roberto.iriondo Get in touch

AI visibility checker

Can ChatGPT, Claude, Perplexity, Gemini and Google reach your site, read it and know who you are? Enter your site to get a score out of 100 and the fixes in order, in plain language.

Checks the page you enter (your homepage if you enter a domain) and the site around it. Takes about 10 seconds.

What the checker looks at

AI assistants answer from pages they can reach and read, and they prefer sources they can tie to a known company or person. The score has three parts:

  • Reach (52 points). The page answers, is not set to noindex, and robots.txt and the server let in Google, Bing and the crawlers behind ChatGPT, Claude and Perplexity answers.
  • Understand (35 points). The text is in the HTML, not drawn only by JavaScript, which most AI crawlers do not run. The page has a title and description, structured data that names the company and links its profiles, a sitemap and an llms.txt.
  • Recognized (13 points). The company has a Wikidata item that lists this site as its official website, a Wikipedia article, and pages in Common Crawl, which many AI models learn from.

Training crawlers such as GPTBot and Google-Extended are shown but not scored: blocking them is a fair choice, and it does not stop AI assistants from citing you.

Check it by hand

If the checker cannot reach a site, or you want to see for yourself:

  1. Open yoursite.com/robots.txt and look for Disallow: / under User-agent: *, OAI-SearchBot, ChatGPT-User, Claude-SearchBot, PerplexityBot, Googlebot or Bingbot.
  2. If you use Cloudflare, look under Security for bot settings that block AI crawlers. Other firewalls and security plugins have similar settings.
  3. View the page source (or use View source on mobile) and check your main text is there. If the body is an empty <div id="root">, AI crawlers see an empty page.
  4. Search the source for noindex and for application/ld+json, and check the structured data names your company with its website and profiles (sameAs).
  5. Search Wikidata for your company and check its "official website" matches your domain.

How it works

A small server, a Cloudflare Worker, fetches the page once as a browser and once as each crawler, then reads robots.txt, llms.txt and the sitemap. It looks the site up in Wikidata and Common Crawl, both free public services. Your browser does the scoring, so the rules are the same for every site. It costs nothing to run: no AI model is called, results are kept for an hour, and there is a per-visitor limit.

Want your site set up to be found and described correctly by AI assistants? Get in touch.

Questions

How do I get my site to show up in ChatGPT, Perplexity or Google AI answers?

AI assistants can only cite pages they can reach and read. Let their search crawlers in through robots.txt and your firewall, make sure the page text is in the HTML and not only drawn by JavaScript, describe the company in structured data, and make sure the company is listed where AI models look things up, such as Wikidata. The checker tests each of these and lists the fixes in order.

Should I block GPTBot and other AI training crawlers?

That is a business choice. Blocking training crawlers such as GPTBot, ClaudeBot, Google-Extended and CCBot keeps your pages out of future model training. It does not stop ChatGPT, Claude or Perplexity from citing you in answers, which use separate crawlers: OAI-SearchBot, ChatGPT-User, Claude-SearchBot, Claude-User, PerplexityBot and Perplexity-User. Google-Extended does not affect Google Search.

My robots.txt allows AI crawlers. Why does the checker say they are blocked?

robots.txt is a request; the server or a firewall such as Cloudflare decides who actually gets in. Bot protection, an AI crawler block or a security plugin can refuse crawlers that robots.txt allows. The checker asks for your page as each crawler and shows the answer it got. Some firewalls also check the crawler's real network address, so the real crawler may be treated differently from the checker.

Does llms.txt matter?

A little. llms.txt is a short plain-language summary of a site for AI tools. Few AI crawlers read it today, so the checker gives it a small weight. It takes a few minutes to write and does no harm.