AI Crawler Log Analyzer

Paste a sample of your web server log and see which AI crawlers actually visited, how often, and what they fetched. The lines are parsed in your browser. Nothing is uploaded.

Nothing is sent to a server.

How to read the result

  • Counts describe the sample you pasted, not your whole log. Paste a full day or a full week for a comparable figure.
  • A retrieval agent fetching a page means an answer was being assembled at that moment. It does not mean your page was cited.
  • Training crawlers visiting says nothing about how your brand is described. That is what the Prompt Tester checks.

What this tool cannot do

  • It cannot verify that a user agent is genuine. Identification strings can be faked, so treat unknown hits as unverified.
  • It cannot read your analytics, your rankings or your citations.
  • It does not store or transmit the lines you paste.

Questions

Is my log uploaded anywhere?

No. The lines are parsed in your browser tab. Nothing is sent to our servers and nothing is stored.

Which log format should I paste?

The common or combined format written by Apache and nginx by default, with one request per line. Lines that cannot be parsed are counted separately and reported as unread.

Why do these numbers differ from my analytics tool?

Analytics tools usually filter automated traffic, and this analyzer filters nothing except lines it cannot read. That makes the totals differ by design.

Does a higher crawler count mean I am being cited?

No. Crawling is a precondition, not a citation. To check whether a brand is being mentioned, run the prompt tester and record what the assistants say.

Built and maintained by the same team that publishes this site. It counts what you paste. It does not promise rankings, rich results or citations.