implement
Pass
Audited by Gen Agent Trust Hub on Mar 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a robust security model for automation. It includes a specific step to scan task documents for dangerous operations (e.g., sudo, rm, chmod) and blocks them or requires explicit user approval.
- [SAFE]: It uses worktree isolation via the EnterWorktree tool, ensuring that changes happen in a dedicated directory/branch without affecting the user's main environment directly.
- [SAFE]: Sub-agents are spawned with a restricted set of allowed tools, limiting their capabilities to safe file operations and standard development tools such as npm and git.
- [SAFE]: References to external resources point to the author's own repository or trusted organizations like Vercel Labs and Supabase.
Audit Metadata