device-interaction
Pass
Audited by Gen Agent Trust Hub on Jun 12, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill processes UI hierarchy data which may contain text from the application under test, creating an indirect prompt injection surface. (1) Ingestion points: UI hierarchy files and screenshots captured during device sessions. (2) Boundary markers: The skill does not instruct the agent to distinguish between its instructions and application UI text. (3) Capability inventory: The skill can build/run applications and perform system interactions. (4) Sanitization: No explicit validation of UI strings is performed.
- [COMMAND_EXECUTION]: The skill utilizes platform tools to build and execute iOS applications on simulators or physical devices for verification purposes.
- [DATA_EXPOSURE]: The skill captures screenshots and structural UI data from the device session to verify the application's state.
Audit Metadata