CiteTrack keeps a local crawler catalog and updates it automatically. The goal is not just to match famous names like GPTBot, but to separate search crawlers, AI search crawlers, assistant-triggered fetchers, archive crawlers and possible impersonators.
Examples included in the catalog
| Company | Examples | Why it matters |
|---|---|---|
| OpenAI | GPTBot, OAI-SearchBot, ChatGPT-User | Training, live search and user-triggered ChatGPT fetches are different signals. |
| Anthropic | ClaudeBot, Claude-SearchBot, Claude-User | Separates Claude crawling from Claude user-triggered access. |
| Perplexity | PerplexityBot, Perplexity-User | Helps connect crawler access with answer-side citation checks. |
| Google / Microsoft / Apple | Googlebot, Google-CloudVertexBot, Bingbot, Applebot | Useful for AI search, search crawling and platform source discovery. |
| Cloudflare AI Crawl Control set | Anchor Browser, Cloudflare Crawler, Manus Bot, MistralAI-User, Novellum AI Crawl, ProRataInc, Terracotta Bot, TikTok Spider, archive.org_bot, Arquivo Web Crawler | Keeps the plugin aligned with newer crawler names users see in Cloudflare and server logs. |
Why names alone are not enough
User-agent names can be spoofed. CiteTrack therefore labels requests as verified network, user-agent only or possible spoof when verification evidence is available. Use verified requests for reporting, and use possible-spoof rows as diagnostic or security evidence.
Crawler catalogs change. The plugin updates the list without storing raw visitor IP addresses as ordinary analytics data.