pm-verification-protocols

Pass

Audited by Gen Agent Trust Hub on Mar 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a workflow guide that establishes a 'QA Gate' to ensure implementation quality. It does not contain executable code or dangerous instructions.
  • [PROMPT_INJECTION]: No attempts to bypass agent safety guidelines or override foundational instructions were detected. The 'Circuit Breaker' logic serves as a behavioral guardrail for the agent's operational persona.
  • [COMMAND_EXECUTION]: The instructions suggest using diagnostic tools like curl, lsof, and pm2, as well as Chrome DevTools MCP functions, exclusively for the purpose of verifying application state and gathering evidence.
  • [DATA_EXFILTRATION]: No patterns for unauthorized data access or external exfiltration were found. Network operations mentioned are limited to local server health checks and user-provided testing URLs.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 30, 2026, 02:30 PM
Security Audit — agent-trust-hub — pm-verification-protocols