qt-ui-design
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues detected. The skill instructions focus on UI design best practices, accessibility rules (WCAG 2.2), and platform-specific constraints for embedded systems. The embedded mode includes safety-positive requirements such as multi-cue alarms and confirmation before actuator commands.\n- [EXTERNAL_DOWNLOADS]: The skill references established industry resources including documentation from the W3C, Nielsen Norman Group, Apple, Microsoft, Google, and official Qt documentation. These are reputable sources and do not involve executable code downloads.\n- [INDIRECT_PROMPT_INJECTION]: The skill features an audit mode that processes external UI code or images.\n
- Ingestion points: Auditing existing UI code, screen captures, or images of UI designs (SKILL.md, Section 5).\n
- Boundary markers: Audit findings are gated behind a specific checklist, though no explicit delimiters for code input are mentioned.\n
- Capability inventory: The skill is limited to producing advisory text and audit findings; it possesses no capabilities for file-system writes, network operations, or shell command execution.\n
- Sanitization: No explicit sanitization or filtering of external design content is specified.
Audit Metadata