With tools like Moltbot(https://clawd.bot/) gaining traction, I got thinking about AI agents executing shell commands with full permissions by default.
I built veto to add a safety layer:
1.Intercepts commands before execution
2.Risk-based authentication (Touch ID for critical, simple confirm for low)
3.Custom rules for project-specific patterns
4.Full audit trail
Currently supports Claude Code (via hooks). Looking into Moltbot integration but it doesn't expose hooks yet — if anyone has ideas, I'm all ears.
Feedback welcome.