pm-verification

Pass

Audited by Gen Agent Trust Hub on Jun 24, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is purely instructional and defines quality control checklists for internal project management workflows.
  • [COMMAND_EXECUTION]: The skill uses local file-system tools (Read, Grep, Glob) to analyze project documents. This access is restricted and aligned with the skill's purpose of verifying content completeness.
  • [PROMPT_INJECTION]: No evidence of prompt injection, role-play bypasses, or instructions to ignore safety guidelines was found.
  • [DATA_EXFILTRATION]: No network access is requested, and there is no access to sensitive system files or credentials.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 24, 2026, 08:31 PM
Security Audit — agent-trust-hub — pm-verification