sui-mcp-server
This project is a proof-of-concept MCP server designed to assist in Retrieval-Augmented Generation by querying a vector database and retrieving relevant documents for processing. It integrates the FAISS vector database with a FastAPI server to support AI-agent queries and uses GitHub Move file extraction and LLM integration for comprehensive functionality.
MCP Server with FAISS for RAG
This project is a proof-of-concept implementation of a Machine Conversation Protocol (MCP) server designed to facilitate Retrieval-Augmented Generation (RAG). It integrates a FastAPI server with MCP endpoints and a FAISS vector database to allow an AI agent to query and retrieve relevant documents. Key features include document chunking and embedding, GitHub Move file extraction, LLM integration for a complete RAG workflow, and a simple client example.
Features
- FastAPI server with MCP endpoints
- FAISS vector database integration
- Document chunking and embedding
- GitHub Move file extraction and processing
- LLM integration for RAG workflow
- Simple client example and sample documents
Installation
Installation can be performed using pipx for isolated environments or manually by cloning the repository and installing dependencies.
Usage
Users can utilize mcp-download for downloading Move files, mcp-search-index for enhanced GitHub searching and indexing, mcp-index for indexing files, and mcp-query for querying the vector database. The project allows for RAG querying via mcp-rag and runs as a server using mcp-server.