Free tool

Which AI crawlers can read your site?

Paste your robots.txt. This reads it the way Google's parser actually does, and tells you two different things for each of 18 crawlers: whether it is allowed, and whether you ever decided that — or inherited it from a wildcard, or from a rule you wrote about a different crawler entirely.

It runs entirely in your browser. Nothing is uploaded, stored, or sent anywhere — view source and check. It is verified cell-for-cell against Protego, the RFC 9309 parser behind our 25,000-domain study, so it cannot quietly disagree with the numbers we published.

We build the link; your browser does the fetching.

This is a snapshot. The problem is that it changes.

A robots.txt is one deploy away from a rule nobody meant to ship. In our scan of the top 25,000 domains, 93.8% of sites blocking GPTBot by name had never written any rule for OpenAI's search crawler — not a decision, a default nobody revisited. Driftwitness watches files like this one and tells you the day they change.

Get early access →