reverse-eng-macos
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (Mach-O binaries and .app bundles) to extract metadata, entitlements, and symbols.
- Ingestion points: The skill operates on binary files and application bundles provided as analysis targets, specifically checking code signatures and entitlements.
- Boundary markers: No explicit delimiters or "ignore embedded instructions" warnings are provided to separate the external target data from the agent's processing logic.
- Capability inventory: Includes system analysis tools and debuggers such as codesign, otool, spctl, lldb, and frida across the analysis workflow.
- Sanitization: No sanitization or validation of the target binary's metadata or symbol names is specified before processing.
Audit Metadata