "except (ValueError, TypeError) as e:\n    logger.error(f\"Error generating abstraction:{str(e)}\")\n    raise CustomAbstractionError(f\"Failed to generate abstraction for code block:{code_block}\")",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 218,
-    "end_line": 219,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Code Readability",
-    "description": "Inconsistent logging levels and messages.",
-    "impact": "medium",
-    "rationale": "Using different logging levels inconsistently can make it difficult to track the application's behavior. For example, using `logger.debug` for important events like initialization can obscure critical information.",
-    "recommendation": "Use appropriate logging levels consistently. For instance, use `logger.info` for initialization messages and `logger.debug` for detailed debugging information.",
-    "current_code": "logger.debug(f\"Successfully parsed file:{file_path}\")",
-    "suggested_code": "logger.info(f\"Successfully parsed file:{file_path}\")",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 107,
-    "end_line": 107,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Performance",
-    "description": "Potential performance issues with database operations.",
-    "impact": "high",
-    "rationale": "The current implementation performs multiple database queries within a loop, which can lead to performance bottlenecks, especially with a large number of files or functions.",
-    "recommendation": "Batch database operations where possible to reduce the number of queries. Consider using bulk inserts or updates to improve performance.",
-    "current_code": "connection.execute(query,{...})",
-    "suggested_code": "with connection.begin() as transaction:\n    # Collect all queries and execute them in a batch\n    transaction.execute(...)",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 298,
-    "end_line": 312,
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Security",
-    "description": "Potential SQL injection risk in dynamic queries.",
-    "impact": "critical",
-    "rationale": "The use of string interpolation in SQL queries can expose the application to SQL injection attacks. Although parameterized queries are used in some places, there are instances where string interpolation is still present.",
-    "recommendation": "Always use parameterized queries instead of string interpolation to prevent SQL injection vulnerabilities.",
-    "current_code": "query = text(f\"SELECT ... WHERE node_content LIKE '%{caller}%' \")",
-    "suggested_code": "query = text(\"SELECT ... WHERE node_content LIKE :caller\").bindparams(caller=f'%{caller}%')",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 303,
-    "end_line": 303,
-    "sentiment": "negative",
-    "severity": 9
-  },
-  {
-    "category": "Maintainability",
-    "description": "Hardcoded values for database connection parameters.",
-    "impact": "medium",
-    "rationale": "Hardcoding values like `embed_dim=1536` makes the code less flexible and harder to maintain. If these values need to change, it requires code modification.",
-    "recommendation": "Consider using configuration files or environment variables to manage such parameters, making the code more adaptable.",
-    "current_code": "embed_dim=1536",
-    "suggested_code": "embed_dim=int(os.getenv('EMBED_DIM', 1536))",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 49,
-    "end_line": 49,
-    "sentiment": "neutral",
-    "severity": 4
-  },
-  {
-    "category": "Error Handling",
-    "description": "Broad Exception Catching",
-    "impact": "high",
-    "rationale": "Catching general exceptions can obscure the root cause of errors and make debugging difficult. Specific exceptions should be caught to provide clearer error handling.",
-    "recommendation": "Catch specific exceptions instead of using a general 'Exception'. This will help in identifying the exact issue and provide more meaningful error messages.",
-    "current_code": "except Exception as e:",
-    "suggested_code": "except (ImportError, ValueError) as e:",
-    "file_path": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 28,
-    "end_line": 30,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Logging",
-    "description": "Inconsistent Logging Levels",
-    "impact": "medium",
-    "rationale": "Using different logging levels (error, warning, info) inconsistently can lead to confusion about the severity of issues. It's important to use appropriate logging levels for different situations.",
-    "recommendation": "Use logging levels consistently. For example, use 'logger.warning' for recoverable issues and 'logger.error' for critical failures.",
-    "current_code": "logger.error(f\"Failed to load language{language}:{str(e)}\")",
-    "suggested_code": "logger.warning(f\"Failed to load language{language}:{str(e)}\")",
-    "file_path": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 108,
-    "end_line": 108,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Code Readability",
-    "description": "Complex Conditional Logic",
-    "impact": "medium",
-    "rationale": "The conditional logic in the `traverse_tree` function is complex and could be simplified for better readability and maintainability.",
-    "recommendation": "Consider refactoring the conditional logic into separate functions or using a dictionary mapping to improve clarity.",
-    "current_code": "if node.type in[\"function_definition\", \"function_declaration\", ...]:",
-    "suggested_code": "def is_function(node): return node.type in[\"function_definition\", \"function_declaration\", ...]\n\nif is_function(node):",
-    "file_path": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 48,
-    "end_line": 88,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Performance",
-    "description": "Potential Memory Leak with Caching",
-    "impact": "medium",
-    "rationale": "Using `lru_cache` without a size limit can lead to excessive memory usage if many different languages are loaded. This could affect performance over time.",
-    "recommendation": "Consider setting a reasonable `maxsize` for `lru_cache` to prevent potential memory issues.",
-    "current_code": "@lru_cache(maxsize=None)",
-    "suggested_code": "@lru_cache(maxsize=128)",
-    "file_path": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 14,
-    "end_line": 14,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Unit Testing",
-    "description": "Lack of Unit Tests for New Functionality",
-    "impact": "high",
-    "rationale": "The new functionalities introduced lack corresponding unit tests, which are crucial for verifying correctness and preventing regressions.",
-    "recommendation": "Add unit tests for the new functions, particularly for `load_language` and `get_parser`, to ensure they behave as expected.",
-    "current_code": "No tests present for new functionality.",
-    "suggested_code": "def test_load_language():\n    assert load_language('tree-sitter-python') is not None",
-    "file_path": "tests/retriever/test_chunker.py",
-    "start_line": 1,
-    "end_line": 1,
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Docker",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to Dockerfile, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_path": "Dockerfile",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Docker",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to docker-compose.yml, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "15",
-    "end_line": "15",
-    "side": "RIGHT",
-    "file_path": "docker-compose.yml",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Version Control",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to .gitignore, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "164",
-    "end_line": "164",
-    "side": "RIGHT",
-    "file_path": ".gitignore",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Database",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to db_setup/init.sql, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_path": "db_setup/init.sql",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/review.md b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/review.md
deleted file mode 100644
index 37e9cbc5..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_222/review.md
+++ /dev/null
@@ -1,152 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/222
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 20
-- Critical: 7
-- Important: 0
-- Minor: 0
-- Files Affected: 10
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Security (7 issues)</strong></summary>
-
-### 1. Potential security risk with hardcoded API keys.
-📁 **File:** `config.json:13`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Potential security risk with hardcoded API keys.
-💡 **Solution:** 
-
-**Current Code:**
-```python
-"api_key": "os.environ/AZURE_API_KEY",
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 2. Potential SQL injection risk in dynamic queries.
-📁 **File:** `kaizen/retriever/llama_index_retriever.py:303`
-⚖️ **Severity:** 9/10
-🔍 **Description:** Potential SQL injection risk in dynamic queries.
-💡 **Solution:** 
-
-**Current Code:**
-```python
-query = text(f"SELECT ... WHERE node_content LIKE '%{caller}%' ")
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 3. Changes made to sensitive file
-📁 **File:** `config.json:4`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 4. Changes made to sensitive file
-📁 **File:** `Dockerfile:4`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 5. Changes made to sensitive file
-📁 **File:** `docker-compose.yml:15`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 6. Changes made to sensitive file
-📁 **File:** `.gitignore:164`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 7. Changes made to sensitive file
-📁 **File:** `db_setup/init.sql:1`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 19847, "completion_tokens": 2900, "total_tokens": 22747}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/comments.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/comments.json
deleted file mode 100644
index 0637a088..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/comments.json
+++ /dev/null
@@ -1 +0,0 @@
-[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/issues.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/issues.json
deleted file mode 100644
index 21f1de13..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/issues.json
+++ /dev/null
@@ -1,198 +0,0 @@
-[
-  {
-    "category": "Code Clarity",
-    "description": "Redundant imports and unused variables.",
-    "impact": "medium",
-    "rationale": "Removing unused imports and variables enhances code readability and reduces confusion for future maintainers. It also slightly improves performance by reducing the bundle size.",
-    "recommendation": "Audit imports and remove any that are not used in the file.",
-    "current_code": "import{MemoriesIcon, NextIcon, SearchIcon, UrlIcon}from \"@repo/ui/icons\";",
-    "suggested_code": "import{MemoriesIcon, NextIcon, UrlIcon}from \"@repo/ui/icons\";",
-    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 4,
-    "end_line": 4,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling",
-    "description": "Lack of error handling in async functions.",
-    "impact": "high",
-    "rationale": "The `handleDeleteSpace` function does not handle potential errors from the `deleteSpace` call, which could lead to unhandled promise rejections and a poor user experience.",
-    "recommendation": "Add a try-catch block around the async call to handle errors gracefully.",
-    "current_code": "const handleDeleteSpace = async (id: number) =>{const response = await deleteSpace(id); ...};",
-    "suggested_code": "const handleDeleteSpace = async (id: number) =>{try{const response = await deleteSpace(id); ...}catch (error){toast.error(\"An error occurred while deleting the space\");}};",
-    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 66,
-    "end_line": 76,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Performance",
-    "description": "Inefficient state updates in `handleDeleteSpace`.",
-    "impact": "medium",
-    "rationale": "Using `spaces.filter` creates a new array on every delete, which can be inefficient. Consider using functional updates to update state based on the previous state.",
-    "recommendation": "Use the functional form of `setSpaces` to avoid unnecessary re-renders.",
-    "current_code": "setSpaces(spaces.filter((space) => space.id !== id));",
-    "suggested_code": "setSpaces((prevSpaces) => prevSpaces.filter((space) => space.id !== id));",
-    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 70,
-    "end_line": 70,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Code Consistency",
-    "description": "Inconsistent naming conventions for components.",
-    "impact": "medium",
-    "rationale": "Inconsistent naming can lead to confusion. For instance, `LinkComponent` and `MemoryComponent` should follow a consistent naming pattern.",
-    "recommendation": "Rename components to follow a consistent naming convention, such as `MemoryLink` and `SpaceLink`.",
-    "current_code": "function LinkComponent({...}){...}",
-    "suggested_code": "function MemoryLink({...}){...}",
-    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 321,
-    "end_line": 321,
-    "sentiment": "neutral",
-    "severity": 4
-  },
-  {
-    "category": "User Experience",
-    "description": "Lack of feedback for loading states.",
-    "impact": "high",
-    "rationale": "Users should receive feedback when actions are being processed, such as during the deletion of a space. This improves user experience and prevents confusion.",
-    "recommendation": "Implement a loading state for actions that involve network requests.",
-    "current_code": "const handleDeleteSpace = async (id: number) =>{...};",
-    "suggested_code": "const[loading, setLoading] = useState(false); const handleDeleteSpace = async (id: number) =>{setLoading(true); ... setLoading(false);};",
-    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 66,
-    "end_line": 76,
-    "sentiment": "negative",
-    "severity": 6
-  },
-  {
-    "category": "Code Structure",
-    "description": "Excessive nesting in the Menu component.",
-    "impact": "high",
-    "rationale": "Deeply nested components can make the code harder to read and maintain. It can also lead to performance issues if not managed properly.",
-    "recommendation": "Consider breaking down the Menu component into smaller, reusable components to improve readability and maintainability.",
-    "current_code": "function Menu(){...}",
-    "suggested_code": "function Menu(){return (<DialogContentContainer setDialogClose={() => setDialogOpen(false)}/>);}",
-    "file_path": "apps/web/app/(dash)/menu.tsx",
-    "start_line": 163,
-    "end_line": 384,
-    "sentiment": "neutral",
-    "severity": 7
-  },
-  {
-    "category": "Error Handling",
-    "description": "Error handling in async functions is insufficient.",
-    "impact": "high",
-    "rationale": "Errors in asynchronous operations are not being caught, which could lead to unhandled promise rejections and application crashes.",
-    "recommendation": "Wrap async calls in try-catch blocks to handle errors gracefully.",
-    "current_code": "const handleSubmit = async (content?: string, spaces?: number[]) =>{...};",
-    "suggested_code": "const handleSubmit = async (content?: string, spaces?: number[]) =>{try{...}catch (error){toast.error(`Error: ${error.message}`);}};",
-    "file_path": "apps/web/app/(dash)/menu.tsx",
-    "start_line": 213,
-    "end_line": 230,
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Performance",
-    "description": "Potential performance issues with useEffect dependency.",
-    "impact": "medium",
-    "rationale": "The useEffect hook does not specify dependencies, which may lead to unnecessary re-renders and API calls.",
-    "recommendation": "Add dependencies to the useEffect hook to control when it runs.",
-    "current_code": "useEffect(() =>{...},[]);",
-    "suggested_code": "useEffect(() =>{...},[setSpaces]);",
-    "file_path": "apps/web/app/(dash)/menu.tsx",
-    "start_line": 38,
-    "end_line": 51,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Readability",
-    "description": "Inconsistent naming conventions.",
-    "impact": "medium",
-    "rationale": "Inconsistent naming can lead to confusion and reduce code readability. For instance, the variable 'spaces' is used both as state and in the function.",
-    "recommendation": "Use more descriptive names for variables and functions to improve clarity.",
-    "current_code": "let spaces = await getSpaces();",
-    "suggested_code": "let fetchedSpaces = await getSpaces();",
-    "file_path": "apps/web/app/(dash)/menu.tsx",
-    "start_line": 40,
-    "end_line": 40,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Security",
-    "description": "Potential security vulnerability with user input.",
-    "impact": "high",
-    "rationale": "User input is not being sanitized before being processed, which could lead to XSS attacks or other vulnerabilities.",
-    "recommendation": "Sanitize user input before processing it to prevent security vulnerabilities.",
-    "current_code": "const content = e.get('content')?.toString();",
-    "suggested_code": "const content = sanitizeInput(e.get('content')?.toString());",
-    "file_path": "apps/web/app/(dash)/menu.tsx",
-    "start_line": 238,
-    "end_line": 238,
-    "sentiment": "negative",
-    "severity": 9
-  },
-  {
-    "category": "Code Readability",
-    "description": "Inconsistent use of destructuring in function parameters.",
-    "impact": "medium",
-    "rationale": "The function `ComboboxWithCreate` uses destructuring for some props but not for others, which can lead to confusion and inconsistency in the codebase.",
-    "recommendation": "Consistently use destructuring for all props in the function parameters.",
-    "current_code": "const ComboboxWithCreate = ({options, onSelect, onSubmit, selectedSpaces, setSelectedSpaces}: ComboboxWithCreateProps) =>{",
-    "suggested_code": "const ComboboxWithCreate: React.FC<ComboboxWithCreateProps> = ({options, onSelect, onSubmit, selectedSpaces, setSelectedSpaces}) =>{",
-    "file_path": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 32,
-    "end_line": 38,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling",
-    "description": "Potential for undefined values in `options.find`.",
-    "impact": "high",
-    "rationale": "Using `?.` operator without a fallback can lead to rendering issues if the option is not found. This could result in a runtime error if the component expects a valid label.",
-    "recommendation": "Provide a fallback value in case the option is not found.",
-    "current_code": "options.find((opt) => opt.value === spaceId.toString())?.label",
-    "suggested_code": "options.find((opt) => opt.value === spaceId.toString())?.label || 'Unknown Space'",
-    "file_path": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 74,
-    "end_line": 74,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Performance",
-    "description": "Unnecessary re-renders due to state updates in `handleKeyDown`.",
-    "impact": "medium",
-    "rationale": "The `handleKeyDown` function updates state on every key press, which can lead to performance issues if the component re-renders frequently.",
-    "recommendation": "Consider using a debouncing technique to limit the frequency of state updates.",
-    "current_code": "setSelectedSpaces((prev) => prev.slice(0, -1));",
-    "suggested_code": "// Implement debouncing logic here",
-    "file_path": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 51,
-    "end_line": 51,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Code Maintainability",
-    "description": "Hardcoded strings in JSX can lead to maintenance issues.",
-    "impact": "medium",
-    "rationale": "Hardcoded strings can make localization and future changes more difficult. It is better to define them as constants or use a localization library.",
-    "recommendation": "Define hardcoded strings as constants at the top of the file.",
-    "current_code": "placeholder=\"Select or create a new space.\"",
-    "suggested_code": "const PLACEHOLDER_TEXT = 'Select or create a new space.'; // Use this constant in JSX",
-    "file_path": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 82,
-    "end_line": 82,
-    "sentiment": "neutral",
-    "severity": 4
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/review.md b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/review.md
deleted file mode 100644
