llms.txt Validator
Check any domain’s llms.txt for the mistakes that stop an assistant using it properly.
No sign-up. We read the public file for you. Nothing is stored.
What you get
- A graded pass or fail on every rule
- The exact line that failed each one
- Missing descriptions and duplicate entries
- Relative links that will not resolve
- That the file has exactly one H1 and a summary block, the two elements the proposed format treats as required.
- That every link is absolute, since a relative path does not resolve against your domain in an assistant’s context.
- That each entry carries a description rather than a bare link.
- That there are no duplicate entries and no empty sections, both of which cause an assistant to skip content.
How to use the llms.txt Validator
Enter your domain
We fetch the file at /llms.txt. Checking a draft? Paste it instead and nothing is uploaded.
Read the graded checks
Each one names the rule and shows the line that failed it.
Fix and re-run
The file is short enough that a clean pass takes one iteration.
Why an llms.txt needs validating at all
It fails without a symptom.
An assistant that cannot parse the file ignores it. Nothing errors, nothing logs, and you carry on believing it works.
Relative links are the usual culprit.
A relative path resolves against the assistant’s context, not your domain, so it points nowhere.
A bare link gets skipped.
Without a description, an assistant has nothing to judge relevance by and moves on to a page that explains itself.
The format is young.
The spec is a proposal, not a standard, which makes a checker more useful here than it would be for something settled.
What to do with the result
llms.txt fails quietly: an assistant that cannot parse it simply ignores it, and nothing tells you. That makes validation the whole value, because the failure mode has no symptom. Once it parses, the open question is whether the pages you nominated are the ones actually being cited, which is what AI Visibility Tracking answers.
AI Search IntelligenceEmbed this tool
Paste this into any page. It loads the llms.txt Validator without our navigation, with a one-line credit under it. Free to use on agency sites, client reports and course pages. Keep the credit line.
<iframe src="https://www.getveritas.io/embed/llms-txt-validator" width="100%" height="980" style="border:0;max-width:100%" title="llms.txt Validator by Veritas" loading="lazy"></iframe>
<p style="font:13px/1.5 system-ui,sans-serif;margin:6px 0 0">Free <a href="https://www.getveritas.io/tools/llms-txt-validator">llms.txt Validator</a> by Veritas</p>Frequently asked questions
No. llms.txt is a community proposal, not a ratified standard, so nothing can claim to validate it in the way an XML schema validates a sitemap. These checks test something more useful: whether an assistant that fetches only this file can tell what your site is, find the right page, and resolve every link.
More free tools
llms.txt from Sitemap
Turn any domain’s sitemap into a first-draft llms.txt, grouped into sections automatically.
llms.txt Generator
Build the llms.txt file that tells ChatGPT, Perplexity, and Claude which of your pages to cite.
robots.txt Checker
Test any URL against any crawler, see the exact rule that decides it, and find the lines crawlers ignore.
See where you actually stand.
A free tool checks one thing once. Veritas watches all of it, continuously, across every engine that decides whether you get cited.
GOOGLE · CHATGPT · PERPLEXITY · GEMINI · AI OVERVIEWS