Control Zero Blog
Thinking in public about AI governance.
Practical perspectives on governing AI agents, policy enforcement, security, audit trails, and what it means to deploy AI responsibly.
Coding Agent Security: How to Roll Out Claude Code and Cursor Without Unlimited Authority
A practical framework for rolling out AI coding agents safely across engineering teams without giving them unlimited shell, file, secret, or production access.
AI Agent Runtime Security: Securing the Moment Intent Becomes Action
The security problem is no longer limited to what the model says. Runtime security decides what an AI agent is actually allowed to do before intent becomes action.
claude --dangerously-skip-permissions: What It Does, and How to Get the Speed Without the Risk
What claude --dangerously-skip-permissions actually disables, real incidents it caused, and how to keep the speed without the risk using policies.
When Your Coding Agent Runs rm -rf: Real Incidents, and Why Prompts Are Not a Control
Five documented cases of AI coding agents deleting files and databases, why prompts and instructions failed to stop them, and what actually works.
Why AI Governance Matters More Than You Think
As AI agents gain autonomy in production systems, the gap between capability and control becomes a critical business risk. Here's why governance is the missing layer.
Shadow AI: The Hidden Risk in Your Organization
Employees are using AI tools your IT team doesn't know about. Shadow AI is the new shadow IT, and the risks are bigger because the surface area is orders of magnitude larger.