Free tool
Can ChatGPT read your website? Check whether AI assistants can read it
Enter your website’s domain and we’ll tell you whether the bots from OpenAI, Google, Microsoft, Anthropic and Perplexity can get in, and what to fix if they can’t: the robots.txt, the home page as a bot sees it, the sitemap and the CDN.
It doesn’t tell you whether ChatGPT or Gemini recommend it, because that depends on many more things and changes every day.
Free, with no sign-up. The domain and the result stay in the server’s memory for an hour, so it doesn’t have to be checked again; they aren’t stored anywhere else, and your IP isn’t stored.
What now?
If you’d like us to look at it with you, there are two ways.
Thirty minutes, free and with no obligation, with the engineer who would lead the project.
Leave us your email
We’ll reply personally, with no automated emails, with what we would do on your website.
Received. We’ll reply personally to the email address you left us.
What it checks and how
- The robots.txt, for thirteen bots from OpenAI, Google, Microsoft, Anthropic, Perplexity, Apple and Common Crawl, with the rules of the protocol (RFC 9309): if a bot has its own group, the general group (“*”) no longer applies to it.
- The home page, as a bot that doesn’t run JavaScript sees it: the response, the redirects, HTTPS, whether it can be indexed, the canonical URL, the title, the description, the language, the structured data, how much text there is in the HTML and how long it takes.
- The sitemap, the llms.txt and the CDN or firewall in front of the website.
- It uses an agent that says who it is (SanfeliuVisibilitat/1.0) and doesn’t pretend to be any bot: real bots are recognised by their IP, and the result would be false.
- At most eight requests to your website per check, and ten checks per hour from each IP address.
- The domain and the result stay in the server’s memory for an hour, so it doesn’t have to be checked again, and are then deleted. Your IP isn’t stored.
Frequently asked questions
Does it tell me whether ChatGPT recommends my website?
No. It tells you whether AI assistants can read it. Whether they recommend it depends on many more things and changes every day.
Why doesn’t it pretend to be GPTBot to check?
Because the result would be false: real bots are recognised by the IP they use, not by their name. A website may let through a “GPTBot” that isn’t one, or stop it. That’s why the tool reads the robots.txt and tells you what it says for each bot.
Do you keep the data?
The domain and the result stay in the server’s memory for an hour so that, if you check it again, it doesn’t have to be done again. The IP isn’t stored. If you leave us your email, it goes to our CRM so we can reply to you.
I’ve changed the robots.txt and the result hasn’t changed
The result for each domain is kept for an hour. Also, OpenAI says a change to the robots.txt can take about 24 hours to take effect in ChatGPT search.
Tell us about your case in thirty minutes.
Book a meeting with the engineer who would lead the project: thirty minutes, free and with no obligation. If we’re a fit, we tell you how we’d start; if not, who to turn to.
Or email me@joansanfeliu.com
Prefer to write?
Tell us about your case in a few lines and we’ll reply by email.
The engineer who would lead the project will reply to the email you gave us. If you like, book a thirty-minute call now.