Grith AI

🔥Hot take 🔥 If your AI agent can run shell commands, read files, and hit the network… …and you’re relying on prompts or “alignment” for security… you don’t have a security model. You have a demo. 🧵👇 #AI #infosec

Vibe Coding Still Needs a Senior Engineer (For Now) grith.ai/blog/security-…

Five AI Agent Failures in 36 Days. Zero Times the Agent Caught It. grith.ai/blog/36-days-5…

The Vercel Breach Needed Malware. The Next One Needs a Bad README. grith.ai/blog/next-verc…

Every Claude 4.7 Improvement Makes the Security Problem Worse grith.ai/blog/every-cla…

They Hacked Claude, Gemini, and Copilot (And No One Told You) grith.ai/blog/we-hacked…

@amorriscode When's it coming to Linux?

Today we're launching a rebuilt version of Claude Code on desktop. The app has been redesigned for the ground up to make it easier than ever to parallelize work with Claude. I haven't opened an IDE or terminal in weeks. Excited for you all to give it a shot!

Prompt Injection Is Unfixable (So We Stopped Trying) grith.ai/blog/prompt-in…

We went deep on this: * why AI agents are fundamentally unsafe today * what actually needs to be controlled * what a real security model looks like →grith.ai Curious who disagrees.

Until you control execution: * prompt safety is cosmetic * alignment is irrelevant * approvals are theatre You’re not securing the system. You’re just hoping the model behaves.

🔥Hot take 🔥 If your AI agent can run shell commands, read files, and hit the network… …and you’re relying on prompts or “alignment” for security… you don’t have a security model. You have a demo. 🧵👇 #AI #infosec

If Your AI Agent Ran npm install During the Axios Attack, You're Compromised grith.ai/blog/axios-sup…

Zero Ambient Authority: The Principle That Should Govern Every AI Agent grith.ai/blog/zero-ambi…