Shruti Kaushik

ToolsFree tool

Can ChatGPT and Perplexity read your website?

Enter your website and this free checker runs 11 tests that affect whether ChatGPT, Perplexity and Google AI answers can read, understand and cite it: AI crawler access in robots.txt, server-rendered text, schema, an answer-first opening, sitemap, llms.txt and more. You get a score out of 100 in seconds, and the five fixes that matter most.

Free, no sign-up. The check reads only public pages, the same way an AI crawler does.

What the checker tests, and why

Each check is pass or fail, and the score is the share passed, out of 100. The fix list puts first whatever stops an engine reading the site at all: blocked crawlers and pages that only render in the browser.

  1. AI crawlers allowed in robots.txt. Whether robots.txt shuts out GPTBot or OAI-SearchBot (OpenAI), PerplexityBot, ClaudeBot (Anthropic) or Google-Extended. OpenAI says OAI-SearchBot is the crawler behind ChatGPT search results and GPTBot gathers training data; PerplexityBot indexes pages for Perplexity answers.
  2. Server-rendered text. Whether the HTML your server sends holds at least 100 words of text. Vercel’s crawler study (17 December 2024) found that none of the major AI crawlers, including GPTBot, ClaudeBot and PerplexityBot, run JavaScript, so a site that only renders in the browser looks empty to them.
  3. Direct answer under the H1. Whether the first paragraph after the H1 is 25 to 80 words: long enough to answer something, short enough to be quoted whole.
  4. Organization or Person schema with sameAs. Whether JSON-LD names the company or person behind the site and links their other profiles, so engines can tie the site to one known entity.
  5. FAQPage or Article schema. On the page you enter or the first blog, article or guide page it links to. Google limited FAQ rich results to a few site types in August 2023, but the markup still labels questions, answers, authors and dates for machines.
  6. Title and meta description. Both set, so every engine has a short, accurate summary of the page.
  7. HTTPS. The page ends up on an https address after redirects.
  8. Sitemap. A real sitemap at /sitemap.xml or on the Sitemap line of robots.txt, so crawlers can find every page.
  9. Canonical tag. Tells engines which address to cite when the same page is reachable at several URLs.
  10. Open Graph tags. og:title, og:description and og:image, which shape how a cited link looks when it is shared.
  11. llms.txt. A plain Markdown summary of the site for language models, proposed in September 2024. No major engine has confirmed it reads the file, so it comes last in the fix order.

What a good score does not tell you

This is a technical check of one page and its site files. It does not show whether ChatGPT or Perplexity name you when buyers ask about your category, or which pages they cite instead. For which vendors Google AI Mode and Perplexity name in 12 SaaS categories, and every page they cite, see Who AI recommends; for how I track my own name, see the AI visibility case study.

Quick questions

What does the AI search readiness checker test?

Eleven things that affect whether AI answer engines can read and cite a site: AI crawler access in robots.txt, server-rendered text, an answer-first paragraph under the H1, Organization or Person schema with sameAs links, FAQPage or Article schema, title and meta description, HTTPS, sitemap, canonical tag, Open Graph tags and llms.txt. The score is the share of checks passed, out of 100.

Does a score of 100 mean ChatGPT will cite my site?

No. A full score means nothing technical stops AI crawlers reading and understanding your pages. Whether an engine names you depends on whether your pages answer the question and whether other sites, threads and listicles mention you. The free AI visibility check tests that directly.

Which AI crawlers does it look for in robots.txt?

GPTBot and OAI-SearchBot from OpenAI, PerplexityBot, ClaudeBot from Anthropic, and Google-Extended. A site fails if robots.txt disallows the whole site for any of them, through their own group or the catch-all group. Google says Google-Extended does not affect inclusion or ranking in Google Search, which follows Googlebot.

Why does a JavaScript-only site fail?

Because most AI crawlers read only the HTML the server sends. Vercel found in December 2024 that none of the major AI crawlers execute JavaScript, so text that only appears after scripts run is invisible to them.

Is the website I check stored?

The checker fetches the public page, robots.txt, sitemap and llms.txt, returns the result and keeps no copy of the pages. The address you checked is logged for 14 days to stop abuse. If you ask for the full fix list, your name, work email, company and the result are sent to me.

Built by Shruti Kaushik. Related: the founder AI visibility guide and the free AI referral tracker.