swiftui-view-refactor
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [Prompt Injection] (LOW): Indirect prompt injection surface detected. The skill is designed to ingest and refactor user-provided SwiftUI source code. \n
- Ingestion points: User-provided SwiftUI files (.swift) processed during refactoring. \n
- Boundary markers: Absent; the instructions do not explicitly tell the agent to ignore instructions embedded in the provided code. \n
- Capability inventory: No capabilities for arbitrary command execution, network access, or unauthorized file system writes were identified. \n
- Sanitization: No validation or sanitization of the input code is performed.
Audit Metadata