Grith AI retweeted
Claude Code’s new Auto Mode removes permission prompts.
But it also shifts security decisions to the model itself.
That’s the real change.
We broke down what this means in practice 👇
grith.ai/blog/claude-auto-mo…
@AnthropicAI @claudeai #AI #Security #DevTools
Grith AI retweeted
A GitHub Issue Title Compromised 4,000 Developer Machines
grith.ai/blog/clinejection-w…
🤖 Made with AI
Grith AI retweeted
AI Agent Backdoors Trivy Security Scanner, Weaponizes a VS Code Extension
grith.ai/blog/ai-agent-compr…
#pwned #aiagent #ai #vibecoding
Grith AI retweeted
We asked Claude Code to add input validation to a single route handler. The task required editing one file. Roughly 20 lines of code.
Here is what happened before it wrote a single character...
grith.ai/blog/syscall-trace-…
@claudeai @AnthropicAI @OpenAI
Grith AI retweeted
There's a malware that lives entirely inside of an AI agent's context window and executes at will. It's called Brainworm.
And then there seem to be people (or agents? :) who care about security. Not sure I have guts to test the Brainworm vs Grith.
Anyone?
We’ve got this covered - grith.ai
Grith AI retweeted
We Audited 2,857 Agent Skills. 12% Were Malicious.
grith.ai/blog/agent-skills-s…
Grith AI retweeted
We audited the security of 7 AI coding agents.
Only 1 has OS-level sandboxing, 0 have per-syscall evaluation. All vulnerable to prompt injection.
grith.ai/blog/security-audit…
@OpenAI @AnthropicAI @cursor_ai @ClineAI @aaborone @OpenInterpreter @HiddenLayerSec @MindgardAI
Grith AI retweeted
Excited to announce Claude for Open Source ❤️
We're giving 6 months of free Claude Max 20x to open source maintainers and core contributors.
If you maintain a popular project or contribute across open source, please apply!
claude.com/contact-sales/cla…
