llms.txt generator & validator
Enter a domain. We fetch its /llms.txt, check it against the spec, and hand you a starter file to publish. No sign-up, no email gate.
What a good llms.txt looks like
The spec is small on purpose. One H1 naming the site, one blockquote summarising it, then H2 sections of markdown links where every link carries a short description of what the page answers. That description is the part most files get wrong — it is what tells an agent whether a URL is worth spending context on.
# Acme > Acme builds warehouse robots for mid-market logistics teams. ## Pages - [Home](https://acme.com/): What Acme does, in one page. - [Pricing](https://acme.com/pricing): Plans, prices and what each includes. ## Optional - [Blog](https://acme.com/blog): Articles and updates. Safe to skip.
A valid llms.txt is worthless if AI crawlers cannot reach you
Plenty of sites publish a perfect llms.txt while their robots.txt quietly blocks GPTBot and ClaudeBot. Check that side too — it takes one click and is also free.
Check AI crawler accessllms.txt questions
What is llms.txt?
A plain-markdown file at the root of your site (yourdomain.com/llms.txt) that gives an LLM a curated map of your most useful pages, each with a one-line description. It is the same idea as robots.txt or sitemap.xml, but written for a model that has to decide what to read under a context limit rather than for a crawler that indexes everything.
Is llms.txt an official standard?
No. It is a community proposal published at llmstxt.org, and no major AI provider has publicly committed to reading it. Anyone telling you it guarantees citations is overselling it. It is cheap to publish and it structures the same information you should be giving crawlers anyway, which is why we ship one — but it is a bet, not a rule.
What does this validator actually check?
It fetches your /llms.txt and checks the structural elements the spec defines: a single H1 title, a blockquote summary, H2 sections, markdown link lists in the form - [name](url): notes, absolute URLs, descriptions on every link, plain markdown with no raw HTML, and an optional section for URLs an agent may skip. It reports what it observed and nothing more.
My site returns 200 for /llms.txt but you say it is not a real one. Why?
Many sites have a catch-all route that answers 200 with the app shell for any unknown path. If we get HTML back instead of text, the file effectively does not exist — an agent asking for llms.txt would receive a web page. We flag that explicitly rather than counting it as a pass.
Do I have to sign up?
No. There is no account, no email gate, and no result held back. The tool is rate-limited per IP to keep it available, and that is the only restriction.
Does publishing llms.txt help my SEO?
Not directly — search rankings are unaffected. Its purpose is generative engines. Treat it as one input to AI visibility alongside letting AI crawlers reach your pages at all, which you can check with our AI crawler tool.
Publishing the file is step one.
Growzo keeps your llms.txt, sitemap and AI-crawler access correct as your site changes — and tracks whether the models actually cite you.
Open-source · Self-host · BYOK · ₹10k/mo hard cap