Check your llms.txt, or build one.
Enter a domain. The checker loads its /llms.txt, validates it against the llmstxt.org proposal and requests up to 15 of its links. No file yet? The generator below builds one.
A valid llms.txt is a small, optional piece. Whether AI assistants can reach and cite your pages depends on crawler access and content.
- 15
- links requested per check
- 1
- required element: the H1
- €0
- no signup
What the check tests
llms.txt is a proposal published on llmstxt.org in 2024. The check follows its format and tells you where a file deviates from it.
- 01
File and response
Loads
/llms.txtover https, follows up to three redirects and checks status code and content type. An HTML page at that address, a common CMS fallback, is flagged. - 02
Title and summary
An H1 with the site or project name as the first line: the only required part. A blockquote summary right below it is recommended.
- 03
Sections and link lists
H2 sections with list items in the form
- [name](url): notes. Items that don’t match, and headings or prose where lists belong, are reported with line numbers. - 04
The Optional section
## Optionalhas a special meaning: tools may skip it when they need a shorter context. The check shows what is in it. - 05
Links, requested live
Relative or absolute, your own host or external, and up to 15 links requested to confirm they load without errors or redirects.
- 06
Size and llms-full.txt
File size, and whether
/llms-full.txtexists, a related convention that is not part of the proposal.
Build your llms.txt
Fill in the fields. The file on the right updates as you type. Links go one per line as Title | URL | note; the note is optional. Lines already in llms.txt format ([Title](URL): note) work too.
Save it as llms.txt in the root of your site so it loads at https://yourdomain/llms.txt, then run the check above on your domain.
Questions
What is llms.txt?
A markdown file in the root of a site, proposed by Jeremy Howard in September 2024 on llmstxt.org. It gives language models a short summary of the site and a curated list of links to its most useful pages, ideally in markdown. It is a proposal, not an adopted web standard.
Do Google, ChatGPT or Perplexity use it?
None of them has documented that it uses llms.txt to find, rank or cite content. Google’s Search documentation says you don’t need new machine-readable files or AI text files to appear in its AI features, and Google staff have said publicly that Google Search does not use llms.txt. The file is most common on developer documentation, where people point coding assistants and other AI tools to it.
Should I create one anyway?
If you run documentation, an API or a software product, it takes little time and helps people who feed your docs to an AI tool. For a typical business site or shop, don’t expect a measurable effect on traffic or AI citations. Crawler access, content in the raw HTML and being mentioned on other sites matter far more.
Where does the file go?
In the root of the domain, so it loads at https://yourdomain/llms.txt. The proposal also allows files in subpaths such as /docs/llms.txt, which cover the URLs below them. Serving it as text/plain or text/markdown with UTF-8 avoids encoding problems.
What belongs in the Optional section?
Links that are useful but not essential: changelogs, older versions, background reading. By convention, tools can drop this section when they need a shorter context, so your key pages belong in the regular sections.
Want to know whether AI search mentions you?
The AI Search Visibility Audit measures whether ChatGPT, Perplexity and AI Overviews cite your site for the questions your buyers ask, and what to change so they do.