write-harness-skill

Pass

Audited by Gen Agent Trust Hub on Jun 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is entirely instructional, providing a framework for creating other skills without including executable code or automated actions.
  • [SAFE]: No instances of prompt injection, obfuscation, or unauthorized data access were found.
  • [SAFE]: The inclusion of a review checklist that enforces transparency and outcome verification demonstrates adherence to safety and reliability best practices.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 14, 2026, 10:13 PM
Security Audit — agent-trust-hub — write-harness-skill