AI Search Readiness Checker

By AISEORainmaker.com Editorial TeamPublished 21 Oct 2026Checks reviewed 21 Oct 2026

Affiliate disclosure: the checker is free and stays free. We earn a commission if you join AI SEO Rainmakers through our links. That does not change the price you pay.

What the score means

The score is out of 100. Each check carries a weight based on how completely it can shut a page out of AI answers. A page that fails the first or third check is invisible to the engines it blocks, whatever else it does well.

When a check cannot be judged, for example because every bot request was rate limited, we leave it out and score the rest. The result says so.

Check 1: AI crawlers allowed in robots.txt

We read your robots.txt the way the crawlers do under RFC 9309: each bot follows the group that names it, or the * group if none does, and the longest matching rule wins. Then we test the exact page you entered against nine crawlers.

Six of them fetch pages to answer a live question: Googlebot (Google Search, AI Overviews and AI Mode), Bingbot (Bing and Copilot), OAI-SearchBot and ChatGPT-User (ChatGPT), PerplexityBot and Claude-SearchBot. Blocking any of these removes the page from that engine's answers, so the score drops for each one.

The other three, GPTBot, ClaudeBot and Google-Extended, collect training data. Plenty of publishers block them on purpose. We list them, and we do not take points off. Google states that Google-Extended has no effect on Search, which includes AI Overviews.

Fix: remove the Disallow line from the group that names the blocked crawler, or add an Allow: / group for it. Then run the check again.

Check 2: Firewall lets AI bots through

robots.txt is a request. A firewall is a wall. We fetch your page four more times, sending the published user agents of GPTBot, OAI-SearchBot, PerplexityBot and ClaudeBot, and compare each answer with the one our normal request got.

This catches the most common silent block we see: bot protection switched on at the host or CDN. Cloudflare began blocking AI crawlers by default for new domains in July 2025, so a site can allow every bot in robots.txt and still refuse all of them.

One limit matters here. Our requests come from our server, not from OpenAI's or Perplexity's published address ranges. A firewall that blocks fake bots and lets verified ones through will fail this check even though the real crawler gets in. That is why a failure here is a warning, not a zero. A rate limit (HTTP 429) counts as neither.

Fix: in Cloudflare, open Security, then Bots, and check the AI crawler setting. On other hosts, look for a bot-fight or AI-scraper toggle in the firewall panel.

Check 3: Page can be indexed

Google's AI Overviews and AI Mode draw on the Google index. Copilot draws on Bing's, and ChatGPT search leans on Bing results too. A page carrying noindex in a meta tag or an X-Robots-Tag header falls out of those indexes. We also flag a canonical tag that points somewhere else, because search engines will usually index that other address instead.

Fix: remove noindex from any page you want cited. Point the canonical at the page itself unless you mean to consolidate it.

Check 4: Content readable without JavaScript

We count the words in the HTML your server sends, before any script runs. Vercel's study of crawler traffic, "The rise of the AI crawler", found that OpenAI's and Anthropic's crawlers (GPTBot, OAI-SearchBot, ChatGPT-User and ClaudeBot) fetch JavaScript files without running them. Googlebot renders pages, but most AI crawlers read only what arrives in the first response.

Under 50 words fails. Between 50 and 199 is a warning. 200 or more passes.

Fix: render the main content on the server. Most frameworks (Next.js, Nuxt, Astro, SvelteKit) do this with a setting or by default. A static site passes already.

Check 5: Organization schema with sameAs

We look for Organization (or a subtype such as LocalBusiness) or Person markup in the JSON-LD on the page you entered, then on your homepage. Full marks need at least one sameAs link to a profile you control elsewhere. Those links let an engine match the site to the same entity on LinkedIn, Crunchbase, X or Wikipedia, and an engine that knows who you are is more willing to name you.

We also report any JSON-LD block that fails to parse. Search engines skip a broken block entirely.

Fix: add an Organization node to your homepage with your name, logo, URL and a sameAs list of your official profiles.

Why llms.txt is not scored

The checker tells you whether /llms.txt exists, and gives it zero points. No major AI engine has said it reads the file, and in the AI SEO Rainmakers classroom, Charles Floate reports that no model requested it in his own tests. A file nobody reads should not move your score.

What a high score does not prove

A perfect 100 means the door is open. It does not mean anyone walks through it. Whether ChatGPT or Perplexity names your brand depends on what is written on the page, who else mentions you and how often, and none of that shows up in a crawl test.

That second half is what AI SEO Rainmakers teaches. Crawler access and the entity schema graph are week-one items in Charles Floate's entity stack checklist inside the community. The rest of the classroom covers measuring which prompts cite you, writing passages an engine can lift, and earning the mentions that make it pick you. Read our full review, see what the curriculum covers or check the current price first.