Free tool · August 2026

Is your site AI-ready?

Buyers ask ChatGPT, Claude and Perplexity before they ever see your homepage. This free check fetches your page and runs six mechanical tests for the groundwork those engines depend on: structured data, llms.txt, robots sanity, title and description shape, viewport, and content shaped like answers. Letter grade plus evidence for every check, in seconds, no email gate.

Free, instant, no email needed for the result. We fetch your page once and run the checks; nothing is stored.

Checks run
6, mechanical
Price
Free, no email
Result
Grade + evidence

Quick Answer: At a glance

The claw.mobile AI-Ready Check is a free instant test of whether a website gives AI engines what they need: it fetches the page and runs six mechanical checks (JSON-LD structured data, llms.txt, robots and noindex sanity, title and meta description shape, viewport tag, and question-shaped headings or FAQPage schema), then returns a letter grade from A to F with one line of evidence per check. No AI calls, no email gate, result in seconds. As of August 2026 it complements the paid $99 AI Visibility Scan, which measures what ChatGPT, Claude, Perplexity and Grok actually say about a business.

The six checks

Every check is mechanical: it inspects what your page actually serves, not what a model thinks of it. Pass or fail, with the evidence stated, so you know exactly what to fix. The full rules, thresholds included, are on the methodology page.

  • Structured data (JSON-LD)

    We parse every application/ld+json block on your page and list the schema.org types we find. Structured data is the machine-readable version of what your site is; engines use it instead of guessing.

  • llms.txt

    We request /llms.txt on your domain and verify it answers with real plain text, not an HTML catch-all page. Optional today, but it is the cheapest way to hand crawlers a map of your site.

  • Robots and noindex sanity

    We look for a noindex robots meta tag, an X-Robots-Tag noindex header, and a robots.txt that disallows everything. Any of the three tells engines to leave, and it overrides everything else you do: a site that fails this check is capped at grade D.

  • Title and meta description shape

    Title present and 15 to 70 characters, meta description present and 50 to 170. Those are the ranges engines quote cleanly instead of improvising their own summary from whatever text they grab first.

  • Viewport meta tag

    A one-line page-quality signal. Its absence usually means the page was never finished for real visitors, and engines read it the same way.

  • Answerable content signals

    We count h1 to h3 headings phrased as questions and look for FAQPage schema. AI answers are assembled from content that asks and answers real questions; a page with neither gives the engines nothing to quote.

The next question

Ready is not the same as recommended

This page checks what your site gives the engines to work with. What the engines actually say when your buyers ask is a separate measurement: the AI Visibility Scan puts 25 real buyer questions for your niche to ChatGPT, Claude, Perplexity and Grok and reports who gets recommended, who gets cited, and where you fit. $99, self-serve, report in hours.

See what the AIs say about you →

Frequently Asked Questions

What does the AI-ready check actually test?
Six mechanical checks, no AI involved: JSON-LD structured data types present, a real llms.txt, robots and noindex sanity (no noindex meta tag, no noindex header, no robots.txt blocking everything), title and meta description shape (15-70 and 50-170 characters), a viewport meta tag, and answerable-content signals (headings phrased as questions or FAQPage schema). Each check returns pass or fail with one line of evidence, and the passes add up to a letter grade from A to F.
Is it really free? Do I have to give an email?
Yes and no. The check is free, the result renders on the page instantly, and no email is needed to see it. There is an optional signup afterwards for the monthly AI recommendation report if you want to follow the data over time, and it is genuinely optional.
How is the grade calculated?
One point per passed check, six checks total: 6 passes is an A, 5 is a B, 4 is a C, 3 is a D, and 2 or fewer is an F. One override: a site that fails the robots check (noindex or a blanket robots.txt block) is capped at D whatever else passes, because telling engines to leave defeats everything else.
Why did the check refuse my URL?
We only fetch public http or https sites. URLs that resolve to localhost, private networks, or other non-public addresses are refused before any request is made, and the same rule is enforced again on every redirect. If your site is behind a login or a strict bot wall, the fetch can also fail even though the URL is fine.
My site got an A. Does that mean ChatGPT recommends me?
No. An A means the mechanical groundwork is in place: engines can read your site and its content is shaped like answers. Whether the AIs actually name you when buyers ask is a different measurement, and that is what the AI Visibility Scan does: it puts real buyer questions to ChatGPT, Claude, Perplexity and Grok and reports who their answers recommend, you included.
What is llms.txt and do I need one?
llms.txt is a plain-text file at the root of your domain that gives language-model crawlers a curated map of your site: what it is, what matters, where the good pages are. It is an emerging convention, not a requirement, which is why it is one check out of six and not a gate. It usually takes ten minutes to add.