index 12226340..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_232/review.md
+++ /dev/null
@@ -1,33 +0,0 @@
-PR URL: https://github.com/supermemoryai/supermemory/pull/232
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 14
-- Critical: 0
-- Important: 0
-- Minor: 0
-- Files Affected: 3
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 14992, "completion_tokens": 2670, "total_tokens": 17662}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/comments.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/comments.json
deleted file mode 100644
index 0637a088..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/comments.json
+++ /dev/null
@@ -1 +0,0 @@
-[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/issues.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/issues.json
deleted file mode 100644
index b1b25dc2..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/issues.json
+++ /dev/null
@@ -1,58 +0,0 @@
-[
-  {
-    "category": "Code Readability",
-    "description": "Inconsistent formatting in function calls.",
-    "impact": "medium",
-    "rationale": "Consistency in function call formatting enhances readability and maintainability. The change to break the function call into multiple lines improves clarity but should be applied uniformly across the codebase.",
-    "recommendation": "Consider applying consistent formatting for function calls throughout the codebase.",
-    "current_code": "linkedin_post = work_summary_generator.generate_linkedin_post(summary, user=\"oss_example\")",
-    "suggested_code": "linkedin_post = work_summary_generator.generate_linkedin_post(\n    summary, user=\"oss_example\"\n)",
-    "file_path": "examples/work_summarizer/main.py",
-    "start_line": 60,
-    "end_line": 61,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling",
-    "description": "Lack of error handling in JSON parsing.",
-    "impact": "high",
-    "rationale": "The code does not handle potential exceptions that may arise from JSON parsing, which could lead to runtime errors and application crashes.",
-    "recommendation": "Wrap the JSON parsing code in a try-except block to handle potential exceptions gracefully.",
-    "current_code": "parsed_data = json.loads(json_data)",
-    "suggested_code": "try:\n    parsed_data = json.loads(json_data)\nexcept json.JSONDecodeError as e:\n    logging.error(f\"JSON decoding failed:{e}\")\n    return{}",
-    "file_path": "kaizen/helpers/parser.py",
-    "start_line": 47,
-    "end_line": 48,
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Code Efficiency",
-    "description": "Redundant calls to extract markdown content.",
-    "impact": "medium",
-    "rationale": "The extraction of markdown content is done twice in the new methods, which can lead to code duplication and potential inconsistencies.",
-    "recommendation": "Consider refactoring the markdown extraction into a separate utility function to avoid code duplication.",
-    "current_code": "return parser.extract_markdown_content(response)",
-    "suggested_code": "return extract_markdown_content(response)",
-    "file_path": "kaizen/reviewer/work_summarizer.py",
-    "start_line": 65,
-    "end_line": 65,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Documentation",
-    "description": "Missing docstrings for new methods.",
-    "impact": "medium",
-    "rationale": "Newly added methods lack docstrings, making it difficult for other developers to understand their purpose and usage.",
-    "recommendation": "Add docstrings to the new methods to describe their functionality and parameters.",
-    "current_code": "def generate_twitter_post(self, summary: Dict, user: Optional[str] = None) -> str:",
-    "suggested_code": "def generate_twitter_post(self, summary: Dict, user: Optional[str] = None) -> str:\n    \"\"\"Generate a Twitter post based on the given summary.\"\"\"",
-    "file_path": "kaizen/reviewer/work_summarizer.py",
-    "start_line": 58,
-    "end_line": 58,
-    "sentiment": "neutral",
-    "severity": 4
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/review.md b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/review.md
deleted file mode 100644
index 63a8b589..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_252/review.md
+++ /dev/null
@@ -1,33 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/252
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 4
-- Critical: 0
-- Important: 0
-- Minor: 0
-- Files Affected: 3
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 3585, "completion_tokens": 758, "total_tokens": 4343}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/comments.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/comments.json
deleted file mode 100644
index 0637a088..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/comments.json
+++ /dev/null
@@ -1 +0,0 @@
-[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/issues.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/issues.json
deleted file mode 100644
index 0935869f..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/issues.json
+++ /dev/null
@@ -1,72 +0,0 @@
-[
-  {
-    "category": "Code Consistency",
-    "description": "Inconsistent naming conventions for prompts.",
-    "impact": "medium",
-    "rationale": "The naming of the new prompts does not follow the established naming conventions seen in the previous code, which can lead to confusion and inconsistency.",
-    "recommendation": "Ensure that all prompt variables follow a consistent naming pattern (e.g., using underscores for separation).",
-    "current_code": "PR_DESCRIPTION_SYSTEM_PROMPT",
-    "suggested_code": "PR_DESC_SYSTEM_PROMPT",
-    "file_path": "kaizen/llms/prompts/pr_desc_prompts.py",
-    "start_line": 1,
-    "end_line": 3,
-    "sentiment": "negative",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling",
-    "description": "Lack of error handling in prompt processing.",
-    "impact": "high",
-    "rationale": "The code does not handle potential errors when processing prompts, which could lead to unhandled exceptions during runtime.",
-    "recommendation": "Implement try-except blocks around critical sections of code that process prompts to ensure graceful error handling.",
-    "current_code": "desc = self._process_full_diff(prompt, user)",
-    "suggested_code": "try:\n    desc = self._process_full_diff(prompt, user)\nexcept Exception as e:\n    self.logger.error(f'Error processing full diff:{e}')\n    return 'Error processing request.'",
-    "file_path": "kaizen/generator/pr_description.py",
-    "start_line": 51,
-    "end_line": 55,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Redundant Code",
-    "description": "Redundant parameter in method signatures.",
-    "impact": "medium",
-    "rationale": "The parameter 'reeval_response' is removed in some methods but still appears in others where it is not used, leading to unnecessary complexity.",
-    "recommendation": "Remove unused parameters from method signatures to enhance clarity and maintainability.",
-    "current_code": "def _process_full_diff(self, prompt: str, user: Optional[str], reeval_response: bool) -> str:",
-    "suggested_code": "def _process_full_diff(self, prompt: str, user: Optional[str]) -> str:",
-    "file_path": "kaizen/generator/pr_description.py",
-    "start_line": 40,
-    "end_line": 41,
-    "sentiment": "neutral",
-    "severity": 4
-  },
-  {
-    "category": "Documentation",
-    "description": "Insufficient comments and documentation for new methods.",
-    "impact": "medium",
-    "rationale": "New methods lack docstrings, which makes it difficult for other developers to understand their purpose and usage.",
-    "recommendation": "Add docstrings to all new methods to describe their functionality, parameters, and return values.",
-    "current_code": "def _process_files(self, pull_request_files: List[Dict], pull_request_title: str, pull_request_desc: str, user: Optional[str]):",
-    "suggested_code": "def _process_files(self, pull_request_files: List[Dict], pull_request_title: str, pull_request_desc: str, user: Optional[str]):\n    \"\"\"\n    Process the provided pull request files and generate descriptions.\n    \n    Args:\n        pull_request_files (List[Dict]): List of pull request files.\n        pull_request_title (str): Title of the pull request.\n        pull_request_desc (str): Description of the pull request.\n        user (Optional[str]): User requesting the review.\n    \"\"\"",
-    "file_path": "kaizen/generator/pr_description.py",
-    "start_line": 85,
-    "end_line": 86,
-    "sentiment": "neutral",
-    "severity": 3
-  },
-  {
-    "category": "Performance",
-    "description": "Potential performance issue with JSON handling.",
-    "impact": "medium",
-    "rationale": "Using json.dumps in a loop can lead to performance degradation when processing large datasets.",
-    "recommendation": "Consider using a more efficient method for handling JSON, such as building the dictionary first and then dumping it once.",
-    "current_code": "prompt = MERGE_PR_DESCRIPTION_PROMPT.format(DESCS=json.dumps(file_descs))",
-    "suggested_code": "descs_dict ={'descs': file_descs}\nprompt = MERGE_PR_DESCRIPTION_PROMPT.format(DESCS=json.dumps(descs_dict))",
-    "file_path": "kaizen/generator/pr_description.py",
-    "start_line": 109,
-    "end_line": 110,
-    "sentiment": "neutral",
-    "severity": 6
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/review.md b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/review.md
deleted file mode 100644
index 23ae6698..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_335/review.md
+++ /dev/null
@@ -1,33 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/335
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 5
-- Critical: 0
-- Important: 0
-- Minor: 0
-- Files Affected: 2
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 6450, "completion_tokens": 1055, "total_tokens": 7505}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/comments.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/comments.json
deleted file mode 100644
index 1ceabeef..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/comments.json
+++ /dev/null
@@ -1,16 +0,0 @@
-[
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "11",
-    "end_line": "11",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/issues.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/issues.json
deleted file mode 100644
index 207b04b8..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/issues.json
+++ /dev/null
@@ -1,380 +0,0 @@
-[
-  {
-    "category": "Security",
-    "description": "Sensitive information exposure in .env.example",
-    "impact": "high",
-    "rationale": "The addition of LITELLM_LOG in the .env.example file may expose sensitive logging configurations if this file is shared or pushed to public repositories.",
-    "recommendation": "Consider using a more generic logging level or provide documentation on how to set sensitive configurations without exposing them in version control.",
-    "current_code": "LITELLM_LOG=\"ERROR\"",
-    "suggested_code": "# LITELLM_LOG=\"ERROR\" # Set in production environment only",
-    "file_path": ".env.example",
-    "start_line": 6,
-    "end_line": 6,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Readability",
-    "description": "Inconsistent formatting in test_create_pr_description.py",
-    "impact": "medium",
-    "rationale": "The formatting of the DESC_COLLAPSIBLE_TEMPLATE string is inconsistent with the rest of the code, which can lead to confusion and reduce readability.",
-    "recommendation": "Ensure consistent formatting across all string definitions, especially for multiline strings.",
-    "current_code": "DESC_COLLAPSIBLE_TEMPLATE = \"<details><summary>Original Description</summary>{desc}</details>\"",
-    "suggested_code": "DESC_COLLAPSIBLE_TEMPLATE = \"<details><summary>Original Description</summary>\\n\\n{desc}\\n\\n</details>\"",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
-    "start_line": 5,
-    "end_line": 5,
-    "sentiment": "neutral",
-    "severity": 4
-  },
-  {
-    "category": "Performance",
-    "description": "Potential performance issues with large input sizes in test_create_pr_description.py",
-    "impact": "medium",
-    "rationale": "The boundary condition tests with large strings could lead to performance degradation or memory issues if not handled properly.",
-    "recommendation": "Consider adding checks to limit the size of inputs or to ensure that the function can handle large inputs efficiently without excessive memory usage.",
-    "current_code": "def test_create_pr_description_boundary_conditions(desc, original_desc):",
-    "suggested_code": "# Add input size validation or handling logic here",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
-    "start_line": 46,
-    "end_line": 60,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Maintainability",
-    "description": "Repeated mock patching in test_create_folder.py",
-    "impact": "medium",
-    "rationale": "The repeated use of mock.patch can lead to code duplication and make it harder to maintain tests.",
-    "recommendation": "Consider using fixtures for mocking to reduce redundancy and improve maintainability.",
-    "current_code": "@mock.patch(\"kaizen.helpers.output.os.makedirs\")\n@mock.patch(\"kaizen.helpers.output.os.path.exists\")\n@mock.patch(\"kaizen.helpers.output.logger\")",
-    "suggested_code": "@pytest.fixture\ndef mock_os():\n    with mock.patch(\"kaizen.helpers.output.os\") as mock_os:\n        yield mock_os",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_folder.py",
-    "start_line": 6,
-    "end_line": 8,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Error Handling",
-    "description": "Lack of error handling in create_folder function",
-    "impact": "high",
-    "rationale": "The create_folder function should handle potential errors, such as permission issues or invalid paths, to prevent crashes.",
-    "recommendation": "Implement try-except blocks to catch and handle exceptions gracefully.",
-    "current_code": "os.makedirs(folder_path)",
-    "suggested_code": "try:\n    os.makedirs(folder_path)\nexcept OSError as e:\n    # Handle error (e.g., log it, raise a custom exception)",
-    "file_path": "kaizen/helpers/output.py",
-    "start_line": 10,
-    "end_line": 10,
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Code Readability",
-    "description": "Inconsistent formatting in PR_COLLAPSIBLE_TEMPLATE.",
-    "impact": "medium",
-    "rationale": "The original template uses a multi-line string with parentheses, while the new version uses triple quotes. This inconsistency can lead to confusion and makes the code harder to maintain.",
-    "recommendation": "Choose one style for multi-line strings and apply it consistently throughout the codebase.",
-    "current_code": "PR_COLLAPSIBLE_TEMPLATE = (\n    \"<details>\\n\"\n    \"<summary>Review</summary>\\n\"\n    ...\n)",
-    "suggested_code": "PR_COLLAPSIBLE_TEMPLATE = \"\"\"\n<details>\n<summary>Review Comment</summary>\n<p>{comment}</p>\n...\"\"\"",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 4,
-    "end_line": 16,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling",
-    "description": "Missing validation for required fields in review comments.",
-    "impact": "high",
-    "rationale": "The tests do not validate if required fields (like comment, reason, solution) are present. This could lead to runtime errors or unexpected behavior when these fields are missing.",
-    "recommendation": "Implement checks to ensure that all required fields are present before processing the review comments.",
-    "current_code": "def test_reviews_with_missing_fields():\n    topics ={...}",
-    "suggested_code": "def validate_review(review):\n    required_fields =['comment', 'reason', 'solution', 'confidence', 'start_line', 'end_line', 'file_name', 'severity']\n    for field in required_fields:\n        if field not in review:\n            raise ValueError(f'Missing required field:{field}')\n\n    # Call this validation function in your tests.",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 143,
-    "end_line": 184,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Test Coverage",
-    "description": "Lack of tests for edge cases in review comments.",
-    "impact": "medium",
-    "rationale": "The current tests do not cover scenarios where fields are missing or contain invalid data. This could lead to unhandled exceptions in production.",
-    "recommendation": "Add unit tests that cover various edge cases, such as missing fields, invalid data types, and empty inputs.",
-    "current_code": "def test_reviews_with_missing_fields():\n    topics ={...}",
-    "suggested_code": "def test_reviews_with_missing_fields():\n    # Test with missing fields\n    topics ={...}\n    with pytest.raises(ValueError):\n        create_pr_review_text(topics)",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 143,
-    "end_line": 184,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Performance",
-    "description": "Inefficient handling of review comments.",
-    "impact": "medium",
-    "rationale": "The current implementation processes review comments in a linear fashion. If the number of comments grows, this could lead to performance issues.",
-    "recommendation": "Consider using a more efficient data structure or algorithm to handle review comments, especially if the number of comments can be large.",
-    "current_code": "for topic, reviews in topics.items():\n    for review in reviews:\n        ...",
-    "suggested_code": "from collections import defaultdict\n\ncomments_by_topic = defaultdict(list)\nfor topic, reviews in topics.items():\n    comments_by_topic[topic].extend(reviews)\n# Process comments from comments_by_topic",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 80,
-    "end_line": 100,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Readability",
-    "description": "Inconsistent naming conventions for test cases.",
-    "impact": "medium",
-    "rationale": "Consistency in naming conventions improves readability and maintainability. Some test functions use underscores while others use camel case.",
-    "recommendation": "Adopt a consistent naming convention for all test functions, preferably using snake_case.",
-    "current_code": "def test_get_parent_folder_normal_case():",
-    "suggested_code": "def test_get_parent_folder_normal():",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_parent_folder.py",
-    "start_line": 6,
-    "end_line": 6,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling",
-    "description": "Lack of error handling for file operations.",
-    "impact": "high",
-    "rationale": "The `create_test_files` function may fail due to various reasons (e.g., permission issues), and this should be handled gracefully to avoid crashes.",
-    "recommendation": "Wrap file operations in try-except blocks to handle potential exceptions and log appropriate error messages.",
-    "current_code": "with open(file_path, 'r') as f:",
-    "suggested_code": "try:\n    with open(file_path, 'r') as f:\n        return f.read()\nexcept OSError as e:\n    # Log the error or handle it appropriately\n    raise e",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 17,
-    "end_line": 17,
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Performance",
-    "description": "Potential inefficiency in sanitizing filenames.",
-    "impact": "medium",
-    "rationale": "The current implementation of `sanitize_filename` iterates over each character in the filename, which could be optimized for better performance.",
-    "recommendation": "Consider using regular expressions to sanitize filenames, which can be more efficient and concise.",
-    "current_code": "return \"\".join(c if c.isalnum() or c in (' ', '.', '_') else '_' for c in filename)",
-    "suggested_code": "import re\nreturn re.sub(r'[^\\w .]', '_', filename)",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 21,
-    "end_line": 22,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Testing Coverage",
-    "description": "Insufficient testing for edge cases.",
-    "impact": "high",
-    "rationale": "While there are tests for normal cases, edge cases such as very long filenames or special characters are not thoroughly tested.",
-    "recommendation": "Add more test cases to cover edge scenarios, such as filenames with special characters or exceeding typical length limits.",
-    "current_code": "def test_special_characters_in_test_names(tmp_path, mock_dependencies):",
-    "suggested_code": "def test_edge_cases_in_filenames(tmp_path, mock_dependencies):\n    # Add tests for various edge cases here",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 104,
