harness-thinking
Pass
Audited by Gen Agent Trust Hub on Jun 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists exclusively of markdown documentation outlining reasoning strategies such as Generator-Evaluator separation and context management. There are no executable scripts, shell commands, or external dependencies. All patterns described are common architectural practices for improving AI performance on complex tasks and do not introduce security vulnerabilities.
Audit Metadata