Skip to content
RookAtlas

Draft and check an llms.txt

Draft an llms.txt from your sitemap in a few seconds, or check a site's own file against the format, line by line. And know before you spend time on it what it does and doesn't do.

We fetch /llms.txt from the site's root once and store nothing.

Before a check runs, Cloudflare Turnstile confirms you're a person; it reads signals from your browser and your IP address for that. Cloudflare's privacy addendum

The llms.txt format

  1. An H1 with the site's name. It is the only required part.
  2. A blockquote with a one-sentence summary, then any notes without headings.
  3. H2 sections, each a list of Markdown links: - [Name](https://…): what it is.
  4. Optionally a section named Optional for links an agent can skip.

The validator checks those rules line by line, and that the file is served as text: many sites built as single-page apps answer /llms.txt with their HTML shell, which no reader can use.

Drafting one from your sitemap

The generator reads the sitemap your robots.txt names, or /sitemap.xml, follows up to three child sitemaps, and groups your pages by their top folder, with the pages at the root first. Names are guessed from the addresses, so treat it as a draft: rename the links, add a note after each, and cut what an agent doesn't need.

Where it ranks among AI search fixes

Low. Crawler access, text that arrives without JavaScript and pages that answer buyers' questions come first, and so does being mentioned on the sites AI answers cite. The AI visibility checklist puts them in order with their evidence.

Questions

What is llms.txt?

A proposed Markdown file at a site's root, /llms.txt, that gives language models a short index of the site: its name, a summary, and links to the pages that matter, grouped in sections. It was proposed at llmstxt.org.

Does llms.txt help with AI search?

Not in Google: Google says Search, including its AI features, doesn't use llms.txt. Several vendor studies report no measurable effect on AI citations, and few requests for the file from AI crawlers. It can still help an agent or a coding tool that reads it on purpose.

Is llms.txt like robots.txt?

No. robots.txt tells crawlers what they may fetch, and they follow it. llms.txt is a reading list with no rules in it; nothing obliges an AI to read it or to stay out of anything.

What's the Optional section?

By the format's convention, links under an H2 called Optional are secondary: an agent can skip them when it needs a shorter context.

Read more

RookAtlas opens soon.

RookAtlas asks your buyers' questions on eight AI engines and shows how often you're named, next to whom and from which sources, with a 95% interval on visibility and citation rates.