-    "end_line": 106,
-    "sentiment": "neutral",
-    "severity": 7
-  },
-  {
-    "category": "Code Structure",
-    "description": "Redundant mock patches in tests.",
-    "impact": "medium",
-    "rationale": "Repeatedly patching the same functions in multiple tests can lead to code duplication and make the tests harder to maintain.",
-    "recommendation": "Use fixtures to handle common mock patches, reducing redundancy and improving readability.",
-    "current_code": "@mock.patch(\"kaizen.helpers.output.create_folder\")\n@mock.patch(\"kaizen.helpers.output.general.clean_python_code\")\n@mock.patch(\"kaizen.helpers.output.logger\")\ndef test_create_test_files_special_characters(...):",
-    "suggested_code": "@pytest.fixture\ndef mock_dependencies():\n    with mock.patch(...) as mock_create_folder, ...:\n        yield mock_create_folder, ...",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 84,
-    "end_line": 88,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Code Efficiency",
-    "description": "Use of synchronous function in an asynchronous context.",
-    "impact": "high",
-    "rationale": "The `get_web_html` function is being called synchronously within an async test, which can lead to blocking behavior and defeat the purpose of using async functions.",
-    "recommendation": "Ensure that `get_web_html` is awaited in all async contexts to maintain non-blocking behavior.",
-    "current_code": "result = get_web_html(url)",
-    "suggested_code": "result = await get_web_html(url)",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 92,
-    "end_line": 92,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Error Handling",
-    "description": "Lack of error handling in tests for invalid URLs.",
-    "impact": "high",
-    "rationale": "The test for invalid URLs should assert that an exception is raised, ensuring that the function behaves correctly under error conditions.",
-    "recommendation": "Wrap the call to `get_web_html` in a pytest.raises context manager to verify that the expected exception is thrown.",
-    "current_code": "result = get_web_html(url)",
-    "suggested_code": "with pytest.raises(Exception, match='Network error'):\n    await get_web_html(url)",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 89,
-    "end_line": 91,
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Readability",
-    "description": "Inconsistent use of single and double quotes.",
-    "impact": "medium",
-    "rationale": "Inconsistent quoting can lead to confusion and makes the code less readable. It's a good practice to stick to one style throughout the codebase.",
-    "recommendation": "Choose either single or double quotes for string literals and apply it consistently across the code.",
-    "current_code": "with patch('kaizen.helpers.output.get_html', new_callable=AsyncMock) as mock:",
-    "suggested_code": "with patch(\"kaizen.helpers.output.get_html\", new_callable=AsyncMock) as mock:",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 12,
-    "end_line": 12,
-    "sentiment": "neutral",
-    "severity": 4
-  },
-  {
-    "category": "Test Coverage",
-    "description": "Lack of tests for edge cases and large inputs.",
-    "impact": "medium",
-    "rationale": "While there are tests for normal cases, edge cases such as extremely large HTML content or malformed HTML should also be tested to ensure robustness.",
-    "recommendation": "Add tests that specifically handle edge cases and large inputs to ensure that the function can handle unexpected scenarios.",
-    "current_code": "No specific tests for edge cases.",
-    "suggested_code": "async def test_get_web_html_large_content(mock_get_html):\n    large_html_content = '<html>' + '<p>Test</p>' * 10000 + '</html>'\n    expected_output = '<html><body>' + '<p>Test</p>' * 10000 + '</body></html>'\n    mock_get_html.return_value = large_html_content\n    result = await get_web_html(url)\n    assert result.strip() == expected_output.strip()",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 93,
-    "end_line": 95,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Code Structure",
-    "description": "Inconsistent use of class attributes for supported languages and prompts.",
-    "impact": "high",
-    "rationale": "The code uses both instance attributes and class attributes for similar purposes, which can lead to confusion and inconsistency in how the data is accessed and modified.",
-    "recommendation": "Standardize the use of either class attributes or instance attributes for supported languages and prompts to improve clarity.",
-    "current_code": "self.supported_languages ={...}",
-    "suggested_code": "SUPPORTED_LANGUAGES ={...}",
-    "file_path": "kaizen/generator/unit_test.py",
-    "start_line": 24,
-    "end_line": 30,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Error Handling",
-    "description": "Lack of error handling when reading files.",
-    "impact": "high",
-    "rationale": "If the file does not exist or cannot be read, the current implementation will raise an unhandled exception, which could crash the application.",
-    "recommendation": "Implement error handling using try-except blocks when reading files to gracefully handle potential issues.",
-    "current_code": "with open(file_path, 'r') as file: content = file.read()",
-    "suggested_code": "try:\n    with open(file_path, 'r') as file:\n        return file.read()\nexcept IOError as e:\n    print(f'Error reading file:{e}')\n    return None",
-    "file_path": "kaizen/generator/unit_test.py",
-    "start_line": 108,
-    "end_line": 110,
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Readability",
-    "description": "Long methods with multiple responsibilities.",
-    "impact": "medium",
-    "rationale": "Methods like `generate_tests_from_dir` and `generate_ai_tests` are doing too much, making them harder to read and maintain.",
-    "recommendation": "Refactor long methods into smaller, single-responsibility methods to enhance readability and maintainability.",
-    "current_code": "def generate_tests_from_dir(self, ...): ...",
-    "suggested_code": "def generate_tests_from_dir(self, ...):\n    self._prepare_tests(...)\n    self._execute_tests(...)",
-    "file_path": "kaizen/generator/unit_test.py",
-    "start_line": 61,
-    "end_line": 61,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Performance",
-    "description": "Repeated calls to `os.makedirs` for the same directory.",
-    "impact": "medium",
-    "rationale": "The code calls `os.makedirs` multiple times for the same directory, which can be inefficient.",
-    "recommendation": "Check if the directory exists before creating it to avoid unnecessary calls.",
-    "current_code": "os.makedirs(self.log_dir, exist_ok=True)",
-    "suggested_code": "if not os.path.exists(self.log_dir):\n    os.makedirs(self.log_dir)",
-    "file_path": "kaizen/generator/unit_test.py",
-    "start_line": 58,
-    "end_line": 58,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Logging",
-    "description": "Lack of logging for critical operations.",
-    "impact": "medium",
-    "rationale": "The absence of logging makes it difficult to trace the flow of execution and diagnose issues.",
-    "recommendation": "Add logging statements to critical operations to provide better visibility into the application's behavior.",
-    "current_code": "print(f'Error: Could not generate tests for{file_path}:{e}')",
-    "suggested_code": "self.logger.error(f'Could not generate tests for{file_path}:{e}')",
-    "file_path": "kaizen/generator/unit_test.py",
-    "start_line": 73,
-    "end_line": 73,
-    "sentiment": "neutral",
-    "severity": 4
-  },
-  {
-    "category": "Logging Best Practices",
-    "description": "Improper handling of logger levels.",
-    "impact": "high",
-    "rationale": "Setting all loggers to ERROR without a conditional check can lead to loss of important log information. This could make debugging difficult, especially in production environments where lower log levels may be necessary for tracing issues.",
-    "recommendation": "Consider allowing configurable log levels instead of hardcoding ERROR for all loggers. This can be done via environment variables or configuration files.",
-    "current_code": "logging.getLogger(name).setLevel(logging.ERROR)",
-    "suggested_code": "level = os.environ.get('DEFAULT_LOG_LEVEL', 'ERROR').upper()\nlogging.getLogger(name).setLevel(getattr(logging, level, logging.ERROR))",
-    "file_path": "kaizen/llms/provider.py",
-    "start_line": 13,
-    "end_line": 28,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Code Readability",
-    "description": "Lack of docstrings and comments in the new function.",
-    "impact": "medium",
-    "rationale": "The function `set_all_loggers_to_ERROR` lacks a docstring, making it unclear what its purpose is. Clear documentation is essential for maintainability and understanding of the code.",
-    "recommendation": "Add a docstring to explain the purpose of the function and its behavior.",
-    "current_code": "def set_all_loggers_to_ERROR():",
-    "suggested_code": "def set_all_loggers_to_ERROR():\n    \"\"\"\n    Set all loggers to the ERROR level.\n    This function iterates through all loggers and sets their level to ERROR, which can help in reducing log verbosity.\n    \"\"\"",
-    "file_path": "kaizen/llms/provider.py",
-    "start_line": 13,
-    "end_line": 13,
-    "sentiment": "negative",
-    "severity": 5
-  },
-  {
-    "category": "Performance",
-    "description": "Inefficient logging of all logger names.",
-    "impact": "medium",
-    "rationale": "Printing all logger names and their levels can be inefficient and clutter the output, especially if there are many loggers. This could lead to performance issues in environments with extensive logging.",
-    "recommendation": "Consider logging only the loggers that are set to a certain level or have specific names of interest.",
-    "current_code": "print(f\"Logger:{name}, Level:{logging.getLevelName(logger.level)}\")",
-    "suggested_code": "if logger.level <= logging.WARNING:\n    print(f\"Logger:{name}, Level:{logging.getLevelName(logger.level)}\")",
-    "file_path": "kaizen/llms/provider.py",
-    "start_line": 14,
-    "end_line": 20,
-    "sentiment": "negative",
-    "severity": 6
-  },
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "11",
-    "end_line": "11",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/review.md b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/review.md
deleted file mode 100644
index bdd46645..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_400/review.md
+++ /dev/null
@@ -1,56 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/400
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 27
-- Critical: 1
-- Important: 0
-- Minor: 0
-- Files Affected: 11
-## 🏆 Code Quality
-[███████████████░░░░░] 75% (Fair)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Configuration (1 issues)</strong></summary>
-
-### 1. Changes made to sensitive file
-📁 **File:** `config.json:11`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 35088, "completion_tokens": 5327, "total_tokens": 40415}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/comments.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/comments.json
deleted file mode 100644
index 0637a088..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/comments.json
+++ /dev/null
@@ -1 +0,0 @@
-[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/issues.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/issues.json
deleted file mode 100644
index e9c23d5a..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/issues.json
+++ /dev/null
@@ -1,44 +0,0 @@
-[
-  {
-    "category": "Code Efficiency",
-    "description": "Redundant repository setup call.",
-    "impact": "medium",
-    "rationale": "The repository setup is called every time the script runs, which may not be necessary if the repository has already been set up. This can lead to unnecessary overhead.",
-    "recommendation": "Check if the repository is already set up before calling setup_repository.",
-    "current_code": "analyzer.setup_repository(\"./github_app/\")",
-    "suggested_code": "if not analyzer.is_repository_setup():\n    analyzer.setup_repository(\"./github_app/\")",
-    "file_path": "examples/ragify_codebase/main.py",
-    "start_line": 7,
-    "end_line": 7,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Code Clarity",
-    "description": "Missing comment on potential duplicate embedding storage.",
-    "impact": "high",
-    "rationale": "The TODO comment indicates a potential issue with duplicate embeddings being pushed to the database, which could lead to data integrity issues. Clear documentation is essential for maintainability.",
-    "recommendation": "Implement a check to prevent duplicate embeddings from being stored and document the logic clearly.",
-    "current_code": "# TODO: DONT PUSH DUPLICATE",
-    "suggested_code": "# Check for duplicates before storing the embedding\nif not self.is_duplicate_embedding(embedding):\n    # Store the embedding in the database",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 157,
-    "end_line": 157,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Dependency Management",
-    "description": "Outdated dependency version.",
-    "impact": "medium",
-    "rationale": "Updating dependencies is crucial for security and performance improvements. The change from version 0.10.47 to 0.10.65 should be validated for compatibility.",
-    "recommendation": "Ensure that the new version is compatible with the existing codebase and run tests after updating.",
-    "current_code": "llama-index-core = \"^0.10.47\"",
-    "suggested_code": "llama-index-core = \"0.10.65\"",
-    "file_path": "pyproject.toml",
-    "start_line": 27,
-    "end_line": 27,
-    "sentiment": "neutral",
-    "severity": 4
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/review.md b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/review.md
deleted file mode 100644
index 44d9c077..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_440/review.md
+++ /dev/null
@@ -1,33 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/440
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 3
-- Critical: 0
-- Important: 0
-- Minor: 0
-- Files Affected: 3
-## 🏆 Code Quality
-[███████████████░░░░░] 75% (Fair)
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 1063, "completion_tokens": 567, "total_tokens": 1630}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/comments.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/comments.json
deleted file mode 100644
index 752f765d..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/comments.json
+++ /dev/null
@@ -1,30 +0,0 @@
-[
-  {
-    "category": "Security",
-    "description": "Potential exposure of sensitive information in logs.",
-    "impact": "critical",
-    "rationale": "Logging sensitive information, such as access tokens or user data, can lead to security vulnerabilities. It is crucial to avoid logging sensitive data.",
-    "recommendation": "Review logging statements to ensure that no sensitive information is being logged. Use obfuscation or redaction where necessary.",
-    "current_code": "logger.debug(f\"Post Review comment response:{response.text}\")",
-    "suggested_code": "logger.debug(f\"Post Review comment response:{response.status_code}\")  # Avoid logging response body",
-    "file_path": "github_app/github_helper/pull_requests.py",
-    "start_line": 181,
-    "end_line": 182,
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/issues.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/issues.json
deleted file mode 100644
index fe021299..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/issues.json
+++ /dev/null
@@ -1,86 +0,0 @@
-[
-  {
-    "category": "Error Handling",
-    "description": "Broad exception handling without specific error types.",
-    "impact": "high",
-    "rationale": "Using a generic Exception in the error handling can mask underlying issues and make debugging difficult. It is better to catch specific exceptions to handle known error cases appropriately.",
-    "recommendation": "Refine the exception handling to catch specific exceptions and log or handle them accordingly.",
-    "current_code": "except Exception:\n    print(\"Error\")",
-    "suggested_code": "except KeyError:\n    raise ValueError(f\"Invalid confidence level:{min_confidence}\")\nexcept ValueError as ve:\n    logger.error(f\"ValueError:{ve}\")",
-    "file_path": "github_app/github_helper/pull_requests.py",
-    "start_line": 140,
-    "end_line": 141,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Code Readability",
-    "description": "Inconsistent naming conventions and lack of comments.",
-    "impact": "medium",
-    "rationale": "Inconsistent naming can lead to confusion and make the code harder to maintain. Comments are essential for explaining complex logic.",
-    "recommendation": "Ensure consistent naming conventions across the codebase and add comments where necessary to explain the purpose of complex logic.",
-    "current_code": "def post_pull_request(url, data, installation_id, tests=None):",
-    "suggested_code": "def post_pull_request(url, data, installation_id, test_files=None):  # Renamed for clarity",
-    "file_path": "github_app/github_helper/pull_requests.py",
-    "start_line": 106,
-    "end_line": 107,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Performance",
-    "description": "Inefficient file sorting algorithm.",
-    "impact": "high",
-    "rationale": "The current sorting algorithm has a time complexity of O(n^2), which can lead to performance issues with larger datasets. Using Python's built-in sorting methods would be more efficient.",
-    "recommendation": "Replace the custom sorting logic with Python's built-in sort() method for better performance.",
-    "current_code": "for file in files:\n    min_index = len(sorted_files)\n    file_name = file[\"filename\"]\n    for i, sorted_file in enumerate(sorted_files):\n        if file_name < sorted_file[\"filename\"]:\n            min_index = i\n            break\n    sorted_files.insert(min_index, file)",
-    "suggested_code": "sorted_files = sorted(files, key=lambda x: x['filename'])",
-    "file_path": "github_app/github_helper/pull_requests.py",
-    "start_line": 184,
-    "end_line": 196,
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Unit Testing",
-    "description": "Lack of unit tests for new functionality.",
-    "impact": "high",
-    "rationale": "New functionality should be accompanied by unit tests to ensure correctness and prevent regressions in the future.",
-    "recommendation": "Add unit tests for the new functions, especially for `generate_tests` and `sort_files`, to validate their behavior.",
-    "current_code": "def generate_tests(pr_files):\n    return[f[\"filename\"] for f in pr_files]",
-    "suggested_code": "# Add unit tests in a separate test file for generate_tests and sort_files",
-    "file_path": "github_app/github_helper/pull_requests.py",
-    "start_line": 199,
-    "end_line": 200,
-    "sentiment": "negative",
-    "severity": 9
-  },
-  {
-    "category": "Security",
-    "description": "Potential exposure of sensitive information in logs.",
-    "impact": "critical",
-    "rationale": "Logging sensitive information, such as access tokens or user data, can lead to security vulnerabilities. It is crucial to avoid logging sensitive data.",
-    "recommendation": "Review logging statements to ensure that no sensitive information is being logged. Use obfuscation or redaction where necessary.",
-    "current_code": "logger.debug(f\"Post Review comment response:{response.text}\")",
-    "suggested_code": "logger.debug(f\"Post Review comment response:{response.status_code}\")  # Avoid logging response body",
-    "file_path": "github_app/github_helper/pull_requests.py",
-    "start_line": 181,
-    "end_line": 182,
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/review.md b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/review.md
deleted file mode 100644
index 11cdc033..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_476/review.md
+++ /dev/null
@@ -1,72 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/476
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 6
-- Critical: 2
-- Important: 0
-- Minor: 0
-- Files Affected: 2
-## 🏆 Code Quality
-[███████████████░░░░░] 75% (Fair)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Security (2 issues)</strong></summary>
-
-### 1. Potential exposure of sensitive information in logs.
