how
Pass
Audited by Gen Agent Trust Hub on Aug 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a controlled orchestration process for analyzing codebases. It utilizes a hierarchical structure where a lead agent delegates specific investigation slices to read-only subagents.
- [SAFE]: Subagent prompts (Explorer and Critic) are strictly focused on technical discovery and architectural evaluation, requiring concrete code evidence for all findings and explicitly forbidding file modifications.
- [SAFE]: Tool usage for 'parallel' exploration and 'runtime verification' is handled through a standardized capability contract ('pstack'), which ensures the agent operates within defined platform boundaries.
- [SAFE]: No evidence of prompt injection, data exfiltration, or obfuscation was found in the skill instructions or reference templates.
Audit Metadata