Bot-Eye View

Most AI crawlers — GPTBot and OAI-SearchBot (ChatGPT), PerplexityBot, ClaudeBot, and others — fetch your raw HTML and do not execute JavaScript. If your text, links or structured data are injected by JavaScript after the page loads (common with React, Vue, Angular and other single-page-app frameworks), the crawler never sees them. This tool fetches your page both ways — raw and fully JS-rendered — and shows you the gap.

It’s the share of your rendered text that’s already present in the raw HTML — i.e. how much of your content AI crawlers can actually read. 100% means your page is fully server-rendered and everything is visible. A low number means most of your content only appears after JavaScript runs, so AI answer engines see an almost-empty page and can’t cite you.

Serve the important content in the initial HTML response. Options include server-side rendering (SSR), static site generation (SSG), or pre-rendering / dynamic rendering for bots. Frameworks like Next.js, Nuxt and Astro do this by default. At minimum, make sure your title, H1, main body text, internal links and JSON-LD schema are in the raw HTML rather than injected by client-side JavaScript.

Not quite. Googlebot does render JavaScript (on a delay), so it can eventually see JS-injected content — but AI answer engines mostly do not. So a page can rank in classic Google yet still be invisible to ChatGPT and Perplexity. The “raw” column here is the safest assumption for AEO/GEO: if it’s in the raw HTML, every crawler can read it.