Qmd
Indexes and searches local Markdown files with high performance, offering BM25 full-text, vector semantic, and hybrid search, alongside an MCP server for AI agent integration.
Qmd is a powerful and private local search engine designed for Markdown files. Built in Rust, it provides a high-performance solution for managing and searching your personal knowledge base or project documentation without relying on cloud services. It features a sophisticated hybrid search mechanism, combining traditional BM25 full-text search with modern vector semantic search, enhanced by LLM query expansion and reranking for highly relevant results. Qmd also includes an MCP server, enabling seamless integration with AI agents like Claude Desktop, transforming your local Markdown collections into a dynamic knowledge source for AI-powered workflows.