Docs MCP Server

Example Use Cases

• 1

  Access and search indexed documentation for libraries and packages (including versioned docs) to answer developer questions.

• 2

  Fetch a single web page and convert it to Markdown for downstream processing or display.

• 3

  Orchestrate and monitor asynchronous scraping and indexing jobs (start, cancel, list, and check job status).

Description

This server crawls and fetches documentation from websites, GitHub, package registries or local files, splits content semantically, and indexes it for fast search and retrieval. It is designed to store documents in SQLite (with FTS5 for full-text search) and originally supported vector embeddings and hybrid search; note the README states embedding functionality has been removed in the current version so only full-text search and chat/completions remain. The project includes asynchronous job handling, versioning of indexed docs, OpenRouter-compatible chat/completions integration, and easy deployment via Docker or npx. Useful for building agents that need to query or navigate library docs and manage indexing workflows.