Getting Started

Follow these steps to install, configure, and run the Document RAG MCP server.

Installation

This project uses uv for python dependency management. Ensure you have uv installed, then synchronize the environment:

git clone https://github.com/user/document-rag-mcp.git
cd document-rag-mcp
uv sync --group dev

Copy the example configuration file and adjust it to your directories:

cp config.example.yaml config.yaml

Edit config.yaml to specify the folders you want to watch.

You can test the ingestion and run semantic searches directly from the command line:

Run a one-shot ingestion scan to populate the vector store:

uv run document-rag-mcp ingest

Query the indexed collections:

uv run document-rag-mcp search "What is project antigravity?"

Check chunk counts and file listings:

uv run document-rag-mcp collections

Start the server using stdio transport (default) for integration with LLM clients:

uv run document-rag-mcp serve --transport stdio

For HTTP/SSE integration, specify the transport and bindings:

uv run document-rag-mcp serve --transport http --host 127.0.0.1 --port 8000