grounded.tools
Grounded.tools (Grounded Docs MCP Server) is an open-source, privacy-first documentation indexing tool that keeps AI assistants informed with up-to-date, version-specific documentation from multiple sources. The tool addresses common issues like stale LLM knowledge, hallucinated code, and version ambiguity by indexing documentation from websites, GitHub repositories, npm, PyPI, and local files, then providing AI assistants with version-aware search capabilities via the Model Context Protocol (MCP). Unlike generic web scrapers, Grounded.tools uses structure-aware RAG (Retrieval-Augmented Generation) that parses documents as hierarchies and intelligently handles code blocks while preserving API signatures and proper formatting. The system runs entirely locally on your machine for privacy, offers semantic search with relevance ranking, supports OAuth security, and provides tools like search_docs for querying indexed documentation and fetch_url for grabbing pages on the fly. Compatible with popular AI development tools including Cursor, Claude Desktop, VS Code, GitHub Copilot, and any MCP-compatible assistant, Grounded.tools serves as an open-source alternative to Context7, Nia, and Ref.Tools, enabling faster AI code assistance while reducing API costs through condensed, highly-relevant context delivery.