Check what AI
assistants can read.
If an assistant cannot fetch your home page, it cannot quote what is on it, and nothing on your site will tell you. Paste your address and we fetch your home page, robots.txt, sitemap and llms.txt once, the way a crawler would, then show you each check that passed and each that did not. It takes a few seconds and stores nothing.
Getting in
Reading the page
Knowing who you are
Training crawlers, not counted.
These mostly decide whether a model may learn from your words, not whether ChatGPT, Claude, Perplexity, Google Search or Copilot may fetch your page and cite you. One exception: Google-Extended also decides whether Google's Gemini app may use your pages in its answers. They do not change your score, and our own robots.txt lists them separately.
How the check works.
-
You paste your address
Your main web address is all it needs. There is no sign-up and no email to run it.
-
We fetch it once
Our server requests your home page, robots.txt, sitemap.xml and llms.txt a single time, the way a crawler would.
-
You see every check
Each one shows as passed, failed or not checked, grouped by what it tells you. Nothing is saved.
What this check
cannot tell you.
Whether an assistant will actually recommend you or cite you. That depends on what the rest of the web says about you, your reviews, and whether your pages answer the question being asked. A clean result means the door is open, not that anyone walks through it. And because only the home page is checked, a problem on a service page will not show up here. It also visits as our own checker, not as each assistant's crawler, so a firewall or security service that turns away a named crawler will not show up here either.
Check questions,
answered.
-
Is this the same as the free audit?
No. This check reads your home page, robots.txt, sitemap.xml and llms.txt in a few seconds and looks only at whether AI assistants can get in and read you. The free audit runs thirty checks across search engines, answer engines and AI assistants on your domain, by hand, and the result is emailed to you.
-
Do you store the address I check?
No. The address is used to run the check and is not saved. If you fill in the report form afterwards, we keep only what you send in it.
-
Why skip training crawlers?
Training crawlers mostly decide whether a model may learn from your words. The crawlers counted above decide whether an assistant can fetch your page when it is asked. For ChatGPT, Claude and Perplexity you can allow one and block the other. Google-Extended is the exception: it also covers the answers in Google's Gemini app. Microsoft has no separate training crawler, so Bing and Copilot are set with page tags instead.
-
Why is llms.txt marked optional?
It is a proposed standard, not an agreed one. We show whether you have one, but it does not count towards your score either way.
-
What does “not checked” mean?
That we could not read that part, for example because the site refused the request or took too long to answer. It is never counted as a pass. Run the check again, or ask for the free audit and we will look by hand.
Get all
thirty checks.
The free audit runs thirty checks across search engines, answer engines and AI assistants on your domain, by hand. It is free, a founder replies personally, and the findings are yours whether you hire us or not.