Alexey Shurov.
Free tool

Can AI assistants read your website

More buyers now ask an assistant before they search. Type your address and see in a few seconds whether an assistant can find your site, read it and describe it correctly.

It reads the home page and a few public files, the same ones an assistant would look at. Nothing about your site is kept.

What it checks

Eight things, in the order they matter. Whether the site answers over a secure connection. Whether your robots.txt shuts out the assistants that fetch pages to answer a question. Whether the home page tells them not to list it. Whether the words arrive with the page or only after scripts run. Whether the page has a title and a description. Whether it describes itself in structured data. Whether there is a sitemap. And whether there is a plain index for assistants.

It also notes three things for information only. Whether crawlers that collect text for training are shut out, whether the page offers a plain text copy, and whether the site describes anything an assistant could do for a person, not only read.

What the result means

Fix this means an assistant cannot read or list the site until it is changed. Could be better means the assistant can read the site but has less to go on, so it may describe you loosely or leave you out.

Passing every check does not make an assistant recommend you. It means nothing stands in the way. What it says about you then depends on what your pages say, how specific they are and who else writes about you.

What it does not do

It reads the home page and a few public files. It does not crawl the whole site, log in, fill in forms or test anything behind a password. It does not store the address you type or what it found.

It cannot tell you how often assistants mention you today. That needs the same buyer questions asked week after week, which I do for my own site and can set up for yours.

Common questions

How do I know if ChatGPT or Claude can read my website?

Check four things. That robots.txt does not shut out the assistants that fetch pages to answer a question, that the page is not marked noindex, that the words arrive with the page itself and not only after scripts run, and that the site answers over a secure connection. This tool checks all four in a few seconds.

Should I block AI crawlers in robots.txt?

There are two kinds. Crawlers that collect text to train models, which you may block if you wish. And assistants that fetch a page when a person asks a question, which you should let in if you want to appear in answers. Blocking the second kind hides your site from those answers.

What is llms.txt and do I need it?

It is a short plain text file at /llms.txt that lists your main pages for assistants. Some assistants look for it. It does not change your position in Google. It is cheap to add and worth having once the more important checks pass.

Why does it matter that the page works without scripts?

Most assistants read the page as it arrives and do not run scripts. If your text only appears after scripts run, the assistant sees an almost empty page and has nothing to quote.

Does the tool keep my website address?

No. It reads a few public files from the address, shows you the result and keeps nothing about the site. It counts only that the tool was used.

Want assistants to describe you correctly

I make sites and products readable and usable by AI assistants, and I measure whether they are being found. Tell me your site and what you want an assistant to say about it.

shurco.aiGuidesFree toolsSolutionsInsightsWork with meRSSllms.txt