Why can’t I trust the user agent?
A user agent is a free-text header the client writes itself. Anyone can send “Googlebot” or “GPTBot” with one command, so scrapers routinely borrow the names of crawlers that sites allow. Check the verdict for the address, not the user agent, to tell a real crawler from an imitation.
What does “Spoofed claim” mean?
The request names a bot, in its user agent or in the list above, but the address does not belong to that bot. Treat it as an impostor, whatever its user agent says.
Does “Not a known bot” mean the visitor is human?
No. It means the address does not belong to any AI crawler, crawler or internet scanner we cover. Plenty of automated traffic comes from ordinary cloud and residential networks.
Which bots does it cover?
AI crawlers and agents such as GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot and PerplexityBot; search and SEO crawlers such as Googlebot, Bingbot, Applebot, YandexBot and Baiduspider; and internet scanners such as Censys and Shadowserver. Every bot is listed on the Bot signals page.
Is this the same answer the API gives?
Yes. The verdict matches risk.is_ai, risk.is_crawler and risk.is_scanner in the lookup API.
Why do I sometimes get “Unknown”?
Sometimes we cannot check an address right now. Try again in a moment.