task-validation-loop

The skill description is internally coherent with its stated purpose of autonomous task validation across PRD, architecture, and design system. It avoids risky behaviors like external downloads or credential exposure and uses controlled, document-based inputs to drive a multi-agent verification loop. The primary risks are operational (potential infinite loop or misconfigurations) and governance-related (need for explicit user approval and transparent iteration logs). Overall, the footprint is benign and proportionate to its stated purpose.