Author: Robin Martherus
Page 2/2
Stop Writing Policies for Agents. Write Policies for Actions.
The Model Can Build Exploit Chains. The Governance Can’t Stop It.
The Agent Did What It Said It Would Do. It Was Still Wrong.
The Three Domains of Agentic Security
I Built the Detector. Then I Built What Defeats It.
I Built an AI Attack Swarm. I’m Not the Only One.
Agent Identity Is a Name Tag. It Needs to Be Character.
The Agent Passed Every Security Check. It Still Shouldn’t Have Acted.
OpenClaw Has Eight Doors. The Industry Is Guarding One.
