to-ios-test-script
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes application source code and local configuration files to generate test logic. This creates a surface where external data could potentially influence the AI's code generation.
- Ingestion points: Application source code and .meu/config.json.
- Boundary markers: No specific delimiters or instructions to ignore embedded commands in source code are present.
- Capability inventory: Writing executable Javascript files.
- Sanitization: The skill lacks explicit sanitization instructions for the ingested source code data.
- [DYNAMIC_EXECUTION]: The skill generates Javascript modules for the Frida runtime. While execution is not performed by the skill itself, the generation of code from external inputs is a dynamic execution pattern.
Audit Metadata