docs: add comprehensive LlamaIndex integration documentation

docs/mint.json

@@ -96,6 +96,7 @@
         "v1/integrations/groq",
         "v1/integrations/langchain",
         "v1/integrations/llama_stack",
+        "v1/integrations/llamaindex",
         "v1/integrations/litellm",
         "v1/integrations/mistral",
         "v1/integrations/multion",

docs/v1/integrations/llamaindex.mdx

@@ -0,0 +1,91 @@
+---
+title: 'LlamaIndex'
+description: 'LlamaIndex is a data framework for LLM applications.'
+---
+
+import CodeTooltip from '/snippets/add-code-tooltip.mdx'
+import EnvTooltip from '/snippets/add-env-tooltip.mdx'
+
+# LlamaIndex Integration
+
+AgentOps provides seamless integration with LlamaIndex, enabling comprehensive monitoring and evaluation of your LlamaIndex applications.
+
+## Installation
+
+<CodeGroup>
+```bash pip
+pip install llama-index-instrumentation-agentops
+```
+```bash poetry
+poetry add llama-index-instrumentation-agentops
+```
+</CodeGroup>
+
+## Usage
+
+<CodeGroup>
+```python python
+from llama_index.core import set_global_handler
+
+# Initialize AgentOps
+set_global_handler("agentops")
+```
+</CodeGroup>
+
+## Features
+
+### LLM Event Tracking
+- Monitor all LLM interactions
+- Track token usage and costs
+- Analyze prompt effectiveness
+
+### Tool Usage Analytics
+- Monitor tool calls and performance
+- Track success/failure rates
+- Identify bottlenecks
+
+### Error Tracking
+- Capture and analyze errors
+- Debug agent behavior
+- Improve reliability
+
+## Examples
+
+For a complete example of using AgentOps with LlamaIndex, see our [example notebook](https://github.com/AgentOps-AI/agentops/blob/main/examples/llamaindex_examples/llamaindex_examples.ipynb).
+
+Here's a simple example demonstrating AgentOps integration with LlamaIndex:
+
+<CodeGroup>
+```python python
+import os
+from llama_index.core import set_global_handler
+from llama_index.llms.openai import OpenAI
+from llama_index.core import VectorStoreIndex, SimpleDirectoryReader
+
+# Set up AgentOps handler
+set_global_handler("agentops")
+
+# Initialize LLM
+llm = OpenAI(model="gpt-3.5-turbo")
+
+# Load and index documents
+documents = SimpleDirectoryReader('data').load_data()
+index = VectorStoreIndex.from_documents(documents)
+
+# Create query engine
+query_engine = index.as_query_engine()
+
+# Run queries (AgentOps will automatically track these)
+response = query_engine.query("What is the capital of France?")
+print(response)
+```
+</CodeGroup>
+
+For more examples and detailed usage, visit our [documentation](https://docs.agentops.ai/integrations/llamaindex).
+
+<script type="module" src="/scripts/github_stars.js"></script>
+<script type="module" src="/scripts/link_to_api_button.js"></script>
+<script type="module" src="/scripts/scroll-img-fadein-animation.js"></script>
+<script type="module" src="/scripts/button_heartbeat_animation.js"></script>
+<script type="css" src="/styles/styles.css"></script>
+<script type="module" src="/scripts/adjust_api_dynamically.js"></script>

examples/README.md

@@ -22,12 +22,13 @@ At a high level, AgentOps gives you the ability to monitor LLM calls, costs, lat
 
 ## Integrations
 - [AI21](./ai21_examples/ai21_examples.ipynb)
-- [Anthropic](./anthropic_examples/)- 
+- [Anthropic](./anthropic_examples/)-
 - [Autogen](./autogen_examples/)
 - [Cohere](./cohere_examples/cohere_example.ipynb)
 - [Crew.ai](./crew_examples/)
 - [Groq](./multi_agent_groq_example.ipynb)
 - [Langchain](./langchain_examples/langchain_examples.ipynb)
+- [LlamaIndex](./llamaindex_examples/llamaindex_examples.ipynb)
 - [LiteLLM](./litlelm_examples/litlelm_example.ipynb)
 - [Mistral](./mistral_examples/mistral_example.ipynb)
 - [MultiOn](./multion_examples/)

examples/llamaindex_examples/.env.tpl

@@ -0,0 +1,5 @@
+# AgentOps API key for monitoring and observability
+AGENTOPS_API_KEY=your_agentops_key
+
+# OpenAI API key for LLM operations
+OPENAI_API_KEY=your_openai_key