-📁 **File:** `github_app/github_helper/pull_requests.py:181`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Potential exposure of sensitive information in logs.
-💡 **Solution:** 
-
-**Current Code:**
-```python
-logger.debug(f"Post Review comment response:{response.text}")
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 2. Changes made to sensitive file
-📁 **File:** `config.json:1`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 3706, "completion_tokens": 1023, "total_tokens": 4729}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/comments.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/comments.json
deleted file mode 100644
index 0637a088..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/comments.json
+++ /dev/null
@@ -1 +0,0 @@
-[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/issues.json b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/issues.json
deleted file mode 100644
index d942a77a..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/issues.json
+++ /dev/null
@@ -1,86 +0,0 @@
-[
-  {
-    "category": "Unused Imports",
-    "description": "The import of 'random' is unnecessary and should be removed.",
-    "impact": "low",
-    "rationale": "Unused imports can clutter the code and increase loading time slightly.",
-    "recommendation": "Remove the import statement for 'random'.",
-    "current_code": "import random  # Unused import",
-    "suggested_code": "",
-    "file_path": "main.py",
-    "start_line": 8,
-    "end_line": 8,
-    "sentiment": "negative",
-    "severity": 3
-  },
-  {
-    "category": "Error Handling",
-    "description": "The API call in 'process_applicant' lacks a retry mechanism.",
-    "impact": "high",
-    "rationale": "If the API call fails, the function will throw an error without any recovery, leading to potential data loss.",
-    "recommendation": "Implement a retry mechanism with exponential backoff for the API call.",
-    "current_code": "response = completion(...)",
-    "suggested_code": "import time\n\nfor attempt in range(3):\n    try:\n        response = completion(...)\n        break\n    except Exception as e:\n        if attempt < 2:\n            time.sleep(2 ** attempt)\n        else:\n            raise e",
-    "file_path": "main.py",
-    "start_line": 65,
-    "end_line": 68,
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Silent Failures",
-    "description": "The JSON parsing failure in 'process_applicant' is silent and does not log the error.",
-    "impact": "high",
-    "rationale": "Silent failures can lead to debugging difficulties and loss of important information.",
-    "recommendation": "Log the error message instead of just printing it.",
-    "current_code": "print(f\"Failed to parse content for applicant\")",
-    "suggested_code": "import logging\n\nlogging.error(f\"Failed to parse content for applicant:{e}\")",
-    "file_path": "main.py",
-    "start_line": 86,
-    "end_line": 86,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Redundant Code",
-    "description": "The check for an empty DataFrame in 'main' is unnecessary.",
-    "impact": "low",
-    "rationale": "If the DataFrame is empty, the process_applicants function will handle it gracefully.",
-    "recommendation": "Remove the check for an empty DataFrame.",
-    "current_code": "if len(df) == 0:\n    return",
-    "suggested_code": "",
-    "file_path": "main.py",
-    "start_line": 141,
-    "end_line": 143,
-    "sentiment": "neutral",
-    "severity": 2
-  },
-  {
-    "category": "Division by Zero",
-    "description": "Potential division by zero when calculating total tokens.",
-    "impact": "high",
-    "rationale": "If total_tokens is zero, it will raise an exception.",
-    "recommendation": "Add a check to prevent division by zero.",
-    "current_code": "print(f\"Total tokens used:{total_tokens:,}\")",
-    "suggested_code": "if total_tokens > 0:\n    print(f\"Total tokens used:{total_tokens:,}\")\nelse:\n    print(\"No tokens used.\")",
-    "file_path": "main.py",
-    "start_line": 159,
-    "end_line": 161,
-    "sentiment": "negative",
-    "severity": 6
-  },
-  {
-    "category": "File Handling",
-    "description": "No error handling for file not found in 'main'.",
-    "impact": "high",
-    "rationale": "If the file does not exist, the program will crash without a user-friendly message.",
-    "recommendation": "Add a try-except block to handle file not found errors.",
-    "current_code": "main(input_file)",
-    "suggested_code": "try:\n    main(input_file)\nexcept FileNotFoundError:\n    print(f\"Error: File '{input_file}' not found.\")",
-    "file_path": "main.py",
-    "start_line": 175,
-    "end_line": 175,
-    "sentiment": "negative",
-    "severity": 8
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/review.md b/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/review.md
deleted file mode 100644
index 112891fd..00000000
--- a/.experiments/code_review/code_reviews_20240827_163829/no_eval/pr_5/review.md
+++ /dev/null
@@ -1,41 +0,0 @@
-PR URL: https://github.com/sauravpanda/applicant-screening/pull/5
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 6
-- Critical: 0
-- Important: 0
-- Minor: 0
-- Files Affected: 1
-## 🏆 Code Quality
-[███████████████░░░░░] 75% (Fair)
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (2 issues)</strong></summary>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 5853, "completion_tokens": 998, "total_tokens": 6851}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/comments.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/comments.json
deleted file mode 100644
index 8b8c06d9..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/comments.json
+++ /dev/null
@@ -1,86 +0,0 @@
-[
-  {
-    "category": "Security Vulnerabilities",
-    "description": "Hard-coded API keys in config.json",
-    "impact": "critical",
-    "rationale": "Storing sensitive information such as API keys directly in the codebase can lead to security breaches if the repository is exposed. This practice violates security best practices and can compromise application integrity.",
-    "recommendation": "Use environment variables or a secure vault to manage sensitive information instead of hard-coding them in configuration files.",
-    "current_code": "\"api_key\": \"os.environ/AZURE_API_KEY\",",
-    "suggested_code": "\"api_key\": \"${AZURE_API_KEY}\",",
-    "file_path": "config.json",
-    "start_line": 13,
-    "end_line": 14,
-    "sentiment": "negative",
-    "severity": 9
-  },
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Docker",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to Dockerfile, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_path": "Dockerfile",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Docker",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to docker-compose.yml, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "15",
-    "end_line": "15",
-    "side": "RIGHT",
-    "file_path": "docker-compose.yml",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Version Control",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to .gitignore, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "164",
-    "end_line": "164",
-    "side": "RIGHT",
-    "file_path": ".gitignore",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Database",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to db_setup/init.sql, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_path": "db_setup/init.sql",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/issues.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/issues.json
deleted file mode 100644
index 81b0a7b6..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/issues.json
+++ /dev/null
@@ -1,282 +0,0 @@
-[
-  {
-    "category": "Security Vulnerabilities",
-    "description": "Hard-coded API keys in config.json",
-    "impact": "critical",
-    "rationale": "Storing sensitive information such as API keys directly in the codebase can lead to security breaches if the repository is exposed. This practice violates security best practices and can compromise application integrity.",
-    "recommendation": "Use environment variables or a secure vault to manage sensitive information instead of hard-coding them in configuration files.",
-    "current_code": "\"api_key\": \"os.environ/AZURE_API_KEY\",",
-    "suggested_code": "\"api_key\": \"${AZURE_API_KEY}\",",
-    "file_path": "config.json",
-    "start_line": 13,
-    "end_line": 14,
-    "sentiment": "negative",
-    "severity": 9
-  },
-  {
-    "category": "Code Structure and Design",
-    "description": "Lack of comments in critical sections of code",
-    "impact": "medium",
-    "rationale": "While the code is generally readable, critical sections, especially in the Dockerfile and SQL scripts, lack comments explaining their purpose. This can hinder maintainability and onboarding for new developers.",
-    "recommendation": "Add comments to explain the purpose of each section, especially in complex scripts and Docker configurations.",
-    "current_code": "RUN apt-get update && apt-get install -y \\",
-    "suggested_code": "# Update package list and install necessary dependencies\nRUN apt-get update && apt-get install -y \\",
-    "file_path": "Dockerfile",
-    "start_line": 8,
-    "end_line": 11,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling and Logging",
-    "description": "Lack of error handling in database operations",
-    "impact": "high",
-    "rationale": "The database operations in the custom_query method do not handle potential exceptions that may arise from database connectivity issues or query failures. This can lead to unhandled exceptions and application crashes.",
-    "recommendation": "Implement try-except blocks around database operations to catch exceptions and log errors appropriately.",
-    "current_code": "with client.cursor() as cur:\n    cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))",
-    "suggested_code": "try:\n    with client.cursor() as cur:\n        cur.execute(query, (query_embedding_normalized.tolist(), repo_id, similarity_top_k))\nexcept Exception as e:\n    # Log the error\n    print(f\"Database query failed:{e}\")",
-    "file_path": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 39,
-    "end_line": 41,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Performance Issues",
-    "description": "Inefficient handling of query results",
-    "impact": "medium",
-    "rationale": "The current implementation fetches all results from the database and processes them in memory. This can lead to performance issues with large datasets.",
-    "recommendation": "Consider using pagination or streaming results to handle large datasets more efficiently.",
-    "current_code": "results = cur.fetchall()",
-    "suggested_code": "# Consider using a generator to yield results one by one\nfor row in cur:\n    # Process each row as it is fetched",
-    "file_path": "kaizen/retriever/custom_vector_store.py",
-    "start_line": 42,
-    "end_line": 42,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Code Structure and Design",
-    "description": "Code duplication in Dockerfiles",
-    "impact": "medium",
-    "rationale": "The Dockerfile and Dockerfile-postgres have similar sections for installing dependencies. This can lead to maintenance challenges if changes are needed in the future.",
-    "recommendation": "Consider creating a base Dockerfile that includes common dependencies and extend it in specific Dockerfiles.",
-    "current_code": "RUN apt-get update && apt-get install -y \\\n    git \\\n    build-essential \\\n    && rm -rf /var/lib/apt/lists/*",
-    "suggested_code": "# Base Dockerfile\n# Common dependencies can be installed here\nRUN apt-get update && apt-get install -y \\\n    git \\\n    build-essential \\\n    && rm -rf /var/lib/apt/lists/*",
-    "file_path": "Dockerfile",
-    "start_line": 8,
-    "end_line": 11,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling and Logging",
-    "description": "Inconsistent error handling in the `generate_abstraction` method.",
-    "impact": "high",
-    "rationale": "The `generate_abstraction` method raises exceptions directly without logging them, which may lead to loss of context in error tracking.",
-    "recommendation": "Log exceptions before raising them to maintain context for debugging.",
-    "current_code": "raise e",
-    "suggested_code": "logger.error(f'Error generating abstraction:{str(e)}'); raise e",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 218,
-    "end_line": 219,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Code Structure and Design",
-    "description": "Hard-coded values for start and end lines in `store_code_in_db` method.",
-    "impact": "medium",
-    "rationale": "Using hard-coded values for line numbers can lead to inaccuracies and maintenance issues when the code changes.",
-    "recommendation": "Calculate actual start and end lines based on the parsed content.",
-    "current_code": "\"start_line\": 1,  # You might want to calculate actual start and end lines",
-    "suggested_code": "\"start_line\": actual_start_line, \"end_line\": actual_end_line,",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 287,
-    "end_line": 287,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Performance Issues",
-    "description": "Potential N+1 query problem in `store_function_relationships` method.",
-    "impact": "high",
-    "rationale": "The current implementation executes a separate query for each edge in the graph, which can lead to performance bottlenecks.",
-    "recommendation": "Batch insert relationships or use a single query to insert multiple relationships at once.",
-    "current_code": "connection.execute(query,{\"caller\": f\"%{caller}%\", \"callee\": f\"%{callee}%\"})",
-    "suggested_code": "batch_insert_query = ...; connection.execute(batch_insert_query, batch_params)",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 295,
-    "end_line": 312,
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Code Structure and Design",
-    "description": "The `parse_file` method has high cyclomatic complexity.",
-    "impact": "medium",
-    "rationale": "High cyclomatic complexity can make the code harder to understand and maintain. The method has multiple nested conditions.",
-    "recommendation": "Refactor the method to break it into smaller, more manageable functions.",
-    "current_code": "if isinstance(items, dict): ... elif isinstance(items, list): ...",
-    "suggested_code": "def handle_dict(items): ...; def handle_list(items): ...; then call these in parse_file",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 98,
-    "end_line": 102,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Resource Management",
-    "description": "File handles in `parse_file` method are not guaranteed to be closed in case of an exception.",
-    "impact": "high",
-    "rationale": "If an exception occurs before the file is closed, it may lead to resource leaks.",
-    "recommendation": "Use a context manager to ensure the file is properly closed.",
-    "current_code": "with open(file_path, \"r\", encoding=\"utf-8\") as file:",
-    "suggested_code": "with open(file_path, \"r\", encoding=\"utf-8\") as file: ...",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 92,
-    "end_line": 92,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Error Handling and Logging",
-    "description": "Broad exception handling in language loading and parser creation.",
-    "impact": "high",
-    "rationale": "Using a broad exception catch can obscure the root cause of errors and make debugging difficult. Specific exceptions should be caught to provide clearer error messages and handling.",
-    "recommendation": "Catch specific exceptions (e.g., ImportError, ValueError) instead of a generic Exception to improve error clarity.",
-    "current_code": "except Exception as e:",
-    "suggested_code": "except (ImportError, ValueError) as e:",
-    "file_path": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 28,
-    "end_line": 30,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Code Structure and Design",
-    "description": "Inconsistent naming conventions for functions and variables.",
-    "impact": "medium",
-    "rationale": "Inconsistent naming conventions can lead to confusion and reduce code readability. Following a consistent style guide improves maintainability.",
-    "recommendation": "Use snake_case for function names and variables to maintain consistency with Python's PEP 8 style guide.",
-    "current_code": "def check_language_files():",
-    "suggested_code": "def check_language_files():",
-    "file_path": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 101,
-    "end_line": 101,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Performance Issues",
-    "description": "Potential inefficiency in dynamic module loading.",
-    "impact": "medium",
-    "rationale": "Dynamically importing modules can introduce overhead and may lead to performance issues if done repeatedly in a loop. Consider caching or pre-loading modules if they are frequently accessed.",
-    "recommendation": "Evaluate if dynamic imports can be replaced with static imports or cached to improve performance.",
-    "current_code": "module = importlib.import_module(module_name)",
-    "suggested_code": "# Consider caching or pre-loading modules if frequently accessed.",
-    "file_path": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 23,
-    "end_line": 23,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Code Structure and Design",
-    "description": "Redundant code in traverse_tree function for JSX elements.",
-    "impact": "medium",
-    "rationale": "The logic for handling JSX elements is duplicated and can be simplified, improving readability and maintainability.",
-    "recommendation": "Refactor the JSX handling logic to reduce duplication.",
-    "current_code": "return{'type': 'component', 'name': (node.child_by_field_name('opening_element').child_by_field_name('name').text.decode('utf8') if node.type == 'jsx_element' else node.child_by_field_name('name').text.decode('utf8')), 'code': code_bytes[node.start_byte : node.end_byte].decode('utf8')}",
-    "suggested_code": "name = node.child_by_field_name('opening_element').child_by_field_name('name').text.decode('utf8') if node.type == 'jsx_element' else node.child_by_field_name('name').text.decode('utf8'); return{'type': 'component', 'name': name, 'code': code_bytes[node.start_byte : node.end_byte].decode('utf8')}",
-    "file_path": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 71,
-    "end_line": 79,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Resource Management",
-    "description": "Lack of resource cleanup in parser creation.",
-    "impact": "high",
-    "rationale": "Not properly managing resources can lead to memory leaks or unclosed connections, especially in a long-running application.",
-    "recommendation": "Ensure that resources are properly closed or disposed of after use, especially in the ParserFactory.",
-    "current_code": "parser = Parser()",
-    "suggested_code": "# Ensure parser is properly disposed of after use.",
-    "file_path": "kaizen/retriever/tree_sitter_utils.py",
-    "start_line": 38,
-    "end_line": 38,
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Docker",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to Dockerfile, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "4",
-    "end_line": "4",
-    "side": "RIGHT",
-    "file_path": "Dockerfile",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Docker",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to docker-compose.yml, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "15",
-    "end_line": "15",
-    "side": "RIGHT",
-    "file_path": "docker-compose.yml",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Version Control",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to .gitignore, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "164",
-    "end_line": "164",
-    "side": "RIGHT",
-    "file_path": ".gitignore",
-    "sentiment": "negative",
-    "severity": 10
-  },
-  {
-    "category": "Database",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to db_setup/init.sql, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_path": "db_setup/init.sql",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/review.md b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/review.md
