Blog
Your AI Agent Is Indistinguishable From Malware
·MMatthew SlipperThe operations that make your coding agent useful are the same ones malware uses. We built a tool to tell the difference.
How To Dangerously Skip Permissions
·MMatthew SlipperFive realistic ways to sandbox a coding agent, what each one actually gets you, and the three things worth protecting against before you pick one.
AI Agents Are Insecure by Design. We're Fixing That
·MMatthew SlipperThe behaviors that make agents dangerous are the same ones that make them useful. The answer isn't to restrict them: it's to build infrastructure that watches what they do.