harness-engineering-zh

Pass

Audited by Gen Agent Trust Hub on Apr 8, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is purely instructional and consists of Markdown-based best practices. It does not contain any executable scripts or hidden malicious logic.
  • [SAFE]: Code examples provided (such as init.sh and .eslintrc.js) are explicitly presented as templates for developers to copy and adapt. They follow standard development practices for dependency installation and architectural linting.
  • [SAFE]: The skill includes explicit security recommendations, such as 'Safety Autonomy Boundaries' that forbid agents from modifying CI/CD configurations, pushing directly to main branches, or using real credentials in environment files.
  • [SAFE]: The instructions promote the use of automated guardrails like type systems, Linters, and testing suites to restrict agent behavior and catch errors early.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 8, 2026, 02:09 PM