review-performance

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns detected. The skill's behavior is restricted to code analysis and reporting.
  • [PROMPT_INJECTION]: The instructions do not contain any bypass markers, role-play jailbreaks, or commands to ignore system safety guidelines.
  • [DATA_EXFILTRATION]: No network operations (curl, wget, fetch) or sensitive file path access were identified. The skill does not handle credentials or secrets.
  • [REMOTE_CODE_EXECUTION]: There are no scripts or configuration files that download or execute external code.
  • [NO_CODE]: The skill is entirely descriptive and instruction-based, containing no executable code or scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 09:05 AM