rag-server-mcp

rag-server-mcp

6
research_and_datadeveloper_tools

MCP RAG Server is a tool for enhancing language model responses by providing contextual documents using Retrieval Augmented Generation (RAG). It employs local models and databases for improved privacy and customization. The project facilitates seamless integration within the MCP ecosystem.

MCP RAG Server

mcp-rag-server is a Model Context Protocol (MCP) server for enabling Retrieval Augmented Generation (RAG) with connected LLMs. It automatically indexes documents from your project to provide relevant context.

Features

  • Automatic Indexing: Scans directory and indexes files like .txt, .md, code files, .json, .csv.
  • Vector Storage: Utilizes ChromaDB for vector storage.
  • Local Embeddings: Uses Ollama for local embedding generation.
  • MCP Tools: Offers functionalities like indexing and querying documents.
  • Dockerized: Easy setup with docker-compose including ChromaDB and Ollama.

Installation & Usage

  1. Install Docker
  2. Clone repository
  3. Start services with Docker Compose
  4. Pull embedding model into Ollama container
  5. Configure MCP client for integration