deleted file mode 100644
index 6183caac..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_222/review.md
+++ /dev/null
@@ -1,136 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/222
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 20
-- Critical: 6
-- Important: 0
-- Minor: 0
-- Files Affected: 8
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Security Vulnerabilities (6 issues)</strong></summary>
-
-### 1. Hard-coded API keys in config.json
-📁 **File:** `config.json:13`
-⚖️ **Severity:** 9/10
-🔍 **Description:** Hard-coded API keys in config.json
-💡 **Solution:** 
-
-**Current Code:**
-```python
-"api_key": "os.environ/AZURE_API_KEY",
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 2. Changes made to sensitive file
-📁 **File:** `config.json:4`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 3. Changes made to sensitive file
-📁 **File:** `Dockerfile:4`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 4. Changes made to sensitive file
-📁 **File:** `docker-compose.yml:15`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 5. Changes made to sensitive file
-📁 **File:** `.gitignore:164`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 6. Changes made to sensitive file
-📁 **File:** `db_setup/init.sql:1`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 20363, "completion_tokens": 2974, "total_tokens": 23337}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/comments.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/comments.json
deleted file mode 100644
index 0637a088..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/comments.json
+++ /dev/null
@@ -1 +0,0 @@
-[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/issues.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/issues.json
deleted file mode 100644
index 1e3db8ca..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/issues.json
+++ /dev/null
@@ -1,156 +0,0 @@
-[
-  {
-    "category": "Code Structure and Design",
-    "description": "Inconsistent naming conventions for components.",
-    "impact": "medium",
-    "rationale": "The naming of components should be consistent to improve readability and maintainability. For example, 'LinkComponent' and 'MemoryComponent' should follow a uniform naming pattern to indicate their purpose clearly.",
-    "recommendation": "Rename components to maintain consistency in naming conventions.",
-    "current_code": "function LinkComponent({...}){...}",
-    "suggested_code": "function MemoryComponent({...}){...}",
-    "file_path": "apps/web/app/(dash)/home/page.tsx",
-    "start_line": 321,
-    "end_line": 321,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling and Logging",
-    "description": "Lack of error handling in asynchronous operations.",
-    "impact": "high",
-    "rationale": "The function 'handleDeleteSpace' does not handle potential errors from the 'deleteSpace' API call. This could lead to unhandled promise rejections and a poor user experience.",
-    "recommendation": "Implement error handling for the asynchronous operation to ensure that errors are caught and logged appropriately.",
-    "current_code": "const response = await deleteSpace(id);",
-    "suggested_code": "try{const response = await deleteSpace(id);}catch (error){console.error('Error deleting space:', error); toast.error('An error occurred while deleting the space.');}",
-    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 66,
-    "end_line": 70,
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Performance Issues",
-    "description": "Inefficient merging and sorting of items.",
-    "impact": "medium",
-    "rationale": "The current approach merges and sorts items in a way that could be optimized. The use of multiple map operations followed by a sort can lead to performance degradation, especially with large datasets.",
-    "recommendation": "Consider using a single iteration to merge and sort items to improve performance.",
-    "current_code": "const unifiedItems =[...memoriesAndSpaces.memories.map(...), ...spaces.map(...)]",
-    "suggested_code": "const unifiedItems =[...memoriesAndSpaces.memories, ...spaces].map(spaceOrMemory => ({item: spaceOrMemory.type, date: new Date(spaceOrMemory.date), data: spaceOrMemory})).sort((a, b) => a.date - b.date);",
-    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 98,
-    "end_line": 118,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Syntax and Logic Errors",
-    "description": "Potential null pointer dereference.",
-    "impact": "high",
-    "rationale": "The code assumes that 'currentSpace' and 'usersWithAccess' are always defined. If they are null or undefined, it could lead to runtime errors.",
-    "recommendation": "Add null checks for 'currentSpace' and 'usersWithAccess' before accessing their properties.",
-    "current_code": "usersWithAccess.length > 0",
-    "suggested_code": "usersWithAccess?.length > 0",
-    "file_path": "apps/web/app/(dash)/(memories)/content.tsx",
-    "start_line": 168,
-    "end_line": 168,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Error Handling and Logging",
-    "description": "Inadequate error handling in asynchronous functions.",
-    "impact": "high",
-    "rationale": "The current implementation of the `handleSubmit` function throws errors without proper handling, which can lead to unhandled promise rejections and poor user experience.",
-    "recommendation": "Implement try-catch blocks around asynchronous calls to handle errors gracefully.",
-    "current_code": "const cont = await createMemory({\n    content: content,\n    spaces: spaces ?? undefined,\n});",
-    "suggested_code": "try{\n    const cont = await createMemory({\n        content: content,\n        spaces: spaces ?? undefined,\n});\n}catch (error){\n    toast.error(`Error: ${error.message}`);\n    return;\n}",
-    "file_path": "apps/web/app/(dash)/menu.tsx",
-    "start_line": 213,
-    "end_line": 223,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Code Structure and Design",
-    "description": "Code duplication in the `handleSubmit` function and the form submission handler.",
-    "impact": "medium",
-    "rationale": "The logic for handling the form submission is repeated in both the `handleSubmit` function and the inline form submission handler, which violates the DRY principle and makes maintenance harder.",
-    "recommendation": "Extract the form submission logic into a separate function to avoid duplication.",
-    "current_code": "toast.promise(handleSubmit(content, selectedSpaces),{\n    loading: (...),\n    success: (data) => \"Memory queued\",\n    error: (error) => error.message,\n});",
-    "suggested_code": "const submitMemory = async () =>{\n    return toast.promise(handleSubmit(content, selectedSpaces),{\n        loading: (...),\n        success: (data) => \"Memory queued\",\n        error: (error) => error.message,\n});\n};\n\n// Use submitMemory in both places.",
-    "file_path": "apps/web/app/(dash)/menu.tsx",
-    "start_line": 186,
-    "end_line": 198,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Performance Issues",
-    "description": "Potential performance issue with frequent state updates in `handleSubmit`.",
-    "impact": "medium",
-    "rationale": "The state updates for `content` and `selectedSpaces` are done sequentially, which can lead to unnecessary re-renders. Using a functional update can optimize this.",
-    "recommendation": "Use functional updates for state setters to ensure they are based on the latest state.",
-    "current_code": "setContent(\"\");\nsetSelectedSpaces([]);",
-    "suggested_code": "setContent((prev) => \"\");\nsetSelectedSpaces((prev) =>[]);",
-    "file_path": "apps/web/app/(dash)/menu.tsx",
-    "start_line": 221,
-    "end_line": 223,
-    "sentiment": "neutral",
-    "severity": 4
-  },
-  {
-    "category": "Code Structure and Design",
-    "description": "Lack of separation of concerns in the `Menu` component.",
-    "impact": "medium",
-    "rationale": "The `Menu` component is handling multiple responsibilities, including state management and UI rendering. This can make the component harder to test and maintain.",
-    "recommendation": "Consider breaking the `Menu` component into smaller components, such as `MemoryForm` and `SpaceSelector`, to improve readability and maintainability.",
-    "current_code": "function Menu(){\n    // ...\n}",
-    "suggested_code": "function Menu(){\n    return <MemoryForm />;\n}\n\nfunction MemoryForm(){\n    // Logic for handling memory submission\n}",
-    "file_path": "apps/web/app/(dash)/menu.tsx",
-    "start_line": 35,
-    "end_line": 36,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Code Structure and Design",
-    "description": "Inconsistent naming conventions for props.",
-    "impact": "medium",
-    "rationale": "The props in the component have inconsistent naming conventions. Some props use camelCase while others use snake_case, which can lead to confusion and reduce readability.",
-    "recommendation": "Standardize the naming convention for props to camelCase to maintain consistency throughout the codebase.",
-    "current_code": "placeholder?: string;\nemptyMessage?: string;\ncreateNewMessage?: string;\nclassName?: string;",
-    "suggested_code": "placeholder?: string;\nemptyMessage?: string;\ncreateNewMessage?: string;\nclassName?: string;",
-    "file_path": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 23,
-    "end_line": 26,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Performance Issues",
-    "description": "Inefficient filtering of options.",
-    "impact": "high",
-    "rationale": "The current implementation filters options on every render, which can lead to performance degradation, especially with a large dataset. This could be optimized by memoizing the filtered options.",
-    "recommendation": "Use `useMemo` to memoize the filtered options based on the selected spaces and the original options.",
-    "current_code": "const filteredOptions = options.filter(\n\t(option) => !selectedSpaces.includes(parseInt(option.value)),\n);",
-    "suggested_code": "const filteredOptions = useMemo(() => options.filter(\n\t(option) => !selectedSpaces.includes(parseInt(option.value)),\n),[options, selectedSpaces]);",
-    "file_path": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 55,
-    "end_line": 57,
-    "sentiment": "neutral",
-    "severity": 7
-  },
-  {
-    "category": "Error Handling and Logging",
-    "description": "Lack of error handling for asynchronous operations.",
-    "impact": "high",
-    "rationale": "The onSubmit function is called asynchronously without any error handling, which can lead to unhandled promise rejections and a poor user experience.",
-    "recommendation": "Wrap the onSubmit call in a try-catch block to handle potential errors gracefully.",
-    "current_code": "onClick={() => onSubmit(inputValue)}",
-    "suggested_code": "onClick={async () =>{\n\ttry{\n\t\tawait onSubmit(inputValue);\n\t}catch (error){\n\t\tconsole.error('Error submitting:', error);\n\t}\n}}",
-    "file_path": "packages/ui/shadcn/combobox.tsx",
-    "start_line": 98,
-    "end_line": 98,
-    "sentiment": "negative",
-    "severity": 8
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/review.md b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/review.md
deleted file mode 100644
index 0403dc15..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_232/review.md
+++ /dev/null
@@ -1,33 +0,0 @@
-PR URL: https://github.com/supermemoryai/supermemory/pull/232
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 11
-- Critical: 0
-- Important: 0
-- Minor: 0
-- Files Affected: 4
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 15508, "completion_tokens": 2390, "total_tokens": 17898}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/comments.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/comments.json
deleted file mode 100644
index 0637a088..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/comments.json
+++ /dev/null
@@ -1 +0,0 @@
-[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/issues.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/issues.json
deleted file mode 100644
index 296cc3b7..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/issues.json
+++ /dev/null
@@ -1,58 +0,0 @@
-[
-  {
-    "category": "Code Structure and Design",
-    "description": "Inconsistent naming conventions for parameters.",
-    "impact": "medium",
-    "rationale": "Inconsistent naming can lead to confusion and make the code harder to read and maintain. For example, the parameter names in `generate_twitter_post` and `generate_linkedin_post` methods could be more consistent.",
-    "recommendation": "Use a consistent naming convention for parameters across similar methods.",
-    "current_code": "def generate_twitter_post(self, summary: Dict, user: Optional[str] = None):",
-    "suggested_code": "def generate_twitter_post(self, summary: Dict, user: Optional[str] = None):",
-    "file_path": "kaizen/reviewer/work_summarizer.py",
-    "start_line": 58,
-    "end_line": 58,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling and Logging",
-    "description": "Lack of error handling when parsing JSON data.",
-    "impact": "high",
-    "rationale": "If the JSON data is malformed, the application will raise an unhandled exception, which can lead to crashes. Proper error handling is essential to ensure robustness.",
-    "recommendation": "Wrap the JSON parsing in a try-except block to handle potential exceptions gracefully.",
-    "current_code": "parsed_data = json.loads(json_data)",
-    "suggested_code": "try:\n    parsed_data = json.loads(json_data)\nexcept json.JSONDecodeError as e:\n    logging.error(f'Failed to parse JSON:{e}')\n    return None",
-    "file_path": "kaizen/helpers/parser.py",
-    "start_line": 47,
-    "end_line": 48,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Performance Issues",
-    "description": "Potential inefficiency in string formatting.",
-    "impact": "medium",
-    "rationale": "Using f-strings for logging can be inefficient if the logging level is set to a level higher than the one being used. This can lead to unnecessary string formatting operations.",
-    "recommendation": "Use logging methods that include the level check, such as `logging.debug` instead of f-strings directly.",
-    "current_code": "print(f\" Work Summary: \n{summary}\n\")",
-    "suggested_code": "logging.debug(\" Work Summary: \n %s\", summary)",
-    "file_path": "examples/work_summarizer/main.py",
-    "start_line": 62,
-    "end_line": 62,
-    "sentiment": "neutral",
-    "severity": 4
-  },
-  {
-    "category": "Resource Management",
-    "description": "Unclosed resources in the context of file handling.",
-    "impact": "high",
-    "rationale": "If there are any file operations in the code (not shown in the provided patch), failing to close files can lead to resource leaks.",
-    "recommendation": "Ensure that all file operations are wrapped in a context manager (with statement) to guarantee closure.",
-    "current_code": "file = open('example.txt', 'r')",
-    "suggested_code": "with open('example.txt', 'r') as file:\n    data = file.read()",
-    "file_path": "kaizen/reviewer/code_review.py",
-    "start_line": 14,
-    "end_line": 14,
-    "sentiment": "negative",
-    "severity": 6
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/review.md b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/review.md
deleted file mode 100644
index 4e6e60f3..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_252/review.md
+++ /dev/null
@@ -1,33 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/252
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 4
-- Critical: 0
-- Important: 0
-- Minor: 0
-- Files Affected: 4
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 3757, "completion_tokens": 786, "total_tokens": 4543}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/comments.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/comments.json
deleted file mode 100644
index 0637a088..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/comments.json
+++ /dev/null
@@ -1 +0,0 @@
-[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/issues.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/issues.json
deleted file mode 100644
index 67c6897f..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/issues.json
+++ /dev/null
@@ -1,58 +0,0 @@
-[
-  {
-    "category": "Code Structure and Design",
-    "description": "Inconsistent naming conventions and prompt usage.",
-    "impact": "medium",
-    "rationale": "The change from `CODE_REVIEW_SYSTEM_PROMPT` to `PR_DESCRIPTION_SYSTEM_PROMPT` is a good improvement for clarity. However, there are still remnants of the old naming conventions that could lead to confusion.",
-    "recommendation": "Ensure all prompt variables follow a consistent naming convention that clearly reflects their purpose. Consider refactoring any remaining instances of old variable names.",
-    "current_code": "self.provider.system_prompt = CODE_REVIEW_SYSTEM_PROMPT",
-    "suggested_code": "self.provider.system_prompt = PR_DESCRIPTION_SYSTEM_PROMPT",
-    "file_path": "kaizen/generator/pr_description.py",
-    "start_line": 29,
-    "end_line": 29,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling and Logging",
-    "description": "Lack of specific exception handling.",
-    "impact": "high",
-    "rationale": "Raising a generic Exception can obscure the root cause of issues and complicate debugging. Specific exceptions provide clearer context for errors.",
-    "recommendation": "Use more specific exception types or create custom exceptions to provide better context. Additionally, consider logging the error details for easier troubleshooting.",
-    "current_code": "raise Exception(\"Both diff_text and pull_request_files are empty!\")",
-    "suggested_code": "raise ValueError(\"Both diff_text and pull_request_files are empty!\")",
-    "file_path": "kaizen/generator/pr_description.py",
-    "start_line": 51,
-    "end_line": 51,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Performance Issues",
-    "description": "Potential inefficiencies in processing logic.",
-    "impact": "medium",
-    "rationale": "The `_process_full_diff` and `_process_files` methods may be called in scenarios where they are not necessary, leading to unnecessary computations.",
-    "recommendation": "Implement checks to ensure that these methods are only called when required, potentially using early returns to simplify logic.",
-    "current_code": "if diff_text and self.provider.is_inside_token_limit(PROMPT=prompt):",
-    "suggested_code": "if not diff_text or not self.provider.is_inside_token_limit(PROMPT=prompt): return None",
-    "file_path": "kaizen/generator/pr_description.py",
-    "start_line": 53,
-    "end_line": 53,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Code Structure and Design",
-    "description": "Code duplication in prompt generation.",
-    "impact": "medium",
-    "rationale": "There are similar patterns in how prompts are generated across different methods, leading to code duplication.",
-    "recommendation": "Consider creating a utility function to handle prompt generation to reduce duplication and improve maintainability.",
-    "current_code": "prompt = PR_DESCRIPTION_PROMPT.format(...)\nprompt = MERGE_PR_DESCRIPTION_PROMPT.format(...)",
-    "suggested_code": "def generate_prompt(template, **kwargs): return template.format(**kwargs)\nprompt = generate_prompt(PR_DESCRIPTION_PROMPT, ...)",
-    "file_path": "kaizen/generator/pr_description.py",
-    "start_line": 45,
-    "end_line": 45,
-    "sentiment": "neutral",
-    "severity": 5
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/review.md b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/review.md
deleted file mode 100644
index 32e3df81..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_335/review.md
+++ /dev/null
@@ -1,33 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/335
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 4
-- Critical: 0
-- Important: 0
-- Minor: 0
-- Files Affected: 1
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 6622, "completion_tokens": 774, "total_tokens": 7396}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/comments.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/comments.json
