Docsheaf
AI-ready docs

Docs that AI tools can read — and get right

Every published page ships a clean Markdown version, and every site publishes llms.txt and llms-full.txt, so assistants and coding agents read your docs without scraping HTML.

Built for machine readers

Nothing to configure: these files are generated the moment you publish.

Markdown twin

Add .md to any page URL to get the page as Markdown — the format language models read best.

llms.txt

An index of every page in the site with links, following the llms.txt convention.

llms-full.txt

The whole site in a single Markdown file, ready to drop into a model’s context.

Sitemap and robots

sitemap.xml and robots.txt for crawlers; unlisted sites tell them to stay out.

Try it on any Docsheaf site

Point an agent — or curl — at your docs.

Terminal
# One page as Markdown
curl https://docs.example.com/getting-started/quickstart.md

# Every page, with links
curl https://docs.example.com/llms.txt

# The entire site in one file
curl https://docs.example.com/llms-full.txt

Frequently asked questions

What is llms.txt?

A plain-text file at the root of a site that lists its pages with links, so language models and agents can find content without crawling HTML. Docsheaf also publishes llms-full.txt with the full text of every page.

How do I get a page as Markdown?

Add .md to the page URL. The Markdown is generated from the same document as the HTML page, so they never disagree.

Are unlisted sites exposed to AI crawlers?

No. Unlisted sites need the share link, their robots.txt disallows crawling, and pages are marked noindex.

Do I need to turn anything on?

No. Markdown pages, llms.txt, llms-full.txt, sitemap.xml and robots.txt are available as soon as a site is published.

Accurate docs. Better answers.

Start writing for free, or talk to us about bringing your team and your existing docs.