Documentation Extractor
Extracts llms.txt and install.md docs from websites for AI agents and MCP clients
- Platforms
- Web, API, CLI
- Version
- 1.0.0
- Listing updated
- today
Similar MCP servers
About Documentation Extractor
Documentation Extractor is an MCP server for ai models and llm tooling. It is available on Web, API and CLI.
The README describes this project as a web app and MCP server that fetches documentation from sites following the llms.txt and install.md conventions, then reorganizes it into agent-friendly files. Output can be split into per-section markdown, or exported as Markdown, JSON, YAML or a ZIP archive.
The MCP server exposes tools for extracting docs from a URL, validating whether a site offers llms.txt, verifying the file, discovering documentation URLs for a domain, and searching a directory of known llms.txt-enabled sites. It also includes a generator that uses Claude to produce install.md files from GitHub repositories or documentation pages, which requires an Anthropic API key.
The project is aimed at developers wiring up coding assistants, RAG pipelines or agents that need current vendor documentation. The README lists a hosted version at llm.energy alongside a local development setup.
Tools this server provides (6)
Taken from the project documentation; the running server may differ.
- extract_docs
- Extract documentation from a URL with llms.txt support
- validate_url
- Check if a URL has llms.txt available
- verify_llms_txt
- Verify llms.txt exists and get file info
- discover_documentation_urls
- Find documentation URLs for a domain
- list_sites
- Get directory of known llms.txt-enabled sites
- search_sites
- Search the site directory by category or keyword
What it does
- Automatic llms.txt detection
- Looks for llms.txt, llms-full.txt and install.md files on a given documentation site.
- Sectioned markdown output
- Splits extracted documentation into individual markdown files organized by section.
- Multiple export formats
- Exports results as Markdown, JSON, YAML or a ZIP archive.
- Batch URL processing
- Handles several documentation URLs in one request.
- install.md generator
- Creates LLM-executable installation instructions from GitHub repos or docs URLs using Claude.
- Curated site directory
- Browses a list of websites the project lists as offering llms.txt support.
Good for
- Feeding vendor documentation into coding assistants such as Cursor or Windsurf
- Building RAG pipelines over up-to-date product docs
- Checking whether a documentation site publishes llms.txt before integrating it
- Generating install.md files for an existing GitHub project
Documentation Extractor FAQ
Which MCP tools does the server expose?
The README lists extract_docs, validate_url, verify_llms_txt, discover_documentation_urls, list_sites and search_sites.
Does install.md generation need an API key?
Yes. The README says the ANTHROPIC_API_KEY variable enables Claude-backed generation and the endpoint returns a configuration error without it.
How is the MCP server started?
The README gives a JSON client configuration that runs the @llm-energy/mcp-server package with npx.
What license applies?
The README states the project is proprietary and all rights are reserved by its author.
What are the best alternatives to Documentation Extractor?
The closest alternatives to Documentation Extractor are llms-text, Llmmcp, llms.txt, Toolbox and gomddoc.
Where this comes from
- Description: written from the project's README and its MCP Registry entry.

