Is your llms.txt in the right place?
Enter your domain. We look for yourdomain.com/llms.txt, check that it can be reached, and check the file against the published format.
Check your llms.txt
Looks for /llms.txt on your domain · 2 free checks a day
Where llms.txt goes
- At the root
- yourdomain.com/llms.txt. This is the place this checker looks.
- Or in a subfolder
- The format also allows one in a subfolder, such as yourdomain.com/docs/llms.txt, where the most specific file applies. This checker looks at the root only.
- As plain text
- A text file in Markdown, answered with status 200. A web page in its place is a common mistake: sites without the file often answer with their not-found page.
What a valid file looks like
Only the title is required. The summary, the link sections and the Optional section are all optional, and this checker treats them that way.
# Example Plumbing
> Family-run plumber in Aarhus. Emergency call-outs, bathroom renovations and boiler servicing.
## Key pages
- [Services](https://example.com/services): What we do and what it costs.
- [Contact](https://example.com/contact): Opening hours and how to reach us.
## Optional
- [About us](https://example.com/about): Who we are.
Questions
A plain text file in Markdown, named llms.txt, that you publish on your own website. It gives an AI model a short description of the site and a list of links to the pages worth reading. It is an open proposal that anyone can use, not a rule that any company has to follow.
Enter your domain above. We request yourdomain.com/llms.txt, follow any redirects, and tell you whether the file was found, whether the server answered, how it was served and whether it matches the format. If you have not published it yet, paste it instead.
Only what stops a model reading the file at all is an error: no title, more than one title, or a file that is empty or is actually a web page. Everything else is a warning or a note, because the format requires only the title.
No. A valid file means the format is right and the file can be reached, nothing more. Whether a given assistant reads an llms.txt file at all is that assistant's choice, and nothing here can change it.
No. robots.txt tells crawlers which parts of a site they may visit. llms.txt describes the site, in text, for a model to read. Having one does not change the other.
No, they do different jobs. llms.txt is plain text that describes your website to a model. An ARD file (ard.json) is a different open standard: a file on your own domain that tells AI assistants what your business offers. An MCP server is the piece that lets an AI assistant ask your business a question and get a real answer back.
llms_txt_checker_page.faqs.faq7.answer
Also have an ARD file? Check it with the ARD validator →New to llms.txt? Read what it is →
Want AI assistants to be able to ask your business directly?
Describe what you offer in a guided setup. The platform builds an MCP server, the piece that lets an AI assistant ask your business a question and get a real answer back, then hosts it and registers it. No code.