Free · no signup · no login
Can AI actually read your site?
Most checks stop at the status code. We send requests from 16 real crawler user agents, then measure what came back — because a page can return 200 and still give a crawler almost nothing to read.
Takes about 15 seconds. Nothing is stored against you, and no signup is required.
What we fetch: your homepage with 16 crawler user agents and one browser user agent, plus robots.txt, llms.txt, sitemap.xml, and three pages picked from your sitemap.
Common questions
- Does a 200 mean AI can read my site?
- No. A bot-check page can return 200 with nothing but a JavaScript challenge in it. A single-page app can return 200 with an empty shell. Both look identical to a checker that stops at the status code — which is why this one measures the body.
- Is blocking GPTBot bad for me?
- Not necessarily. GPTBot collects training data; OAI-SearchBot is the one that feeds answers with citations. Blocking the first and allowing the second is a coherent position, and this check treats it as one rather than marking everything red.
- Why did you check pages other than my homepage?
- Because homepages and inner pages usually fail differently. Homepages are built around layout and often assemble their text in the browser; inner pages are built around content, so the text is already in the HTML. Checking only one of them gives you half an answer.
- Do I need an llms.txt?
- No engine requires one today. What matters far more is that your robots.txt is right, your content is in the HTML, and your pages have headings and structured data.
- Do you store my site’s data?
- The report is kept briefly so the page can be reloaded and so repeat checks of the same domain do not keep hitting your server. No account, no email, and nothing that identifies you is attached to it.
Once crawlers can read you
Getting in is step zero. The next question is which searches you are close to winning, and what to change to win them.