
FastMCP RAG Template
Supports UIby harangju
Production-ready FastMCP and LangChain RAG template with multi-provider LLM support.
What it does
This server implements a Retrieval-Augmented Generation (RAG) system that connects LLMs to local or remote documents. It enables AI agents to query specific knowledge bases using semantic search and hybrid retrieval.
Tools
setup_rag(): Initializes the RAG system with auto-provider detection.query(): Executes a RAG query with conversation history support.add_documents(): Performs bulk document addition for indexing.search(): Provides semantic and keyword hybrid search across indexed docs.get_config(): Displays current RAG configuration and metrics.switch_provider(): Hot-swaps between LLM providers (OpenAI, Anthropic, Gemini).clear_index(): Resets the vector store.
Installation
Add the following to your claude_desktop_config.json:
"rag-mcp": {
"command": "uv",
"args": ["run", "/path/to/server.py"]
}
Supported hosts
- Claude Desktop
Quick install
uv pip install -e .Information
- Pricing
- free
- Published
- 4/30/2026
- stars