deleted file mode 100644
index 252fbeac..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/comments.json
+++ /dev/null
@@ -1,30 +0,0 @@
-[
-  {
-    "category": "Error Handling and Logging",
-    "description": "Lack of input validation for create_folder function.",
-    "impact": "critical",
-    "rationale": "The create_folder function should validate inputs to prevent unexpected behavior or crashes. The tests now include checks for empty and invalid paths, which is a good start, but the function itself should also handle these cases.",
-    "recommendation": "Implement input validation in the create_folder function to ensure that it handles invalid inputs gracefully.",
-    "current_code": "def create_folder(folder_path):\n    # Function logic here",
-    "suggested_code": "def create_folder(folder_path):\n    if not folder_path:\n        raise ValueError('Folder path cannot be empty')\n    # Function logic here",
-    "file_path": "kaizen/helpers/output.py",
-    "start_line": 1,
-    "end_line": 1,
-    "sentiment": "negative",
-    "severity": 9
-  },
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "11",
-    "end_line": "11",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/issues.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/issues.json
deleted file mode 100644
index 548c5215..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/issues.json
+++ /dev/null
@@ -1,338 +0,0 @@
-[
-  {
-    "category": "Code Structure and Design",
-    "description": "Inconsistent use of mocking in tests.",
-    "impact": "high",
-    "rationale": "The original code used decorators for mocking which can lead to confusion and inconsistency in test setup. The new approach uses fixtures, which improves readability and maintainability by clearly defining the mock setup in one place.",
-    "recommendation": "Continue using fixtures for mocking to maintain consistency across tests. This will also help in reducing boilerplate code.",
-    "current_code": "@mock.patch('kaizen.helpers.output.os.makedirs')\n@mock.patch('kaizen.helpers.output.os.path.exists')\ndef test_create_folder_success(mock_logger, mock_exists, mock_makedirs):",
-    "suggested_code": "@pytest.fixture\ndef mock_os_path_exists():\n    with mock.patch('os.path.exists') as mock_exists:\n        yield mock_exists",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_folder.py",
-    "start_line": 6,
-    "end_line": 8,
-    "sentiment": "positive",
-    "severity": 7
-  },
-  {
-    "category": "Error Handling and Logging",
-    "description": "Lack of input validation for create_folder function.",
-    "impact": "critical",
-    "rationale": "The create_folder function should validate inputs to prevent unexpected behavior or crashes. The tests now include checks for empty and invalid paths, which is a good start, but the function itself should also handle these cases.",
-    "recommendation": "Implement input validation in the create_folder function to ensure that it handles invalid inputs gracefully.",
-    "current_code": "def create_folder(folder_path):\n    # Function logic here",
-    "suggested_code": "def create_folder(folder_path):\n    if not folder_path:\n        raise ValueError('Folder path cannot be empty')\n    # Function logic here",
-    "file_path": "kaizen/helpers/output.py",
-    "start_line": 1,
-    "end_line": 1,
-    "sentiment": "negative",
-    "severity": 9
-  },
-  {
-    "category": "Performance Issues",
-    "description": "Potential performance issue with deep folder creation.",
-    "impact": "medium",
-    "rationale": "Creating deeply nested folders can lead to performance degradation if not handled properly. The tests should ensure that the function can handle such cases efficiently.",
-    "recommendation": "Consider implementing a method to create nested directories in a single call if the underlying OS supports it, or handle it in a way that minimizes performance hits.",
-    "current_code": "create_folder(folder_path)",
-    "suggested_code": "os.makedirs(folder_path, exist_ok=True)",
-    "file_path": "kaizen/helpers/output.py",
-    "start_line": 1,
-    "end_line": 1,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Code Structure and Design",
-    "description": "Duplication of test cases.",
-    "impact": "medium",
-    "rationale": "There are multiple test cases that check similar functionality with slight variations. This can lead to maintenance challenges and bloated test suites.",
-    "recommendation": "Consolidate similar test cases into parameterized tests to improve maintainability and clarity.",
-    "current_code": "def test_create_pr_description(desc, original_desc, expected):\n    assert create_pr_description(desc, original_desc) == expected",
-    "suggested_code": "@pytest.mark.parametrize('desc, original_desc, expected',[\n    ('This is a test description.', 'This is the original description.', 'Expected output'),\n    # Other cases\n])\ndef test_create_pr_description_normal_and_edge_cases(desc, original_desc, expected):\n    assert create_pr_description(desc, original_desc) == expected",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py",
-    "start_line": 8,
-    "end_line": 8,
-    "sentiment": "positive",
-    "severity": 5
-  },
-  {
-    "category": "Code Structure and Design",
-    "description": "Inconsistent naming conventions for severity levels.",
-    "impact": "medium",
-    "rationale": "The code uses both 'severity_level' and 'severity' interchangeably, which can lead to confusion and inconsistency in the codebase.",
-    "recommendation": "Standardize the naming convention for severity across the codebase.",
-    "current_code": "\"severity_level\": 9,",
-    "suggested_code": "\"severity\": 9,",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 30,
-    "end_line": 30,
-    "sentiment": "negative",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling and Logging",
-    "description": "Missing error handling for potential issues in input data.",
-    "impact": "high",
-    "rationale": "The current implementation does not validate the input data structure or handle cases where expected fields are missing, which could lead to runtime errors.",
-    "recommendation": "Implement input validation to ensure all required fields are present before processing.",
-    "current_code": "def test_reviews_with_missing_fields():",
-    "suggested_code": "def test_reviews_with_missing_fields():\n    # Validate input data structure here\n    assert 'comment' in review and 'reason' in review, 'Missing required fields'",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 144,
-    "end_line": 144,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Performance Issues",
-    "description": "Repeated use of PR_COLLAPSIBLE_TEMPLATE can lead to performance overhead.",
-    "impact": "medium",
-    "rationale": "The template is being formatted multiple times within loops, which could be optimized by caching or reusing the formatted string.",
-    "recommendation": "Consider pre-formatting the template outside of loops or using a more efficient string handling method.",
-    "current_code": "PR_COLLAPSIBLE_TEMPLATE.format(...)",
-    "suggested_code": "formatted_template = PR_COLLAPSIBLE_TEMPLATE.format(...)\n# Use formatted_template in the loop",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 85,
-    "end_line": 85,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Readability",
-    "description": "Use of inline comments to explain complex logic.",
-    "impact": "low",
-    "rationale": "While the code is generally readable, adding comments to explain the purpose of complex blocks can improve maintainability.",
-    "recommendation": "Add inline comments to clarify the purpose of key sections of the code.",
-    "current_code": "def test_multiple_topics_multiple_reviews(setup_multiple_topics_multiple_reviews):",
-    "suggested_code": "def test_multiple_topics_multiple_reviews(setup_multiple_topics_multiple_reviews):\n    # Test case for multiple topics and reviews",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py",
-    "start_line": 99,
-    "end_line": 99,
-    "sentiment": "positive",
-    "severity": 3
-  },
-  {
-    "category": "Code Structure and Design",
-    "description": "Inconsistent naming conventions for importance levels.",
-    "impact": "medium",
-    "rationale": "The importance levels in the JSON structure are sometimes written as 'High' and other times as 'high'. This inconsistency can lead to confusion and bugs when processing these values.",
-    "recommendation": "Standardize the naming convention for importance levels across all tests.",
-    "current_code": "\"importance\": \"High\"",
-    "suggested_code": "\"importance\": \"high\"",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 32,
-    "end_line": 32,
-    "sentiment": "negative",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling and Logging",
-    "description": "Lack of logging for error cases in file operations.",
-    "impact": "high",
-    "rationale": "While the tests handle exceptions, there is no logging to capture the context of the errors, making it harder to debug issues when they arise in production.",
-    "recommendation": "Add logging statements to capture exceptions and provide context for failures.",
-    "current_code": "with pytest.raises(PermissionError):",
-    "suggested_code": "try:\n    create_test_files(json_tests, tmp_path)\nexcept PermissionError as e:\n    mock_logger.error(f'Permission error:{e}')\n    raise",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 194,
-    "end_line": 195,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Performance Issues",
-    "description": "Inefficient handling of long test names.",
-    "impact": "medium",
-    "rationale": "The test for very long test names does not account for potential performance issues when generating and checking file names. This could lead to unnecessary resource consumption.",
-    "recommendation": "Implement a check to limit the length of generated test names before creating files.",
-    "current_code": "long_test_name = \"Test \" + \"Example \" * 50",
-    "suggested_code": "long_test_name = \"Test Example Long Name\"[:250]",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 131,
-    "end_line": 131,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Code Structure and Design",
-    "description": "Redundant mock patches in tests.",
-    "impact": "medium",
-    "rationale": "The same mock patches are repeated across multiple tests, leading to code duplication and increased maintenance effort.",
-    "recommendation": "Use a fixture to apply common mock patches to reduce redundancy.",
-    "current_code": "@mock.patch(\"kaizen.helpers.output.create_folder\")\n@mock.patch(\"kaizen.helpers.output.general.clean_python_code\")\n@mock.patch(\"kaizen.helpers.output.logger\")",
-    "suggested_code": "@pytest.fixture\ndef mock_dependencies():\n    with mock.patch(\"kaizen.helpers.output.create_folder\") as mock_create_folder,\n         mock.patch(\"kaizen.helpers.output.general.clean_python_code\") as mock_clean_python_code,\n         mock.patch(\"kaizen.helpers.output.logger\") as mock_logger:\n        yield mock_create_folder, mock_clean_python_code, mock_logger",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py",
-    "start_line": 48,
-    "end_line": 50,
-    "sentiment": "neutral",
-    "severity": 4
-  },
-  {
-    "category": "Code Structure and Design",
-    "description": "Inconsistent use of async/await in test functions.",
-    "impact": "high",
-    "rationale": "Some test functions are defined as async but do not use await properly, which can lead to unexpected behavior and test failures.",
-    "recommendation": "Ensure all async functions use await where necessary to handle asynchronous calls correctly.",
-    "current_code": "result = get_web_html(url)",
-    "suggested_code": "result = await get_web_html(url)",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 92,
-    "end_line": 92,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Error Handling and Logging",
-    "description": "Lack of error handling for network errors in async tests.",
-    "impact": "high",
-    "rationale": "The tests do not handle exceptions that may arise from network calls, which can lead to false negatives in test results.",
-    "recommendation": "Add exception handling in the async tests to ensure that network errors are properly caught and asserted.",
-    "current_code": "result = await get_web_html(url)",
-    "suggested_code": "with pytest.raises(Exception, match='Network error'):\n    await get_web_html(url)",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 88,
-    "end_line": 90,
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Performance Issues",
-    "description": "Potential inefficiency in handling large HTML content.",
-    "impact": "medium",
-    "rationale": "The test for large HTML content may lead to performance issues if the content size grows significantly, affecting test execution time.",
-    "recommendation": "Consider using a smaller sample size or mocking the HTML content for performance-sensitive tests.",
-    "current_code": "large_html_content = '<html><body>' + '<p>Test</p>' * 10000 + '</body></html>'",
-    "suggested_code": "large_html_content = '<html><body><p>Test</p></body></html>'",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 94,
-    "end_line": 95,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Code Structure and Design",
-    "description": "Use of hard-coded values in tests.",
-    "impact": "medium",
-    "rationale": "Hard-coded values can lead to maintenance issues and reduce the flexibility of the tests.",
-    "recommendation": "Define constants for commonly used values to improve maintainability and readability.",
-    "current_code": "url = 'https://cloudcode.ai'",
-    "suggested_code": "BASE_URL = 'https://cloudcode.ai'\nurl = BASE_URL",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "start_line": 48,
-    "end_line": 48,
-    "sentiment": "positive",
-    "severity": 4
-  },
-  {
-    "category": "Code Structure and Design",
-    "description": "Inconsistent naming conventions for constants.",
-    "impact": "medium",
-    "rationale": "Constants like SUPPORTED_LANGUAGES and LANGUAGE_PROMPTS should follow a consistent naming convention, typically ALL_CAPS, to improve readability and maintainability.",
-    "recommendation": "Rename constants to follow the ALL_CAPS convention for clarity.",
-    "current_code": "SUPPORTED_LANGUAGES ={...}",
-    "suggested_code": "SUPPORTED_LANGUAGES ={...}",
-    "file_path": "kaizen/generator/unit_test.py",
-    "start_line": 24,
-    "end_line": 40,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Error Handling and Logging",
-    "description": "Lack of error handling in file operations.",
-    "impact": "high",
-    "rationale": "File operations can fail due to various reasons (e.g., file not found, permission issues). Proper error handling should be implemented to avoid crashes and provide meaningful feedback.",
-    "recommendation": "Wrap file operations in try-except blocks to handle potential exceptions gracefully.",
-    "current_code": "with open(file_path, 'r') as file: ...",
-    "suggested_code": "try:\n    with open(file_path, 'r') as file:\n        return file.read()\nexcept IOError as e:\n    self.logger.error(f'Error reading file{file_path}:{e}')\n    return None",
-    "file_path": "kaizen/generator/unit_test.py",
-    "start_line": 108,
-    "end_line": 110,
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Performance Issues",
-    "description": "Potential inefficiency in string operations.",
-    "impact": "medium",
-    "rationale": "Using string concatenation in a loop can lead to performance issues, especially with large datasets. It's better to use a list and join at the end.",
-    "recommendation": "Change string concatenation to use a list and join method for better performance.",
-    "current_code": "formatted_scenarios += f'{category.replace('_', ' ').title()}:\n'\nformatted_scenarios += '\n'.join(f'-{case}' for case in cases)",
-    "suggested_code": "formatted_scenarios =[]\nformatted_scenarios.append(f'{category.replace('_', ' ').title()}:\n')\nformatted_scenarios.append('\n'.join(f'-{case}' for case in cases))\nreturn ''.join(formatted_scenarios)",
-    "file_path": "kaizen/generator/unit_test.py",
-    "start_line": 240,
-    "end_line": 250,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Error Handling and Logging",
-    "description": "Inconsistent logging practices.",
-    "impact": "medium",
-    "rationale": "Some methods log their actions while others do not, leading to inconsistent behavior and difficulty in tracing execution flow.",
-    "recommendation": "Ensure that all significant actions, especially those that could fail, are logged consistently.",
-    "current_code": "print(f'Error: Could not generate tests for{file_path}:{e}')",
-    "suggested_code": "self.logger.error(f'Could not generate tests for{file_path}:{e}')",
-    "file_path": "kaizen/generator/unit_test.py",
-    "start_line": 73,
-    "end_line": 74,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Code Structure and Design",
-    "description": "Hardcoded logger settings and lack of configurability.",
-    "impact": "high",
-    "rationale": "The function `set_all_loggers_to_ERROR` hardcodes the log level for specific loggers, which reduces flexibility and makes it difficult to change logging behavior without modifying the code.",
-    "recommendation": "Consider using a configuration file or environment variables to manage logger settings dynamically.",
-    "current_code": "logging.getLogger(\"LiteLLM\").setLevel(logging.ERROR)",
-    "suggested_code": "logging.getLogger(\"LiteLLM\").setLevel(os.environ.get(\"LITELLM_LOG_LEVEL\", logging.ERROR))",
-    "file_path": "kaizen/llms/provider.py",
-    "start_line": 25,
-    "end_line": 27,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Error Handling and Logging",
-    "description": "Lack of error handling when setting logger levels.",
-    "impact": "medium",
-    "rationale": "If there are issues with setting the logger levels (e.g., invalid logger names), the application will fail without any meaningful error message.",
-    "recommendation": "Add try-except blocks around logger level settings to handle potential exceptions gracefully.",
-    "current_code": "logging.getLogger(name).setLevel(logging.ERROR)",
-    "suggested_code": "try:\n    logging.getLogger(name).setLevel(logging.ERROR)\nexcept Exception as e:\n    print(f\"Failed to set logger level for{name}:{e}\")",
-    "file_path": "kaizen/llms/provider.py",
-    "start_line": 18,
-    "end_line": 18,
-    "sentiment": "negative",
-    "severity": 6
-  },
-  {
-    "category": "Readability",
-    "description": "Inconsistent naming conventions.",
-    "impact": "medium",
-    "rationale": "The function name `set_all_loggers_to_ERROR` uses snake_case, while the logger names use CamelCase. This inconsistency can lead to confusion.",
-    "recommendation": "Standardize naming conventions across the codebase to improve readability and maintainability.",
-    "current_code": "set_all_loggers_to_ERROR()",
-    "suggested_code": "set_all_loggers_to_error()",
-    "file_path": "kaizen/llms/provider.py",
-    "start_line": 23,
-    "end_line": 23,
-    "sentiment": "neutral",
-    "severity": 4
-  },
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "11",
-    "end_line": "11",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/review.md b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/review.md
deleted file mode 100644
index 2cffb7a1..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_400/review.md
+++ /dev/null
@@ -1,81 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/400
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 24
-- Critical: 2
-- Important: 0
-- Minor: 0
-- Files Affected: 9
-## 🏆 Code Quality
-[███████████████░░░░░] 75% (Fair)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Error Handling and Logging (2 issues)</strong></summary>
-
-### 1. Lack of input validation for create_folder function.
-📁 **File:** `kaizen/helpers/output.py:1`
-⚖️ **Severity:** 9/10
-🔍 **Description:** Lack of input validation for create_folder function.
