Check my logs
For a site operator: paste or drop your web server's access log once, or a set of them (several files, a folder, or .zip, .tar, .tar.gz and .gz archives). The page runs the same rules as the workbench's access-log preset over the requests to your host, and checks every request that says it comes from an AI provider's agent against the ranges the provider publishes, then sums both up in plain words. It runs in this page: your log is never uploaded, and nothing is fetched from anyone.
Spoofing cannot be checked with this site's files.
This site was built without the providers' published address ranges and signing keys, so a request that names an AI provider's agent stays a claim: nothing is verified and nothing is shown spoofed. The rules still run. The agent identity page takes copies of the providers' files you load.
1. Your log
Paste your log below, drop files or a folder here, or try the synthetic example.
Common or Combined Log Format (Combined carries the user agent the identity check reads). Behind a proxy, log the forwarded client address, not the proxy's. Rotated and compressed logs can go in together: each file is read on its own and all of them are checked as one, in this page.
An access log does not name the site it belongs to; the analysis needs it to read each request as a URL.
Example loaded: a synthetic log from documentation addresses on a reserved host, checked against synthetic range files that stand in for the providers' files. None of it was collected.
2. What the page found
3. What to do next
Flagged requests and spoofed agents are hints to check against the rest of your logs, not proof of who sent them (what to expect in your own logs).