Documentation Server MCP

This MCP server is designed to fetch, process, and search documentation for various software libraries. It can scrape content from multiple sources like websites, GitHub repositories, or local files, then semantically split and generate embeddings (using specified models). The data is stored in an SQLite database optimized for vector similarity and full-text search, enabling hybrid search across different library versions. It allows targeted documentation indexing, version management, and provides tools for searching, listing, removing, and fetching documentation URLs. It supports asynchronous job management and simple deployment via Docker or npx, making it suitable for developers and technical teams managing extensive software documentation.