MCPIndependentshinpr
MCP Local RAG
Searches your PDF, DOCX and Markdown documents locally without sending them to an API.
The server indexes PDF, DOCX, Markdown and text files from a chosen folder and combines semantic search with keyword matching, returning passages with their heading, line or page number. It needs no API key, Docker or Python; after the model is downloaded the first time, it works offline.
Installation
Manual install
$claude mcp add local-rag --scope user --env BASE_DIR=<PATH> -- npx -y mcp-local-ragDetails
- Works in
- Claude Code, Claude, Cursor, Codex, Gemini CLI, VS Code, Windsurf, Zed
- Transport
- local
- License
- MIT
- Tasks
- Research, Docs
- Rating
- Source
- github.com/shinpr/mcp-local-rag (opens in a new tab)
Tags
- rag
- documents
- semantic search
- offline
A pack is your own set of skills and MCP servers. sciver loads it in your AI with one command.