knowledge-rag

by lyonzin · Python · ★ 67

Drop docs, search instantly from Claude Code — 12 MCP tools, 20 format parsers, hybrid search + reranking. Zero servers, zero API keys, 100% local.

#bm25#chromadb#claude#claude-code#document-search#embeddings#fastembed#hybrid-search#knowledge-base#local-ai#mcp#mcp-server#onnx#privacy#python#rag#reranking#retrieval-augmented-generation#semantic-search#vector-database

Install

pip install git+https://github.com/lyonzin/knowledge-rag.git

Claude Desktop config

Add this to your claude_desktop_config.json:

{
  "mcpServers": {
    "knowledge-rag": {
      "command": "uvx",
      "args": [
        "git+https://github.com/lyonzin/knowledge-rag.git"
      ]
    }
  }
}

From the README

[](https://pypi.org/project/knowledge-rag/) [](https://www.npmjs.com/package/knowledge-rag) [](https://pepy.tech/project/knowledge-rag) [](https://github.com/lyonzin/knowledge-rag/actions/workflows/ci.yml) [](https://github.com/lyonzin/knowledge-rag/actions/workflows/security.yml) [](https://glama.ai/mcp/servers/lyonzin/knowledge-rag) Drop your PDFs, markdown, code, notebooks — **1800+ files, 39K chunks, indexed in under 3 minutes.** Hybrid search (BM25 + semantic vectors + cross-encoder reranking) through 12 MCP tools. Everything runs locally via ONNX. No Docker, no Ollama, no API keys, no…

Read full README on GitHub →