ui-skills
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides UI design guidelines and does not contain any malicious code or instructions.
- [PROMPT_INJECTION]: No attempts to override safety filters or agent behavior were detected. The skill includes a feature to review files, which presents an indirect prompt injection surface.
- Ingestion points: User-provided files via
/ui-skills <file>. - Boundary markers: None present.
- Capability inventory: Limited to generating text-based review reports.
- Sanitization: None explicitly defined.
- [DATA_EXFILTRATION]: No network operations or access to sensitive credentials or files were found.
- [REMOTE_CODE_EXECUTION]: No remote scripts are downloaded or executed. References to UI libraries are for implementation guidelines only.
Audit Metadata