Referenced \u2013 AI Crawler Tracking counts visits from known AI crawlers, such as GPTBot, OAI-SearchBot, ClaudeBot, PerplexityBot and Googlebot, and posts hourly day totals to your Referenced AI account. You can then see which AI engines are reading which pages of your site.<\/p>\n\n
The plugin is a reporter only. It writes nothing to your pages or content, and it records no visitor data: no IP addresses, cookies or user accounts. For each known crawler it keeps a count per bot, path and day in one small table of its own.<\/p>\n\n
You need a Referenced AI<\/a> account and a connection token. Until you save a token, nothing is sent anywhere.<\/p>\n\n This plugin sends crawler visit totals to Referenced AI<\/a>, the service whose account you connect it to.<\/p>\n\n Privacy policy: getreferenced.ai\/privacy<\/a>\nTerms of service: getreferenced.ai\/terms<\/a><\/p>\n\n Pages served from a full-page cache or a CDN never reach PHP, so the plugin cannot count them. Static files such as images, scripts and PDFs are served without PHP too, so they are not counted either. This is a known blind spot: the numbers are a floor, not an exact total.<\/p><\/dd>\n Only the following, and only after you have saved a token: the bot name, the page path, the date, the visit count, the plugin version, the WordPress version and your site's home URL. No visitor IP addresses, cookies, user agents or personal data are sent.<\/p><\/dd>\n No. Each request costs a handful of cheap string comparisons (about twenty substring checks) and one small row write only when a known crawler matched.<\/p><\/dd>\n With plain permalinks ( Yes. Use Disconnect on Settings \u2192 Referenced to remove the token and stop all reporting, or deactivate the plugin. Deleting the plugin removes its table and options.<\/p><\/dd>\n\n<\/dl>\n\n\nExternal Services<\/h3>\n\n
\n
Credits<\/h3>\n\n
\n
\n
\n
Why do some crawler visits not appear?<\/h3><\/dt>\n
What leaves my site?<\/h3><\/dt>\n
Does it slow my site down?<\/h3><\/dt>\n
Why does every page show as \"\/\" with plain permalinks?<\/h3><\/dt>\n
\/?p=123<\/code>) the page is identified by its query string, and query strings are not sent. Every page is then reported as \/<\/code>. Pretty permalinks (Settings \u2192 Permalinks) give each page its own path.<\/p><\/dd>\nCan I stop it?<\/h3><\/dt>\n
1.0.0<\/h4>\n\n
\n