mcp-registry/ragie-mcp-server

    ==================
      
       /// MCP ///
      /// RAG ///
        
    ==================
        
    [server:online]
    [protocol:ready]

ragie-mcp-server

by ragieai

Ragie Model Context Protocol Server – exposes Ragie knowledge-base retrieval through the Model Context Protocol (MCP) with a single "retrieve" tool.

57
14
Open Source
GitHub

01

retrieve

Query the Ragie knowledge base to return relevant content chunks based on a search query, with optional parameters for result count, reranking, and recency bias.

Installation

1. Prerequisites
• Node.js 18 LTS or higher
• npm (shipped with Node) or pnpm/yarn
2. Clone the repository
git clone https://github.com/ragieai/ragie-mcp-server.git cd ragie-mcp-server
3. Install dependencies
npm install # or: pnpm install | yarn install
4. Configure environment
• Copy the example env file (if present) and edit values:
cp .env.example .env # then set PORT, AUTH_TOKEN, DATA_DIR, etc.
• Generate and store any API keys the server needs to reach your model provider (OpenAI, Anthropic, etc.).
5. Run in development mode
npm run dev # hot-reload with nodemon if configured
6. Run in production
npm run build # optional – if TypeScript npm start # launches the MCP server on PORT (default 3000)
7. Reverse-proxy (optional)
• Place Nginx/Caddy/Traefik in front of the server and forward public traffic to the PORT you configured.
8. Docker (if Dockerfile exists)
docker build -t ragie-mcp-server . docker run -d -p 3000:3000 --env-file .env ragie-mcp-server

Documentation

License: MIT License
Updated 7/30/2025