Mcp Local Rag

Mcp Local Rag is an open-source Model Context Protocol (MCP) server and CLI tool for fast, offline, retrieval-augmented search of local documents (PDF, DOCX, Markdown, and TXT) on your machine. It combines semantic similarity and keyword search, providing precise context—including headings and line/page numbers—in results, making it ideal for developers, researchers, and technical teams who need private, high-quality search over their documentation without uploading files to external APIs. Works offline after initial model download, requires no external database or Docker, and integrates with MCP clients or can operate via CLI.