-💡 **Solution:** 
-
-**Current Code:**
-```python
-def create_folder(folder_path):
-    # Function logic here
-```
-
-**Suggested Code:**
-```python
-
-```
-
-### 2. Changes made to sensitive file
-📁 **File:** `config.json:11`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (1 issues)</strong></summary>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 36120, "completion_tokens": 4596, "total_tokens": 40716}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/comments.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/comments.json
deleted file mode 100644
index 0637a088..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/comments.json
+++ /dev/null
@@ -1 +0,0 @@
-[]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/issues.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/issues.json
deleted file mode 100644
index 65970832..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/issues.json
+++ /dev/null
@@ -1,44 +0,0 @@
-[
-  {
-    "category": "Code Structure and Design",
-    "description": "Potential for duplicate embeddings in the database.",
-    "impact": "high",
-    "rationale": "The comment 'TODO: DONT PUSH DUPLICATE' indicates that there is a risk of inserting duplicate embeddings into the database. This can lead to data inconsistency and increased storage usage.",
-    "recommendation": "Implement a check before inserting embeddings to ensure duplicates are not added.",
-    "current_code": "embedding_query = text(\n                  \"\"\"\n",
-    "suggested_code": "if not self.is_duplicate_embedding(embedding):\n    # Proceed with insertion logic here\n",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 157,
-    "end_line": 157,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Error Handling and Logging",
-    "description": "Lack of error handling for repository setup.",
-    "impact": "medium",
-    "rationale": "The `setup_repository` method is called without any error handling. If this operation fails, it could lead to unhandled exceptions and application crashes.",
-    "recommendation": "Wrap the `setup_repository` call in a try-except block to handle potential exceptions gracefully.",
-    "current_code": "analyzer.setup_repository(\"./github_app/\")",
-    "suggested_code": "try:\n    analyzer.setup_repository(\"./github_app/\")\nexcept Exception as e:\n    logging.error(f\"Failed to set up repository:{e}\")\n",
-    "file_path": "examples/ragify_codebase/main.py",
-    "start_line": 7,
-    "end_line": 7,
-    "sentiment": "negative",
-    "severity": 6
-  },
-  {
-    "category": "Performance Issues",
-    "description": "Potential inefficiency in embedding retrieval.",
-    "impact": "medium",
-    "rationale": "The line `embedding = embedding[0][\"embedding\"]` assumes that the embedding list will always have at least one element. If it doesn't, this could lead to an index error.",
-    "recommendation": "Check if the embedding list is not empty before accessing its first element.",
-    "current_code": "embedding = embedding[0][\"embedding\"]",
-    "suggested_code": "if embedding:\n    embedding = embedding[0][\"embedding\"]\nelse:\n    raise ValueError(\"No embeddings found\")\n",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 155,
-    "end_line": 155,
-    "sentiment": "negative",
-    "severity": 5
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/review.md b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/review.md
deleted file mode 100644
index 3219c661..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_440/review.md
+++ /dev/null
@@ -1,33 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/440
-
-# 🔍 Code Review Summary
-
-✅ **All Clear:** This commit looks good! 👍
-
-## 📊 Stats
-- Total Issues: 3
-- Critical: 0
-- Important: 0
-- Minor: 0
-- Files Affected: 2
-## 🏆 Code Quality
-[███████████████░░░░░] 75% (Fair)
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 1235, "completion_tokens": 619, "total_tokens": 1854}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/comments.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/comments.json
deleted file mode 100644
index 9cbf8d49..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/comments.json
+++ /dev/null
@@ -1,16 +0,0 @@
-[
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/issues.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/issues.json
deleted file mode 100644
index 28b2c5e9..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/issues.json
+++ /dev/null
@@ -1,72 +0,0 @@
-[
-  {
-    "category": "Error Handling and Logging",
-    "description": "Broad exception handling without specific error management.",
-    "impact": "high",
-    "rationale": "Using a generic `except Exception` can mask underlying issues and makes debugging difficult. It\u2019s crucial to handle specific exceptions to ensure that errors are logged and managed appropriately.",
-    "recommendation": "Refine the exception handling to catch specific exceptions and log meaningful error messages.",
-    "current_code": "except Exception:\n    print(\"Error\")",
-    "suggested_code": "except KeyError:\n    raise ValueError(f\"Invalid confidence level:{min_confidence}\")\nexcept Exception as e:\n    logger.error(f\"Unexpected error:{str(e)}\")",
-    "file_path": "github_app/github_helper/pull_requests.py",
-    "start_line": 140,
-    "end_line": 141,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Code Structure and Design",
-    "description": "Potential code duplication in sorting logic.",
-    "impact": "medium",
-    "rationale": "The sorting logic for files appears to be implemented in a way that could lead to duplication if similar logic is used elsewhere. This can lead to maintenance challenges.",
-    "recommendation": "Consider using Python's built-in sorting capabilities or refactor the sorting logic into a separate utility function to promote reusability.",
-    "current_code": "for file in files:\n    min_index = len(sorted_files)\n    file_name = file[\"filename\"]\n    for i, sorted_file in enumerate(sorted_files):\n        if file_name < sorted_file[\"filename\"]:\n            min_index = i\n            break\n    sorted_files.insert(min_index, file)",
-    "suggested_code": "sorted_files = sorted(files, key=lambda f: f[\"filename\"])",
-    "file_path": "github_app/github_helper/pull_requests.py",
-    "start_line": 184,
-    "end_line": 196,
-    "sentiment": "neutral",
-    "severity": 5
-  },
-  {
-    "category": "Performance Issues",
-    "description": "Inefficient file sorting algorithm.",
-    "impact": "medium",
-    "rationale": "The current sorting algorithm has a time complexity of O(n^2) due to the nested loops. This can lead to performance issues with larger datasets.",
-    "recommendation": "Use Python's built-in sorting functions which are optimized and have a time complexity of O(n log n).",
-    "current_code": "for file in files:\n    min_index = len(sorted_files)\n    for i, sorted_file in enumerate(sorted_files):\n        if file_name < sorted_file[\"filename\"]:\n            min_index = i\n            break\n    sorted_files.insert(min_index, file)",
-    "suggested_code": "sorted_files = sorted(files, key=lambda f: f[\"filename\"])",
-    "file_path": "github_app/github_helper/pull_requests.py",
-    "start_line": 184,
-    "end_line": 196,
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Code Structure and Design",
-    "description": "Inconsistent naming conventions.",
-    "impact": "low",
-    "rationale": "The naming of variables such as `pr_files` and `pull_request_files` can lead to confusion. Consistent naming conventions improve readability and maintainability.",
-    "recommendation": "Standardize variable names throughout the codebase to follow a consistent naming convention.",
-    "current_code": "pull_request_files=pr_files",
-    "suggested_code": "pull_request_files=sorted_files",
-    "file_path": "github_app/github_helper/pull_requests.py",
-    "start_line": 55,
-    "end_line": 56,
-    "sentiment": "positive",
-    "severity": 4
-  },
-  {
-    "category": "Configuration",
-    "description": "Changes made to sensitive file",
-    "impact": "critical",
-    "recommendation": "Changes were made to config.json, which needs review",
-    "current_code": "NA",
-    "fixed_code": "",
-    "start_line": "1",
-    "end_line": "1",
-    "side": "RIGHT",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/review.md b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/review.md
deleted file mode 100644
index 30369f46..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_476/review.md
+++ /dev/null
@@ -1,64 +0,0 @@
-PR URL: https://github.com/Cloud-Code-AI/kaizen/pull/476
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 5
-- Critical: 1
-- Important: 0
-- Minor: 0
-- Files Affected: 2
-## 🏆 Code Quality
-[█████████████████░░░] 85% (Good)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Configuration (1 issues)</strong></summary>
-
-### 1. Changes made to sensitive file
-📁 **File:** `config.json:1`
-⚖️ **Severity:** 10/10
-🔍 **Description:** Changes made to sensitive file
-💡 **Solution:** 
-
-**Current Code:**
-```python
-NA
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (1 issues)</strong></summary>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 3878, "completion_tokens": 867, "total_tokens": 4745}
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/comments.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/comments.json
deleted file mode 100644
index 26261207..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/comments.json
+++ /dev/null
@@ -1,16 +0,0 @@
-[
-  {
-    "category": "Error Handling and Logging",
-    "description": "Potential division by zero error.",
-    "impact": "critical",
-    "rationale": "If 'total_tokens' is zero, it will cause a division by zero error when calculating the summary statistics.",
-    "recommendation": "Add a check to ensure 'total_tokens' is not zero before performing calculations.",
-    "current_code": "print(f'Total tokens used:{total_tokens:,}')",
-    "suggested_code": "if total_tokens > 0:\n    print(f'Total tokens used:{total_tokens:,}')\nelse:\n    print('No tokens used.')",
-    "file_path": "main.py",
-    "start_line": 159,
-    "end_line": 161,
-    "sentiment": "negative",
-    "severity": 9
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/issues.json b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/issues.json
deleted file mode 100644
index f8fd1693..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/issues.json
+++ /dev/null
@@ -1,72 +0,0 @@
-[
-  {
-    "category": "Code Structure and Design",
-    "description": "Unused import statement.",
-    "impact": "low",
-    "rationale": "The import of 'random' is unnecessary and can clutter the codebase, potentially leading to confusion.",
-    "recommendation": "Remove the unused import statement to improve code clarity.",
-    "current_code": "import random  # Unused import",
-    "suggested_code": "",
-    "file_path": "main.py",
-    "start_line": 8,
-    "end_line": 8,
-    "sentiment": "negative",
-    "severity": 3
-  },
-  {
-    "category": "Error Handling and Logging",
-    "description": "Silent failure during JSON parsing.",
-    "impact": "high",
-    "rationale": "The code silently fails when JSON parsing fails, which can lead to undetected issues in the application. Proper logging or exception handling should be implemented.",
-    "recommendation": "Log the error message or raise an exception to ensure issues are visible during execution.",
-    "current_code": "except json.JSONDecodeError:\n        # Critical: Silent failure without logging",
-    "suggested_code": "except json.JSONDecodeError as e:\n        print(f'Failed to parse content for applicant:{e}')",
-    "file_path": "main.py",
-    "start_line": 82,
-    "end_line": 84,
-    "sentiment": "negative",
-    "severity": 7
-  },
-  {
-    "category": "Error Handling and Logging",
-    "description": "Potential division by zero error.",
-    "impact": "critical",
-    "rationale": "If 'total_tokens' is zero, it will cause a division by zero error when calculating the summary statistics.",
-    "recommendation": "Add a check to ensure 'total_tokens' is not zero before performing calculations.",
-    "current_code": "print(f'Total tokens used:{total_tokens:,}')",
-    "suggested_code": "if total_tokens > 0:\n    print(f'Total tokens used:{total_tokens:,}')\nelse:\n    print('No tokens used.')",
-    "file_path": "main.py",
-    "start_line": 159,
-    "end_line": 161,
-    "sentiment": "negative",
-    "severity": 9
-  },
-  {
-    "category": "Error Handling and Logging",
-    "description": "Lack of error handling for file not found.",
-    "impact": "high",
-    "rationale": "If the user inputs a file name that does not exist, the program will crash without providing a user-friendly message.",
-    "recommendation": "Implement a try-except block around the file reading operation to handle file not found errors gracefully.",
-    "current_code": "main(input_file)",
-    "suggested_code": "try:\n    main(input_file)\nexcept FileNotFoundError:\n    print('Error: The specified file was not found.')",
-    "file_path": "main.py",
-    "start_line": 175,
-    "end_line": 175,
-    "sentiment": "negative",
-    "severity": 8
-  },
-  {
-    "category": "Performance Issues",
-    "description": "Inefficient way to print progress.",
-    "impact": "medium",
-    "rationale": "The current method of printing progress can be inefficient, especially with larger datasets, as it updates the console frequently.",
-    "recommendation": "Consider using a logging library or a more efficient method to report progress.",
-    "current_code": "print(f'\\rProgress:[{('=' * int(50 * progress)):<50}]{progress:.0%}', end='', flush=True)",
-    "suggested_code": "if index % 10 == 0:  # Update every 10 applicants\n    print(f'Progress:[{('=' * int(50 * progress)):<50}]{progress:.0%}')",
-    "file_path": "main.py",
-    "start_line": 121,
-    "end_line": 122,
-    "sentiment": "neutral",
-    "severity": 5
-  }
-]
\ No newline at end of file
diff --git a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/review.md b/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/review.md
deleted file mode 100644
index a1e7c56a..00000000
--- a/.experiments/code_review/code_reviews_20240827_165253/no_eval/pr_5/review.md
+++ /dev/null
@@ -1,64 +0,0 @@
-PR URL: https://github.com/sauravpanda/applicant-screening/pull/5
-
-# 🔍 Code Review Summary
-
-❗ **Attention Required:** This push has potential issues. 🚨
-
-## 📊 Stats
-- Total Issues: 5
-- Critical: 1
-- Important: 0
-- Minor: 0
-- Files Affected: 1
-## 🏆 Code Quality
-[███████████████░░░░░] 75% (Fair)
-
-## 🚨 Critical Issues
-
-<details>
-<summary><strong>Error Handling and Logging (1 issues)</strong></summary>
-
-### 1. Potential division by zero error.
-📁 **File:** `main.py:159`
-⚖️ **Severity:** 9/10
-🔍 **Description:** Potential division by zero error.
