compliance-automation

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • Prompt Injection (SAFE): No instructions attempt to override safety filters or extract system prompts. The instructions explicitly reinforce grounding in provided documentation.
  • Data Exposure & Exfiltration (SAFE): No hardcoded credentials, sensitive file paths, or network operations are present.
  • Obfuscation (SAFE): No encoded strings, homoglyphs, or hidden characters were detected.
  • Unverifiable Dependencies & Remote Code Execution (SAFE): No external dependencies are declared, and no remote script execution patterns exist.
  • Privilege Escalation (SAFE): No commands involving sudo or permission modification are present.
  • Persistence Mechanisms (SAFE): No attempts to modify shell profiles or startup tasks.
  • Metadata Poisoning (SAFE): The metadata is consistent with the skill's purpose.
  • Indirect Prompt Injection (SAFE): While the skill is designed to help analyze user-provided policies, it lacks executable capabilities (like subprocess calls or file writing) that could be exploited via injection.
  • Time-Delayed / Conditional Attacks (SAFE): No logic gates or date-based triggers detected.
  • Dynamic Execution (SAFE): No use of eval, exec, or runtime compilation.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:15 PM