examples/llamaindex_examples/data/sample.txt

@@ -0,0 +1,22 @@
+# AgentOps LlamaIndex Integration Example
+
+This document demonstrates the integration between AgentOps and LlamaIndex for monitoring and observability of LLM applications.
+
+## Key Features
+1. Document Loading: LlamaIndex can efficiently load and process various document formats
+2. Vector Indexing: Documents are indexed for quick retrieval and semantic search
+3. Query Processing: Natural language queries are processed against the indexed documents
+4. Observability: AgentOps provides comprehensive monitoring of all LLM operations
+
+## Benefits
+- Real-time monitoring of LLM interactions
+- Cost tracking and optimization
+- Performance metrics and analytics
+- Error detection and debugging
+- Session replay capabilities
+
+## Use Cases
+- Document question-answering systems
+- Knowledge base search and retrieval
+- Content summarization and analysis
+- Information extraction pipelines

examples/llamaindex_examples/llamaindex_examples.ipynb

@@ -0,0 +1,183 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "82aa4ae2",
+   "metadata": {},
+   "source": [
+    "# LlamaIndex Integration Example\n",
+    "\n",
+    "This notebook demonstrates how to use AgentOps with LlamaIndex for monitoring and observability of your LLM applications.\n",
+    "\n",
+    "## Setup\n",
+    "\n",
+    "First, ensure you have the required packages installed:\n",
+    "```bash\n",
+    "pip install agentops llama-index python-dotenv openai\n",
+    "```"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "cc6137f4",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Import required packages\n",
+    "from llama_index.core import VectorStoreIndex, SimpleDirectoryReader\n",
+    "from llama_index.llms.openai import OpenAI\n",
+    "from llama_index.instrumentation.agentops import AgentOpsHandler\n",
+    "from agentops import Client\n",
+    "import os\n",
+    "from dotenv import load_dotenv"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "1df168b3",
+   "metadata": {},
+   "source": [
+    "## Environment Setup\n",
+    "\n",
+    "Load environment variables from `.env` file. Make sure to copy `.env.tpl` to `.env` and fill in your API keys."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "0dfcc445",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Load environment variables\n",
+    "load_dotenv()\n",
+    "\n",
+    "# Configure API keys\n",
+    "AGENTOPS_API_KEY = os.getenv(\"AGENTOPS_API_KEY\")\n",
+    "OPENAI_API_KEY = os.getenv(\"OPENAI_API_KEY\")\n",
+    "\n",
+    "if not all([AGENTOPS_API_KEY, OPENAI_API_KEY]):\n",
+    "    raise ValueError(\"Please set AGENTOPS_API_KEY and OPENAI_API_KEY in your .env file\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "c14a16bc",
+   "metadata": {},
+   "source": [
+    "## Initialize AgentOps Handler\n",
+    "\n",
+    "Set up the AgentOps handler to track LlamaIndex operations."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "1b54b2d4",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Initialize AgentOps handler\n",
+    "AgentOpsHandler.init(api_key=AGENTOPS_API_KEY)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "5b38d476",
+   "metadata": {},
+   "source": [
+    "## Document Processing Example\n",
+    "\n",
+    "Let's create a simple example using LlamaIndex to process and query documents while tracking the operations with AgentOps."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "3a3b44f4",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Load documents from the data directory\n",
+    "documents = SimpleDirectoryReader('data').load_data()\n",
+    "\n",
+    "# Create an index from the documents\n",
+    "index = VectorStoreIndex.from_documents(documents)\n",
+    "\n",
+    "# Create a query engine\n",
+    "query_engine = index.as_query_engine()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "5c1c3a56",
+   "metadata": {},
+   "source": [
+    "## Query Example\n",
+    "\n",
+    "Now let's query our indexed documents. AgentOps will automatically track the LLM calls and their responses."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "18a191f6",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Query the index\n",
+    "response = query_engine.query(\"What is this document about?\")\n",
+    "print(f\"Response: {response}\")\n",
+    "\n",
+    "# Try another query\n",
+    "response = query_engine.query(\"What are the main topics covered?\")\n",
+    "print(f\"Response: {response}\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "c03fdffa",
+   "metadata": {},
+   "source": [
+    "## Error Handling Example\n",
+    "\n",
+    "Let's demonstrate how AgentOps tracks errors in LlamaIndex operations."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "63187bb6",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "try:\n",
+    "    # Attempt to load documents from a non-existent directory\n",
+    "    documents = SimpleDirectoryReader('nonexistent_directory').load_data()\n",
+    "except Exception as e:\n",
+    "    print(f\"Error caught and tracked by AgentOps: {str(e)}\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "bf4de9ab",
+   "metadata": {},
+   "source": [
+    "## Viewing Results\n",
+    "\n",
+    "You can view the tracked operations, including LLM calls, errors, and performance metrics in the AgentOps dashboard:\n",
+    "https://app.agentops.ai/\n",
+    "\n",
+    "The dashboard will show:\n",
+    "- LLM interactions and their costs\n",
+    "- Query performance metrics\n",
+    "- Error tracking and debugging information\n",
+    "- Document processing statistics"
+   ]
+  }
+ ],
+ "metadata": {},
+ "nbformat": 4,
+ "nbformat_minor": 5
+}