skill-acquisition

Pass

Audited by Gen Agent Trust Hub on Apr 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is designed for self-improvement by drafting new instruction files. It includes robust safeguards such as the 'NEVER auto-save' rule, which ensures no executable content is added to the system without explicit user approval.
  • [EXTERNAL_DOWNLOADS]: The skill performs web research to gather information for new skills. This ingestive behavior is mitigated by restricting access to allowlisted domains and limiting the number of fetches to five per session, reducing the risk of encountering malicious external content.
  • [COMMAND_EXECUTION]: The skill writes file content to the local filesystem within the 'skills/' directory. While this is a sensitive operation, it is a core functional requirement of the skill and is strictly gated by a mandatory human review step before the file is written.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 18, 2026, 12:42 AM
Security Audit — agent-trust-hub — skill-acquisition