You probably found this address in your server logs. Here is exactly what our crawler is and how to stop it.
AscribeGEO-Audit/1.0 (+https://ascribe.dev/bot)AscribeGEO-Audit is the crawler behind Ascribe, a tool that checks whether AI engines can read and cite a website. It only runs when a signed-in user asks for an audit of a specific domain — it does not crawl the web on its own.
robots.txt first, then at most 20 HTML pages of the same site, plus a HEAD request per link to see if it is broken. It never submits forms, never logs in, and never requests the same page twice in one run.
Yes. We read robots.txt before anything else and obey it path by path, including Crawl-delay. If we cannot read it — a block, a server error, a timeout — we do not crawl at all.
Add this to your robots.txt. We check it on every run and will stop immediately.
User-agent: AscribeGEO-Audit Disallow: /
To slow it down instead of blocking it, use Crawl-delay — we honour it.