ui-ux-polish
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is entirely instructional, containing documentation and code snippets (CSS, TSX) for design patterns like glassmorphism and bento grids. It does not perform network requests, file system modifications, or shell command execution.
- [SAFE]: The iterative polish prompts provided in the references are designed to guide the model toward higher-quality visual outputs and do not contain any instructions to bypass safety filters or ignore previous instructions.
- [SAFE]: All external tools mentioned (such as axe-core, NVDA, and VoiceOver) are standard, legitimate accessibility testing utilities.
- [SAFE]: The skill adheres to security best practices, such as recommending native HTML elements over ARIA attributes where possible and respecting user system preferences for reduced motion.
Audit Metadata