What this tool measures

When someone asks an AI assistant a question, the assistant can only cite pages it has been able to read and understand. That depends on a handful of public, technical signals your site sends out: whether AI crawlers are allowed in, whether there is a machine-readable summary of your content, whether your pages carry structured data, and whether the page structure is clean enough to quote.

This checker fetches your URL (plus the standard files at your site root) and scores eight of those signals from 0 to 100. Each failed check comes with one concrete fix. It is one of the free tools we maintain for freelancers.

What it does not measure

It never queries ChatGPT, Claude or any other AI assistant, so it cannot tell you whether your site currently appears in AI answers. A high score means your site is easy for AI systems to read and cite — it is not a promise that they will. Rankings and citations also depend on your content itself: original experience, clear answers and genuine expertise still do the heavy lifting.

The eight checks, explained

llms.txt — a short Markdown file at your site root summarising what the site covers. AI assistants read it the way humans read an “about” page. If you don’t have one, you’re asking every crawler to figure your site out from scratch.

robots.txt and AI crawlers — your robots.txt decides which crawlers may enter. We check six widely documented AI crawler and opt-out signals: GPTBot, ClaudeBot, PerplexityBot, OAI-SearchBot and CCBot (crawlers), plus Google-Extended (Google’s AI-training opt-out token). Blocking them is a valid choice for some sites, but then you should not expect AI citations either.

Sitemap — a sitemap.xml listing your pages, referenced from robots.txt, so crawlers can discover everything without guessing your URL structure.

Structured data (JSON-LD) — typed facts about the page (an Article, an Organization, a FAQPage) that machines can quote with confidence. Plain HTML tells a reader what a page says; structured data tells a machine what it means.

Head tags — title, meta description, canonical URL and Open Graph tags. These are the labels AI systems and link previews read first. A page without a title is a book without a cover.

H1 heading — exactly one, stating the page’s topic. Multiple H1s (or none) leave automated readers unsure what the page is actually about.

Q&A structure — pages that answer questions directly, especially with FAQPage structured data, are among the pages AI answers quote most often. If your page answers questions, say so in a way machines can see.

Word count — reported for context, not scored. AI systems prefer pages with substance, but length alone says nothing about quality, so it doesn’t move the score.

Who this is for

Freelancers with a portfolio, blog or service site. When a potential client asks an AI assistant “find me a freelance designer in Manchester” or “how does IR35 affect contractors”, the sites with clean signals are the ones that get read — and cited. Ten minutes with the fix list above is one of the cheaper ways to be among them.

This tool checks public technical signals only, as observed on the day you run it. Site owners change robots.txt and page structure; re-run the check after you make fixes. It is not legal or SEO advice, and it makes no claim about search rankings.