AC2 Protocol: The missing security layer for AI agents

AC2 Protocol is a security framework that sits between AI agents and sensitive operations, requiring hardware-bound user signatures to verify intent before execution rather than giving agents direct access to credentials. It addresses gaps in current human-in-the-loop systems by providing cryptographic proof of authorization, phishing resistance, and credential isolation—keeping private keys on the user's device while agents request approval through a signing channel. The open protocol can be integrated as a plugin into existing agent frameworks and works without blockchain or central message relays.