See which AI crawlers actually visit your site

No. The file is read by JavaScript in this browser tab and the results are calculated there. Nothing from your log is sent to us or anyone else, and nothing is stored. The only network requests are the license check (it sends the key only) and, for the Pro IP check, a download of the IP lists that OpenAI, Perplexity, Google, Bing and Apple publish; the matching happens in your browser.

nginx and Apache combined or common log format (including variants with a virtual host in front), JSON lines from Cloudflare Logpush, Caddy and most JSON loggers, W3C extended logs from IIS and CloudFront, and most other formats with a quoted request line such as "GET /page HTTP/1.1". Lines we can't read are counted and a few are shown so you can see why.

On a Linux server with nginx it's usually /var/log/nginx/access.log, with Apache /var/log/apache2/access.log or /var/log/httpd/access_log. cPanel hosts offer Raw Access logs under Metrics. Cloudflare can export HTTP request logs with Logpush (an Enterprise feature). Many managed hosts let you download logs from their dashboard.

One file or one paste, up to 10 MB of plain text: roughly 40,000 to 60,000 lines of a typical combined log, which covers a day of traffic on many small sites. A larger file is read up to 10 MB and the results say so. Pro reads up to 1 GB at a time, from several files, and unpacks .gz and .zip archives in the browser.

A user agent is only a claim: anyone can send "GPTBot" in a request. Pro checks each bot's IP addresses against the lists OpenAI, Perplexity, Google, Bing and Apple publish, so you can see which hits really came from them. Anthropic, Meta, ByteDance and others don't publish lists we can check automatically.

They aren't crawlers. Google-Extended and Applebot-Extended are robots.txt tokens that control how Google and Apple may use what Googlebot and Applebot already crawled, so they never appear in access logs.

No. The log shows which pages AI crawlers fetched and whether your server answered with 200, a redirect or an error. It can't show whether an AI answer cited or ranked you. It's most useful for finding blocks (403, 429, 5xx), pages crawlers never reach, and how much training-crawler traffic you serve.

A license key with no expiry: logs up to 1 GB at a time, several files at once, .gz and .zip, CSV and JSON downloads, a standalone HTML summary you can share, the IP check against published crawler lists, and your own bot patterns and excluded paths saved in your browser. It's a one-time payment by PayPal or USDT, no subscription. It's a simple log reader, not a hosted analytics service.