agent-orchestration-multi-agent-optimize
Pass
Audited by Gen Agent Trust Hub on Jul 31, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a guide for performance engineering and does not contain executable malicious code or dangerous instructions.
- [PROMPT_INJECTION]: No evidence of prompt injection or bypass attempts was found. The use of $ARGUMENTS for specifying an optimization target is standard practice and does not present a high-risk surface in this context.
- [EXTERNAL_DOWNLOADS]: The skill does not perform any external network requests or script downloads.
Audit Metadata