CrawlCheck
Free tool · crawler verification

Is this IP really Googlebot?

Paste an address from your access log. You get a verdict against Google’s own published range files, which of Google’s crawler types it is, whether that type follows robots.txt, and the address’s reverse DNS name.

Nothing about the address is stored. The same check is free for machines at /api/v1/visitor.

How the check works

  • Range files first. The address is matched against the five files Google publishes. Four are Google’s own crawlers and fetchers; the fifth, user-triggered-fetchers.json, is Google Cloud customers’ traffic and never counts as Google.
  • Reverse DNS second. Googlebot’s addresses resolve to crawl-*.googlebot.com or geo-crawl-*.geo.googlebot.com. A name under googleusercontent.com forward-confirms for any rented Google Cloud machine, so it is never accepted. We measured 198 of 200 Cloud addresses passing the usual rule.
  • Three answers, never two. Verified, spoofed, or unverifiable when a file cannot be read. A name in a user agent is a claim anyone can type.
  • Type decides what robots.txt controls. Common crawlers read it. Special-case crawlers may or may not, per Google. User-triggered fetchers and agents act for a person and, Google says, generally ignore it.