Product
AI agent detection: see which AI crawlers and agents visit your site
AI traffic is not one thing: some bots collect pages to train models, some fetch a page because a person asked an assistant a question, and some act for a user. FingerPrint Checker recognises the named ones from a public directory of 142 bots, tells you which operator each one belongs to, and lets you allow or block each by name or by category.
What we recognise
The directory groups AI traffic into four categories. The counts are the entries listed today.
- AI crawlers: bots that collect pages for model training or indexes (20 listed).
- AI search: bots that build the index behind an AI answer engine (10 listed).
- AI assistants: bots that fetch a page when a person asks an assistant about it (9 listed).
- AI agents: bots that browse and act on behalf of a user (2 listed).
Agents that sign their requests can be confirmed
Some agents sign their requests so that a website can tell them from imitations. When the operator is listed in the directory as Signed and the request carries a valid signature, the request is reported as a good bot with that agent’s name and operator.
Impersonators are flagged
Anyone can put a well-known bot’s name in a request. When a request claims a name we can confirm and fails the confirmation, it is reported as spoofed and counts as a bad bot. A named bot that we have no way to confirm is also reported as a bad bot, so a copied name never gets a free pass.
You decide per agent
Rules can allow or deny a bot by name, by category or by whether it was confirmed. You might let search engines and a few AI assistants through, and refuse the AI crawlers you do not want. The first rule that matches decides.
Check a request from your backend
If your pages are served without the browser script, your backend can send the details of a plain HTTP request to the bot-check call and get the same good, bad or not detected answer, plus what is known about the network. You can also see what your own request looks like on the free bot check.
What this does not do
An AI browser that looks like an ordinary Chrome browser is not recognised as an AI agent; it is judged like any other visitor. A bot that is not in the directory is not blocked just for being absent. We do not promise to catch every agent.
Frequently asked questions
Can I allow ChatGPT but block other AI crawlers?
Yes, if the bot you want is in the directory. Allow it by name and deny the AI crawler category. Put the allow rule first. The "block crawlers" request filter runs before rules and blocks every self-declared crawler, so leave it off if you want rules to allow ChatGPT.
Does this catch every AI agent?
No. It recognises the bots in the directory and flags requests that imitate them. Agents that are not listed and do not identify themselves are judged like any other visitor.
What happens to a bot I allow?
It is reported as a good bot in your events and webhooks, with its name, operator and category, so you can see what visited.
Can I try it without an account?
The free bot check shows how your own request is classified from the request alone.
Try it on your own sign-up flow
Every new account gets 2,000 free checks. No card required.