orchestrator
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a procedural guide for agent coordination and does not contain any executable scripts, remote downloads, or hidden payloads.\n- [COMMAND_EXECUTION]: The skill requests standard developer tool access (Bash, Read, Write, Edit) for its orchestration and bug-fixing duties, which is consistent with its primary purpose and includes mandatory validation steps to maintain code quality.\n- [PROMPT_INJECTION]: The instructions focus on structured task decomposition and quality metrics, explicitly requiring verification and approval steps, with no evidence of bypass markers or safety-override instructions.
Audit Metadata