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.shand.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