liquid-glass
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill acts as a comprehensive documentation and pattern library for UI/UX designers and SwiftUI developers. A thorough analysis of all instruction files confirms that no command execution, credential harvesting, persistence mechanisms, or obfuscation techniques are present.
- [EXTERNAL_DOWNLOADS]: The skill references several external resources, including official Apple Human Interface Guidelines and a community-curated design repository on GitHub. These are included solely for reference and do not involve runtime downloading or execution of untrusted scripts.
- [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: The skill interacts with user-supplied design requirements (e.g., requests to build specific SwiftUI components) within SKILL.md.
- Boundary markers: No explicit delimitation or instructions to ignore embedded commands are present in the guidelines.
- Capability inventory: The skill possesses no dangerous capabilities; it does not utilize file system writes, network tools, or shell command execution.
- Sanitization: No input filtering is implemented, but the lack of executable capabilities renders the injection surface benign.
Audit Metadata