Docs MCP Server

Example Use Cases

• 1

  Access and search up-to-date, version-specific library documentation (e.g., exact API behavior for a given dependency version).

• 2

  Reduce hallucinations by grounding responses in scraped official docs (websites, GitHub, npm, PyPI, or local files).

• 3

  Local, private documentation indexing and semantic search with optional vector embeddings and multiple provider integrations.

Description

Docs MCP Server scrapes and indexes documentation from websites, GitHub, npm, PyPI and local files, then serves that content to MCP-compatible AI assistants. It provides semantic chunking, optional vector search (via multiple embedding providers), and version-specific search to reduce hallucinations and ensure answers match a project's dependencies. The project includes a web UI, CLI, and can run via Docker or npx for local, private deployments. Optional enterprise features include OAuth2/OIDC authentication and scalable Docker Compose architecture.