Docs MCP Server

Example Use Cases

• 1

  Fetch and index documentation from websites, GitHub, npm, PyPI or local files for fast lookup.

• 2

  Run full‑text searches across library docs, versions, and unversioned content using a local SQLite store.

• 3

  Automated scraping and background job management (scrape jobs, progress, cancelation) with a CLI and Docker/npx deployment.

Description

This server crawls and converts documentation into markdown, splits content semantically, and stores it in SQLite with FTS5 full‑text search for fast retrieval. Embedding/vector features have been removed in this version, so vector search is not available; the project retains full‑text search and integrated large‑model chat/completions (including OpenRouter support). It exposes MCP tools and a CLI to manage scraping jobs, check job status, search docs, list libraries/versions, remove documents, and fetch single URLs. Deployable via Docker or npx for easy local or containerized usage.