# Git2Docs

> Git2Docs turns a code repository into an accurate, always-current documentation site. It parses the code (tree-sitter across nine languages), comprehends the repo into a product brief, plans a table of contents, and synthesizes user-facing docs that re-sync on every push — so the docs never drift from the code.

Git2Docs is built to be read by both people and AI agents. Every published docs page is also available as clean Markdown by appending `.md` to its URL, each docs site exposes an `llms.txt` index, and there is a public MCP endpoint agents can query for citation-backed answers.

## Product

- [How it works](https://git2docs.com/#how): the comprehend → plan → synthesize pipeline
- [Documentation](https://git2docs.com/docs): product docs — pipeline, architecture, configuration, best practices
- [Quick start](https://git2docs.com/quick-start): connect a repo and generate docs
- [Pricing](https://git2docs.com/pricing): plans and limits
- [Gallery](https://git2docs.com/gallery): example generated docs sites

## For agents

- Append `.md` to any published docs URL for clean Markdown (e.g. `https://git2docs.com/{org}/docs/{product}/{space}/{page}.md`)
- Per-site agent index: `https://git2docs.com/{org}/docs/llms.txt`
- Public MCP server for querying published docs: `https://git2docs.com/api/mcp`

## Compare

- [Comparisons](https://git2docs.com/compare): Git2Docs vs other documentation tools

## Company

- [Blog](https://git2docs.com/blog): product and engineering writing
- [Privacy & security](https://git2docs.com/security): how repositories and code are handled
- [Terms](https://git2docs.com/terms)

## Notes

- We clone a connected repository only for the duration of a documentation run and delete the working copy afterward; we never modify your repositories.
- Sitemap: https://git2docs.com/sitemap.xml