Check whether your llms.txt exists, parses, and actually links anywhere — read the way an AI
system reads it. Free, no account.
What this checks — and what it refuses to score
The mechanical half: the file exists at the root, returns success, has a title and sections,
and its links are well-formed absolute URLs. That is verifiable from outside and this tool
verifies it.
The other half — whether the summary is accurate, complete and well-written — cannot be checked
by fetching a file, so this tool does not pretend to grade it. A valid file with a bad summary
is still a bad file; that judgement needs a person who knows the site. This site publishes its
own at /llms.txt
if you want an example to compare against. Run the
AI crawler checker
alongside this one to confirm bots may actually fetch what the file points at.
Frequently asked questions
Do I need an llms.txt?
No. It is an emerging convention, not a requirement, and no AI system requires it. HTML remains the source of truth — the file is a concise machine-readable summary that helps systems which look for it navigate a site faster. Missing one is a missed convenience, not an error.
Where does the file live?
At the root of the host: https://example.com/llms.txt. Like robots.txt it is per-host, so subdomains need their own. This tool only checks the root, because that is the only location AI systems look.
What belongs in it?
A single H1 naming the site, a short description, ## sections grouping markdown links to the important pages — documentation, API references, key guides — using absolute URLs. Keep it lean and link to detail rather than pasting everything in; very large files get truncated by the systems reading them.
Will an llms.txt get me cited in AI answers?
No file guarantees citations. Citations come from content that is crawlable, current and worth quoting. What the file does is lower the cost of understanding your site correctly — which matters most for documentation-heavy and product-heavy sites that AI summaries otherwise garble.
Should llms.txt be blocked in robots.txt?
No — blocking it defeats its entire purpose. It must be crawlable by the very bots it is written for. If the AI crawler checker shows your AI bots blocked, fix that before polishing the file.
How does this relate to IndexNow?
They cover different halves of freshness. IndexNow tells participating search engines — Bing included, and Bing's index feeds several AI answer surfaces — that a page changed. llms.txt tells AI systems what a site contains. One announces changes, the other explains the site.