create-skill

Pass

Audited by Gen Agent Trust Hub on Aug 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides instructional content for developers on how to structure agent skills, manage metadata, and handle progressive disclosure of information.
  • [SAFE]: The documentation mentions the dynamic context injection syntax (!command) using benign examples such as git status and git diff. As this is purely for documentation and teaching purposes within a developer-focused utility, it does not constitute a security risk.
  • [SAFE]: No evidence of prompt injection, data exfiltration, obfuscation, or malicious persistence mechanisms was found. The skill operates within expected development workflows.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 23, 2026, 06:03 PM
Security Audit — agent-trust-hub — create-skill