llms.txt checker: does your file follow the format?
Enter your domain. We fetch your llms.txt, check it against the llmstxt.org format and separate what breaks it from what is only recommended. With no file, or a broken one, you get a starter file built from your home page.
What the check reads
- Your home pageWhere your address lands after redirects, and its title and description for the starter file.
- Your llms.txtThe file at /llms.txt on that site: whether it is there, whether it is markdown rather than an error page, and how it is laid out.
- The formatErrors for what breaks it, recommendations for what it suggests, as llmstxt.org sets them out.
llms.txt is a proposal, not a rule search engines or assistants must follow. It allows and blocks nothing; robots.txt does that.
What llms.txt is
llms.txt is a proposal, published at llmstxt.org, for a markdown file at the root of a website. Where robots.txt tells crawlers what they may fetch, llms.txt tells an AI assistant what is worth reading: the site's name, a one-line summary and a short list of links to the pages that matter, each with a note. The proposal's reasoning is that an assistant cannot read a whole website at once, and a hand-picked list written in plain markdown is quicker to use than pages full of navigation, scripts and adverts.
What the format requires, and what it only recommends
The format requires one line: an H1 title with the site's name, first in the file. Everything else is recommended rather than required: a summary in a > blockquote straight after the title, optional paragraphs with details, then H2 sections that each hold a markdown list of links in the form - [Page name](https://…): what it covers. A section titled Optional, by convention, lists links an assistant can skip when it needs less.
This checker follows that split. Errors are what break the format: an empty file, an HTML page served at /llms.txt (often a "page not found" page that answers as if it were found), a first line that is not an H1, or more than one H1. Recommendations are the optional parts your file lacks.
What to put in it
Start with the pages a new customer asks about first: what you sell or do, your prices, where you work, how to order or book, and how to reach you. Give each link a short note that says what the page answers, in plain words. Put secondary pages, such as your history or press coverage, under Optional. Save the file as plain text named llms.txt at the root of your site, so it opens at yourdomain.com/llms.txt, and link only to pages anyone can open.
A short example
# Corner Bakery > Sourdough bread and pastries baked daily, with same-day pickup. ## Pages - [Menu](https://example.com/menu/): breads, cakes and prices - [Order](https://example.com/order/): same-day pickup ## Optional - [Our story](https://example.com/about/)
What it will not do
llms.txt is a proposal, not a rule that search engines or assistants have to follow, and support for it varies between AI tools. It allows and blocks nothing: to control which AI crawlers may fetch your pages, use robots.txt, and check it with the AI crawler checker. Keep the file short, link to pages that already answer your customers' questions well, and update it when those pages change. Ours is at serpcompass.com/llms.txt.
Questions and answers
What is llms.txt?
A proposal, published at llmstxt.org, for a markdown file at the root of a website, /llms.txt, that gives AI assistants a short map of the site: its name, a one-line summary and links to the pages worth reading.
Do search engines or AI assistants require llms.txt?
No. It is a proposal, not a rule that search engines or assistants have to follow, and support for it varies between AI tools. It does not replace robots.txt or a sitemap.
What does the llms.txt format require?
One thing: an H1 title with the name of the site, as the first line. The format recommends a one-line summary in a '>' blockquote under it, then H2 sections that list links as '- [Page name](URL): notes'. A section titled Optional holds links an assistant can skip.
How is llms.txt different from robots.txt and sitemap.xml?
robots.txt tells crawlers what they may fetch. sitemap.xml lists the URLs you want search engines to find. llms.txt is a short, hand-picked reading list for AI assistants, written in markdown; it allows and blocks nothing.
Where does the starter file come from?
From your home page: its site name, or else its title, becomes the H1, and its meta description becomes the summary. Everything in square brackets is a placeholder only you can fill in. Read it through before you publish it.
More free checks
- AI crawler checker: which search and AI crawlers your robots.txt lets read your home page.
- Free site scan: a crawl of up to 40 pages for technical and on-page SEO problems.
- All free tools
Start with a free scan of your site.
Two to five minutes, no account. You get your three most urgent findings, and the exact count of everything else.