Signet lets a business set the level of human assurance an AI agent needs before it can take an action, and apply its limits to the person behind the agent, not to each account.

You set a policy for each action. Browsing can stay open, joining a waitlist can require basic assurance, and buying can require a verified person. An agent presents the assurance its user has established with Signet. Signet's SDK on your server checks it against your policy and answers ALLOW, VERIFY FIRST or DENY. Each person reaches you under a private identifier of their own, so your limits count people, not accounts, and you never learn who they are.

Status

Signet is in private beta (November 2026). Beta partners get the parts marked Coming Soon first. To take part, book a call.