integration of Plugins #58
kevinseabourne
started this conversation in
Feature request
Replies: 1 comment
-
this is amazing suggestions 🔥🔥 |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
I propose a feature enhancement for Dialoqbase that involves the integration of AI plugins for retrieval, inspired by Langchain's Plug-and-Plai system. This feature would entail:
Plugin Indexing: Establish a system to load and index plugins from a dedicated directory. This would expand the AI's toolset and allow for more dynamic responses.
Embedding Creation: For each tool description, generate embeddings. These embeddings serve as a numerical representation of the tool's description, enabling efficient similarity searches.
Similarity Search: Implement a similarity search function that uses the embeddings of incoming queries to find relevant tools. This would allow the AI to select the most appropriate tool for handling a specific query.
Dynamic Tool Selection: Introduce a 'tools_getter' function in the prompt template. This function would dynamically select the tools to use based on the input, enhancing the AI's adaptability to different queries.
https://python.langchain.com/docs/use_cases/agents/custom_agent_with_plugin_retrieval_using_plugnplai
Beta Was this translation helpful? Give feedback.
All reactions