Five checks, run live against your domain. Each one is a plain HTTP fetch of your own site — the same kind AI crawlers make — with the response shown next to the verdict. Nothing is queried but your site.
No email. No score. Every verdict reproduces with curl.
robots.txt decides which AI crawlers may read your site. We parse yours and show the allow or deny verdict for GPTBot, ClaudeBot, PerplexityBot, Google-Extended and CCBot, with the lines we read.
Most AI fetchers don't execute JavaScript. If your homepage is a shell that fills in client-side, a crawler gets a 200 status and no content — the page "works" and says nothing. We fetch without executing anything and show the text that came back.
Organization structured data tells a machine who you are: a name, and sameAs links to profiles that corroborate it. We list what your homepage declares.
We check whether /llms.txt and /llms-full.txt exist and serve real content. We treat this as hygiene — it's not a confirmed ranking input, and some platforms already serve it automatically.
We fetch /sitemap.xml and check it's a valid urlset or sitemap index.
Every verdict here reproduces from your own terminal. If our row and your terminal disagree, trust the terminal — then tell us.