What does the "llms.txt is actually useful" check mean?
It sits in the Agent Contract pillar of the Agent Experience rubric and is worth 6 points within that pillar, which works out at about 1.6 points of the overall score out of 100.
Agent Contract · llms-txt-quality
llms.txt is actually useful
What it is worth
| Check id | llms-txt-quality |
|---|---|
| Pillar | Agent Contract — Have you told agents the rules: what they may use, where the docs are, who to contact? |
| Weight within the pillar | 6 points |
| Share of the overall score | about 1.6 of 100 |
| Kind of change | Engineering — a code change and a deploy |
Two numbers rather than one, because a check's weight is inside its pillar and the pillar is a share of the whole. The largest check in the smallest pillar moves your score less than a middling one in the largest, and knowing which you are looking at is the difference between fixing the right thing and fixing the loudest thing.
Why an agent cares
Worth saying plainly: this is not a search-engine ranking factor and passing it will not move you up a results page. It measures something else — whether software with no browser, no JavaScript and a few hundred tokens of budget can tell what your page is and do something with it. Those are the conditions an AI assistant reads you under, and they are much harsher than a browser's.
How to satisfy it
This check does not carry a copy-pasteable snippet, and the reason is worth stating rather than papering over: what it measures is a property of your server or your whole document, not a tag that can be pasted into it.
Scan the page and the report will say what was actually observed — the header that was missing, the status that came back, the passage that could not stand alone — which is the part that tells you what to change. A generic snippet here would be a guess about a site nobody has looked at.
How often the web fails this
Measured on 2026-09-06 against AX specification v1.5.1, across a sample of 292 public websites.
| Result | Sites | Share of those assessed |
|---|---|---|
| Failed | 0 | 0% |
| Warned | 7 | 11% |
| Passed | 57 | 89% |
| Did not apply | 228 | excluded from the shares |
The last row is kept out of the percentages deliberately. A check that did not apply to 228 of 292 sites is not a check most of the web passes, and folding those in is the commonest way this sort of table is used to say something untrue.
At 0% this is something most sites already get right, which makes failing it more costly than the point value suggests — you are behind the field rather than level with it.
Testing it on your own site
npx axray-cli your-site.com
All 70 checks against one URL, this one among them, with what was found rather than only whether it
passed. Add --min 70 in a pipeline and the command exits non-zero below that score, so a
change that breaks what agents can read fails the build instead of shipping.
Questions about this check
Is it unusual to fail llms-txt-quality?
No. Of 64 public sites where this check applied, 0 failed it — 0%. The measurement was taken on 2026-09-06 against AX specification v1.5.1, and no site is named.
How do I test llms-txt-quality on my own site?
Run npx axray-cli your-site.com from a terminal. It runs all 70 checks against one URL and prints this one among them, with what it found rather than only whether it passed. It is free, needs no account and installs nothing.
The rest of the Agent Contract pillar
Have you told agents the rules: what they may use, where the docs are, who to contact?
- Publishes an llms.txt — 14 points
- States an explicit AI policy — 6 points
- Declares AI usage preferences the standard way — 5 points
- Exposes an agent-callable manifest — 6 points
- Identity is machine-verifiable — 5 points
- Says when the content changed — 5 points
- Declares content licensing — 5 points
- Publishes security.txt — 4 points
See how your own page does on this check and the other 69, with the fixes ordered by how many points each one recovers. One scan, no account.