Most AI crawlers, like GPTBot, ClaudeBot and PerplexityBot, don’t run JavaScript. See how much of any page they can actually read, and what they miss.
Checking the page…
This usually takes 10 to 40 seconds.
What this tool checks
Read it like a crawler
We request the page the way an AI crawler does: one plain request, no JavaScript. That’s the HTML the crawler gets to work with.
Read it like a person
We load the same page in a real Chrome browser, let the JavaScript run, and scroll to the bottom so lazy-loaded content appears.
Compare the two
Anything a visitor sees that isn’t in the crawler’s copy is invisible to most AI assistants. We score the gap and show you exactly what’s missing.
What the report means
- Visibility score
- 0 to 100. Weighted 55% on text, 25% on key elements, 15% on internal links and 5% on images. 90 or more means AI crawlers see essentially the whole page.
- Text visible to crawlers
- How much of the text a visitor sees is already in the HTML the server sends. Text added later by JavaScript doesn’t count.
- Key elements
- The title, meta description, canonical URL, H1, structured data (JSON-LD), language and Open Graph title. These tell AI systems what the page is and which URL to cite.
- JS-only links
- Internal links that only appear once JavaScript runs. Crawlers can’t follow them, so the pages they lead to are harder to discover.
- Script-only content
- Text that’s in the HTML but only inside a script or data block (common with some JavaScript frameworks). Most crawlers extract readable page text, so treat it as missing.
- robots.txt access
- Whether the site’s robots.txt allows or blocks each major AI crawler from this URL, and the rule that decides it.