Paste any page. We read its robots.txt, then request it 25 more times — once as each AI crawler — and show you which ones got in.
Most crawler checkers read one file and call it a report. That file describes a policy — it says nothing about what your CDN, WAF or bot-protection rule does when a crawler actually knocks.
We read the file and work out the verdict for each bot individually — including wildcard groups and the precedence rules that decide which line actually wins. This part is exact.
Then we request the page once per crawler, sending that crawler's real User-Agent, and record the status code. A CDN or firewall can refuse a bot your robots.txt welcomes, and robots.txt cannot tell you that.
Allowed in robots.txt, refused by the server, is the silent failure — and it is what a one-click "block AI scrapers" toggle produces. A robots-only checker reports that site as perfectly healthy.
Blocking a search or answer bot costs you visibility today. Blocking a training bot is a normal editorial choice — so the report reports it and never counts it as a problem.
A page every crawler can reach still loses if the answer names a competitor. orqly runs your real buying prompts across ChatGPT, Perplexity, Gemini, Copilot and Grok — then shows you which prompts, sources and rivals to go after.