goal-skill

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists exclusively of instructional content and prompt templates, with no executable scripts, binaries, or automated subprocess calls.
  • [SAFE]: The framework promotes security best practices by requiring the definition of explicit boundaries, iteration policies, and blocked stop conditions for agent tasks, reducing the risk of autonomous drift or unsafe actions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 03:10 PM
Security Audit — agent-trust-hub — goal-skill