-💡 **Solution:** 
-
-**Current Code:**
-```python
-print(f'Total tokens used:{total_tokens:,}')
-```
-
-**Suggested Code:**
-```python
-
-```
-
-</details>
-
-## 📝 Minor Notes
-Additional small points that you might want to consider:
-
-<details>
-<summary><strong>Click to expand (1 issues)</strong></summary>
-
-</details>
-
----
-
-> ✨ Generated with love by [Kaizen](https://cloudcode.ai) ❤️
-
-<details>
-<summary>Useful Commands</summary>
-
-- **Feedback:** Reply with `!feedback [your message]`
-- **Ask PR:** Reply with `!ask-pr [your question]`
-- **Review:** Reply with `!review`
-- **Explain:** Reply with `!explain [issue number]` for more details on a specific issue
-- **Ignore:** Reply with `!ignore [issue number]` to mark an issue as false positive
-- **Update Tests:** Reply with `!unittest` to create a PR with test changes
-</details>
-
-
------ Cost Usage (gpt-4o-mini)
-{"prompt_tokens": 6025, "completion_tokens": 900, "total_tokens": 6925}
\ No newline at end of file
diff --git a/.experiments/code_review/dataset/pr_222/issues.json b/.experiments/code_review/dataset/pr_222/issues.json
index 61607c58..be109f8d 100644
--- a/.experiments/code_review/dataset/pr_222/issues.json
+++ b/.experiments/code_review/dataset/pr_222/issues.json
@@ -1,19 +1,4 @@
 [
-  {
-    "category": "Security",
-    "description": "Hardcoding API keys in `config.json` can lead to security vulnerabilities.",
-    "impact": "critical",
-    "rationale": "Exposing API keys in the codebase can lead to unauthorized access. This was highlighted by multiple models and is a critical security issue.",
-    "recommendation": "Use environment variables to store API keys instead of hardcoding them.",
-    "suggested_code": "\"api_key\": \"os.environ/AZURE_API_KEY\"",
-    "fixed_code": "\"api_key\": \"${AZURE_API_KEY}\"",
-    "file_path": "config.json",
-    "start_line": 13,
-    "end_line": 13,
-    "side": "RIGHT",
-    "sentiment": "negative",
-    "severity": 9
-  },
   {
     "category": "SQL Injection",
     "description": "Potential SQL injection vulnerability in the query construction.",
@@ -74,21 +59,6 @@
     "sentiment": "neutral",
     "severity": 5
   },
-  {
-    "category": "Documentation",
-    "description": "Lack of comments and docstrings in complex functions and methods.",
-    "impact": "high",
-    "rationale": "Several models noted the need for better documentation to improve code understanding and maintainability.",
-    "recommendation": "Add comments explaining complex logic and docstrings to functions and methods.",
-    "suggested_code": "",
-    "fixed_code": "",
-    "file_path": "kaizen/retriever/llama_index_retriever.py",
-    "start_line": 1,
-    "end_line": 1,
-    "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity": 6
-  },
   {
     "category": "Code Duplication",
     "description": "Duplicate code found in test cases and database connection string creation.",
@@ -101,22 +71,7 @@
     "start_line": 98,
     "end_line": 101,
     "side": "RIGHT",
-    "sentiment": "neutral",
-    "severity": 6
-  },
-  {
-    "category": "Dependency Management",
-    "description": "New dependencies added without justification.",
-    "impact": "high",
-    "rationale": "Adding dependencies increases the attack surface and maintenance burden. This was noted as an important consideration by multiple models.",
-    "recommendation": "Review and justify new dependencies, ensuring they are necessary and compatible.",
-    "suggested_code": "",
-    "fixed_code": "",
-    "file_path": "pyproject.toml",
-    "start_line": 27,
-    "end_line": 49,
-    "side": "RIGHT",
-    "sentiment": "neutral",
+    "sentiment": "negative",
     "severity": 6
   },
   {
diff --git a/.experiments/code_review/dataset/pr_252/issues.json b/.experiments/code_review/dataset/pr_252/issues.json
index 1f0c3e8e..74b266ec 100644
--- a/.experiments/code_review/dataset/pr_252/issues.json
+++ b/.experiments/code_review/dataset/pr_252/issues.json
@@ -2,7 +2,7 @@
   {
     "category": "Code Structure and Consistency",
     "description": "There are inconsistencies in code formatting and structure across different function calls.",
-    "impact": "medium",
+    "impact": "high",
     "rationale": "Consistent code structure and formatting improves readability and maintainability. This issue was noted by multiple models.",
     "recommendation": "Standardize the formatting of function calls, particularly for `generate_twitter_post` and `generate_linkedin_post`. Consider using multi-line formatting for both for consistency.",
     "suggested_code": "twitter_post = work_summary_generator.generate_twitter_post(summary, user=\"oss_example\")\n\nlinkedin_post = work_summary_generator.generate_linkedin_post(\n    summary, user=\"oss_example\"\n)",
@@ -11,7 +11,7 @@
     "start_line": 59,
     "end_line": 62,
     "side": "RIGHT",
-    "sentiment": "neutral",
+    "sentiment": "negative",
     "severity": 4
   },
   {
@@ -26,7 +26,7 @@
     "start_line": 0,
     "end_line": 0,
     "side": "RIGHT",
-    "sentiment": "neutral",
+    "sentiment": "negative",
     "severity": 6
   },
   {
@@ -73,20 +73,5 @@
     "side": "RIGHT",
     "sentiment": "neutral",
     "severity": 4
-  },
-  {
-    "category": "Prompt Formatting",
-    "description": "The TWITTER_POST_PROMPT and LINKEDIN_POST_PROMPT could be improved for better readability.",
-    "impact": "medium",
-    "rationale": "Well-formatted prompts are easier to read and maintain.",
-    "recommendation": "Break the prompts into multiple lines, use string formatting, and add comments to explain different sections.",
-    "suggested_code": "",
-    "fixed_code": "",
-    "file_path": "kaizen/llms/prompts/work_summary_prompts.py",
-    "start_line": 44,
-    "end_line": 65,
-    "side": "RIGHT",
-    "sentiment": "positive",
-    "severity": 4
   }
 ]
\ No newline at end of file
diff --git a/.experiments/code_review/dataset/pr_335/issues.json b/.experiments/code_review/dataset/pr_335/issues.json
index 5ec8ba95..ff0bc0b7 100644
--- a/.experiments/code_review/dataset/pr_335/issues.json
+++ b/.experiments/code_review/dataset/pr_335/issues.json
@@ -56,7 +56,7 @@
     "start_line": 1,
     "end_line": 92,
     "side": "RIGHT",
-    "sentiment": "positive",
+    "sentiment": "negative",
     "severity": 7
   },
   {
diff --git a/.experiments/code_review/dataset/pr_400/issues.json b/.experiments/code_review/dataset/pr_400/issues.json
index 037ac7a3..03f01269 100644
--- a/.experiments/code_review/dataset/pr_400/issues.json
+++ b/.experiments/code_review/dataset/pr_400/issues.json
@@ -1,24 +1,4 @@
 [
-  {
-    "category": "Test Coverage",
-    "description": "Improved test coverage with more comprehensive test cases, including edge cases and error handling scenarios.",
-    "impact": "high",
-    "rationale": "Comprehensive test coverage is crucial for maintaining code quality, catching potential bugs, and ensuring the reliability of the code.",
-    "recommendation": "Continue to add test cases for edge cases and ensure all new functionality is covered by tests.",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_pr_description.py, .kaizen/unit_test/kaizen/helpers/test_create_pr_review_text.py, .kaizen/unit_test/kaizen/helpers/test_create_test_files.py, .kaizen/unit_test/kaizen/helpers/test_get_web_html.py",
-    "sentiment": "positive",
-    "severity": 7
-  },
-  {
-    "category": "Code Structure and Organization",
-    "description": "Improved code structure with better modularity, organization, and use of pytest fixtures.",
-    "impact": "high",
-    "rationale": "Good organization improves readability, maintainability, and allows for better separation of concerns.",
-    "recommendation": "Continue to monitor for potential further improvements in code structure and organization.",
-    "file_path": ".kaizen/unit_test/kaizen/helpers/test_create_test_files.py, kaizen/generator/unit_test.py",
-    "sentiment": "positive",
-    "severity": 3
-  },
   {
     "category": "Error Handling",
     "description": "Insufficient error handling in some methods, particularly for file operations.",
@@ -38,15 +18,5 @@
     "file_path": "kaizen/llms/provider.py",
     "sentiment": "negative",
     "severity": 7
-  },
-  {
-    "category": "Configuration Update",
-    "description": "Changes made to sensitive configuration file",
-    "impact": "critical",
-    "rationale": "Changes to config.json may affect system functionality and security",
-    "recommendation": "Review the changes in config.json carefully, especially the new 'base_model' fields, and ensure that the code using this configuration is updated accordingly.",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
   }
 ]
\ No newline at end of file
diff --git a/.experiments/code_review/dataset/pr_476/issues.json b/.experiments/code_review/dataset/pr_476/issues.json
index 035e3712..2e0b9369 100644
--- a/.experiments/code_review/dataset/pr_476/issues.json
+++ b/.experiments/code_review/dataset/pr_476/issues.json
@@ -11,7 +11,7 @@
     "start_line": 140,
     "end_line": 141,
     "sentiment": "negative",
-    "severity": 9
+    "severity": 7
   },
   {
     "category": "Code Efficiency",
@@ -54,15 +54,5 @@
     "end_line": 22,
     "sentiment": "negative",
     "severity": 6
-  },
-  {
-    "category": "Configuration Changes",
-    "description": "Changes made to sensitive configuration file",
-    "impact": "critical",
-    "rationale": "Changes to config.json may affect system functionality and security. The removal of 'enable_observability_logging' option needs to be properly documented.",
-    "recommendation": "Review all changes in config.json carefully. If removing features, provide a migration guide or deprecation notice for existing users.",
-    "file_path": "config.json",
-    "sentiment": "negative",
-    "severity": 10
   }
 ]
\ No newline at end of file
diff --git a/.experiments/code_review/evaluate.py b/.experiments/code_review/evaluate.py
index 3f13c579..dc7c8bc5 100644
--- a/.experiments/code_review/evaluate.py
+++ b/.experiments/code_review/evaluate.py
@@ -23,32 +23,27 @@ def compare_issues(ground_truth, model_issues):
     for gt_issue in ground_truth:
         found_match = False
         for model_issue in model_issues:
-            category_similarity = calculate_similarity(
-                gt_issue["category"], model_issue["category"]
-            )
             description_similarity = calculate_similarity(
                 gt_issue["description"], model_issue["description"]
             )
-
             if (
-                category_similarity > 0.5
-                and description_similarity > 0.5
+                description_similarity > 0.1
                 and gt_issue["file_path"] == model_issue["file_path"]
                 and abs(
                     int(gt_issue.get("start_line", 0))
                     - int(model_issue.get("start_line", -10))
                 )
-                <= 2
+                <= 1
                 and abs(
                     int(gt_issue.get("end_line", 0))
                     - int(model_issue.get("end_line", -10))
                 )
-                <= 2
+                <= 1
                 and abs(
                     int(gt_issue.get("severity", 0))
                     - int(model_issue.get("severity", -10))
                 )
-                <= 2
+                <= 1
             ):
                 matched.append((gt_issue, model_issue))
                 found_match = True
diff --git a/kaizen/llms/prompts/code_review_prompts.py b/kaizen/llms/prompts/code_review_prompts.py
index 86cf82c1..bc064407 100644
--- a/kaizen/llms/prompts/code_review_prompts.py
+++ b/kaizen/llms/prompts/code_review_prompts.py
@@ -24,56 +24,51 @@
   ]
 }}
 
-## Guidelines
-1. Syntax and Logic Errors:
-   - Incorrect use of language constructs
-   - Off-by-one errors
-   - Null pointer dereferences
-   - Type mismatches
-   - Incorrect boolean logic
-
-2. Performance Issues:
-   - Inefficient algorithms or data structures
-   - Unnecessary computations or memory allocations
-   - Improper use of caching mechanisms
-   - N+1 query problems in database operations
-   - Unoptimized loops or recursions
-
-3. Resource Management:
-   - Memory leaks
-   - Unclosed resources (file handles, database connections)
-   - Improper disposal of heavy objects
-
-4. Concurrency and Threading:
-   - Race conditions
-   - Deadlocks or livelocks
-   - Improper use of synchronization primitives
-   - Thread safety issues
-
-5. Security Vulnerabilities:
-   - SQL injection
-   - Cross-site scripting (XSS)
-   - Insecure cryptographic practices
-   - Improper input validation
-   - Hard-coded credentials or sensitive information
-
-6. Code Structure and Design:
-   - Violations of SOLID principles
-   - Excessive complexity (high cyclomatic complexity)
-   - Code duplication
-   - Poor separation of concerns
-   - Inconsistent naming conventions or coding style
-
-7. Error Handling and Logging:
-   - Inadequate or overly broad exception handling
-   - Lack of proper logging or monitoring
-   - Swallowed exceptions without proper handling
-
-Field Guidelines:
-- "suggested_code": Corrected code for additions only, between start_line and end_line.
-- "start_line" and "end_line": Actual line numbers in the additions.
-- "severity": 1 (least severe) to 10 (most critical).
+## Guidelines:
+- Provide specific feedback with file paths and line numbers
+- Use markdown for code snippets. Make sure all code is following the original indentations.
+- Merge duplicate feedback
+- Examine: syntax/logic errors, resource leaks, race conditions, security vulnerabilities
+- If no issues found: {{"review": []}}
+
+## Patch Data Format:
+- First column: Original file line numbers
+- Second column: New file line numbers (spaces for removed lines)
+- Third column: Change type
+  '-1:[-]': Line removed
+  '+1:[+]': Line added
+  '0:[.]': Unchanged line
+- Remaining columns: Code content
+
+Example:
+
+1    -    -1:[-] def old_function(x):
+2         +1:[+] def new_function(x, y):
+3    3     0:[.]     result = x * 2
+4         +1:[+]     result += y
+4    5     0:[.]     return result
 
+This snippet shows a diff (difference) between two versions of a function:
+
+1. The function name changed from 'old_function' to 'new_function'.
+2. A new parameter 'y' was added to the function.
+3. The line 'result = x * 2' remained unchanged.
+4. A new line 'result += y' was added, incorporating the new parameter.
+5. The return statement remained unchanged.
+
+
+## Review Focus:
+1. Removals (-1:[-]): Identify if removal causes problems in remaining code. Remember any line having -1:[-] is removed line from the new code.
+2. Additions (+1:[+]): Provide detailed feedback and suggest improvements. Remember any line having +1:[+] is added line.
+3. Consider impact of changes on overall code structure and functionality.
+4. Note unchanged lines (0:[.]) for context.
+5. For 'fixed_code' -> always suggest changes for Additions. 
+
+## Field Guidelines:
+- "fixed_code": Corrected code for additions only, between start_line and end_line. make sure start_line you suggest does not has `0:[.]`.
+- "actual_code": Current Code line which you think has error. make sure it always done on `+1:[+]` lines. If not, keep it empty ''.
+- "start_line" and "end_line": Actual line numbers in the additions.
+- "severity_level": 1 (least severe) to 10 (most critical).
 Prioritize issues based on their potential impact on code quality, functionality, and maintainability. Provide concrete examples or code snippets when suggesting improvements.
 
 ## PATCH DATA:
@@ -103,55 +98,51 @@
 }}
 
 
-## Guidelines
-1. Syntax and Logic Errors:
-   - Incorrect use of language constructs
-   - Off-by-one errors
-   - Null pointer dereferences
-   - Type mismatches
-   - Incorrect boolean logic
-
-2. Performance Issues:
-   - Inefficient algorithms or data structures
-   - Unnecessary computations or memory allocations
-   - Improper use of caching mechanisms
-   - N+1 query problems in database operations
-   - Unoptimized loops or recursions
-
-3. Resource Management:
-   - Memory leaks
-   - Unclosed resources (file handles, database connections)
-   - Improper disposal of heavy objects
-
-4. Concurrency and Threading:
-   - Race conditions
-   - Deadlocks or livelocks
-   - Improper use of synchronization primitives
-   - Thread safety issues
-
-5. Security Vulnerabilities:
-   - SQL injection
-   - Cross-site scripting (XSS)
-   - Insecure cryptographic practices
-   - Improper input validation
-   - Hard-coded credentials or sensitive information
-
-6. Code Structure and Design:
-   - Violations of SOLID principles
-   - Excessive complexity (high cyclomatic complexity)
-   - Code duplication
-   - Poor separation of concerns
-   - Inconsistent naming conventions or coding style
-
-7. Error Handling and Logging:
-   - Inadequate or overly broad exception handling
-   - Lack of proper logging or monitoring
-   - Swallowed exceptions without proper handling
-
-Field Guidelines:
-- "suggested_code": Corrected code for additions only, between start_line and end_line.
+## Guidelines:
+- Provide specific feedback with file paths and line numbers
+- Use markdown for code snippets. Make sure all code is following the original indentations.
+- Merge duplicate feedback
+- Examine: syntax/logic errors, resource leaks, race conditions, security vulnerabilities
+- If no issues found: {{"review": []}}
+
+## Patch Data Format:
+- First column: Original file line numbers
+- Second column: New file line numbers (spaces for removed lines)
+- Third column: Change type
+  '-1:[-]': Line removed
+  '+1:[+]': Line added
+  '0:[.]': Unchanged line
+- Remaining columns: Code content
+
+Example:
+
+1    -    -1:[-] def old_function(x):
+2         +1:[+] def new_function(x, y):
+3    3     0:[.]     result = x * 2
+4         +1:[+]     result += y
+4    5     0:[.]     return result
+
+This snippet shows a diff (difference) between two versions of a function:
+
+1. The function name changed from 'old_function' to 'new_function'.
+2. A new parameter 'y' was added to the function.
+3. The line 'result = x * 2' remained unchanged.
+4. A new line 'result += y' was added, incorporating the new parameter.
+5. The return statement remained unchanged.
+
+
+## Review Focus:
+1. Removals (-1:[-]): Identify if removal causes problems in remaining code. Remember any line having -1:[-] is removed line from the new code.
+2. Additions (+1:[+]): Provide detailed feedback and suggest improvements. Remember any line having +1:[+] is added line.
+3. Consider impact of changes on overall code structure and functionality.
+4. Note unchanged lines (0:[.]) for context.
+5. For 'fixed_code' -> always suggest changes for Additions. 
+
+## Field Guidelines:
+- "fixed_code": Corrected code for additions only, between start_line and end_line. make sure start_line you suggest does not has `0:[.]`.
+- "actual_code": Current Code line which you think has error. make sure it always done on `+1:[+]` lines. If not, keep it empty ''.
 - "start_line" and "end_line": Actual line numbers in the additions.
-- "severity": 1 (least severe) to 10 (most critical).
+- "severity_level": 1 (least severe) to 10 (most critical).
 
 Prioritize issues based on their potential impact on code quality, functionality, and maintainability. Provide concrete examples or code snippets when suggesting improvements.
 

From bfc8514f54acae1fff463827f4d9e45be2f3eb5c Mon Sep 17 00:00:00 2001
From: Saurav Panda <sgp65@cornell.edu>
Date: Tue, 27 Aug 2024 20:33:38 -0700
Subject: [PATCH 19/19] feat: added file count in scan output

---
 kaizen/reviewer/code_scan.py | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/kaizen/reviewer/code_scan.py b/kaizen/reviewer/code_scan.py
index d72f34c2..c86c4bd4 100644
--- a/kaizen/reviewer/code_scan.py
+++ b/kaizen/reviewer/code_scan.py
@@ -21,6 +21,8 @@ class CodeScanOutput:
     issues: List[Dict]
     usage: Dict[str, int]
     model_name: str
+    total_files: int
+    files_processed: int
 
 
 class CodeScanner:
@@ -60,6 +62,7 @@ def review_code_dir(
         self.logger.info(f"Starting code review for directory: {dir_path}")
         self.reevaluate = reevaluate
         issues = []
+        files_processed = 0
         for file_path in Path(dir_path).rglob("*.*"):
             if self.should_ignore(file_path):
                 continue
@@ -68,6 +71,7 @@ def review_code_dir(
                     file_data = f.read()
                 self.logger.debug(f"Reviewing file: {file_path}")
                 code_scan_output = self.review_code(file_data=file_data, user=user)
+                files_processed += 1
                 for issue in code_scan_output.issues:
                     issue["file_path"] = str(file_path)
                     issues.append(issue)
@@ -81,6 +85,8 @@ def review_code_dir(
             usage=self.total_usage,
             model_name=self.provider.model,
             issues=issues,
+            total_files=files_processed,
+            files_processed=files_processed,
         )
 
     def review_code(self, file_data: str, user: Optional[str] = None) -> CodeScanOutput: