swiftui-liquid-glass
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEMETADATA_POISONINGINDIRECT_PROMPT_INJECTION
Full Analysis
- [METADATA_POISONING]: The skill references a fictional 'iOS 26+' and 'Liquid Glass API' throughout all files as if they were official Apple platform features. This metadata is deceptive regarding the actual existence of the API and the authenticity of the provided documentation.
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the processing of external user-supplied SwiftUI code for review and refactoring.
- Ingestion points: User-provided Swift source code and UI design requests as specified in the workflow sections of SKILL.md.
- Boundary markers: The instructions lack specific delimiters or instructions to prevent the agent from potentially following malicious directions embedded within the analyzed user code.
- Capability inventory: The skill is limited to text and code generation within the agent's context; it does not utilize shell commands, network access, or file-writing tools.
- Sanitization: There are no instructions for sanitizing, validating, or escaping external content before it is processed by the agent.
Audit Metadata