A1-User
Run by A1
How to detect A1-User
A1-User signs its requests, so you detect it by verifying the cryptographic signature on each request. A valid signature proves the request came from A1; a copied user agent without the signature does not.
- Detection method
- SignedThe agent signs its requests, so its identity is verified cryptographically.
- Signals that identify it
- CryptographicA signed request verified cryptographically
- BehavioralAutomated clicks, typing and page flow instead of a person's
Every signal Verisoul checks for A1-User
A1-User signs its requests, so we verify who it is cryptographically, then check the rest of the session.
- Used
Cryptographic
A signed request verified cryptographically
- Not used
Network
Not needed to identify it
- Not used
Device
Not needed to identify it
- Used
Behavioral
Automated clicks, typing and page flow instead of a person's
How often you'll see it
Very uncommon
Rare, brand new, or not seen on Verisoul traffic yet
Questions about A1-User
How do you detect A1-User?
A1-User signs its requests, so you detect it by verifying the cryptographic signature on each request. A valid signature proves the request came from A1; a copied user agent without the signature does not.
What is A1-User?
A1-User is an AI agent from A1 that works in a browser and signs its requests.
Does A1-User identify itself?
Yes. A1-User signs its requests cryptographically, which is the strongest way an agent can identify itself.
Can a bot pretend to be A1-User?
Not convincingly. A bot can copy A1-User's user agent, but without A1 signing keys its requests fail verification and it is not labeled A1-User.
How common is A1-User?
Very uncommon. Rare, brand new, or not seen on Verisoul traffic yet.
What does the Verisoul API return for A1-User?
Every session Verisoul identifies as A1-User carries an agent block with id "a1_user", name, provider, type "ai_agent", detection_method "signed" and the detection_signals used (cryptographic, behavioral). Build rules on the id, type and detection method.