Every agent. Every department. Under control.

Confused Deputy At Machine Speed

A Hidden Project Link Can Make Claude Code Silently Send Your Files to an Attacker

Tego AI Finds Anthropic’s Claude Tag Slack Integration Can Trigger Unauthorized Enterprise Actions

Inside Anthropic's MCP tunnels

Same Words, Different Doors: Notes on Fable 5’s Safety Mechanism and Its Position-Sensitive Routing

The Hidden Layer: How Agent Skill Instructions Quietly Become Your Biggest Attack Surface

AI Security Has a New Perimeter. Most Stacks Haven't Moved Yet.

ChatGPT Enterprise Said “Read-Only”. But It Could Delete Your Files and Push to Your Repos.

Agentic AI CTF: Teaching AI Security by Breaking AI Systems