You've optimized your page — but what does an AI actually see when it fetches it? This tool answers that literally: it fetches your page the way an AI crawler would, extracts the readable text, and has a language model report back how it would describe your page, what it would cite, and what would make it hesitate.
Why it matters for AI search: every other GEO tool measures proxies — structure scores, schema validity, crawler access. This one measures something closer to the outcome: the actual impression your page makes on the kind of model that decides whether to cite you. The "most citable passage" output is especially actionable — if the model picks a throwaway sentence rather than your key claim, your best content isn't positioned where machines look for it. And the "citation risk" callout names the specific weakness (buried answer, unverifiable claims, stale dates) most likely to cost you citations.
How this tool works: enter a URL and the tool fetches it server-side (same proxy as our other tools, respecting the same safety limits), strips it to readable text — removing scripts, navigation, and footers — and passes it to an AI model with instructions to answer three questions honestly: how would an assistant describe this page to a user, which single sentence would it most likely quote, and what's the biggest reason it might skip this page for another source. The output is deliberately critical rather than flattering. Free usage is capped at 3 simulations per day.
Limitations: this is one model's reading, not a guarantee of how ChatGPT, Claude, Perplexity, or Google AI Overviews will actually treat your page — each has its own pipeline, and citation also depends on factors outside your page (competition, authority, query phrasing) that a single-page analysis can't see. The simulator reads server-rendered HTML only, and pages behind logins, paywalls, or aggressive bot protection can't be fetched. Results vary slightly between runs, as language model output does. For the deterministic technical checklist to pair with this qualitative read, run the AI-Readiness Audit.