cross-harness-skill-authoring

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill package consists solely of Markdown reference files and documentation guidelines.
  • [NO_CODE]: There are no executable scripts, binaries, or active commands included in the skill.
  • [SAFE]: The instructions emphasize security principles, including explicit authority boundaries, evidence-backed progress, and the use of non-destructive validation methods.
  • [SAFE]: External links point to official documentation domains of well-known technology providers like OpenAI and Anthropic.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 08:45 AM
Security Audit — agent-trust-hub — cross-harness-skill-authoring