macos-reverse
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes standard macOS utilities including
otool,codesign,spctl, andfileto perform static inspection of binaries and their security metadata. These are established diagnostic tools used for their intended purpose. - [INDIRECT_PROMPT_INJECTION]: The skill provides instructions for analyzing untrusted binary files which represents an attack surface for indirect prompt injection. 1. Ingestion points: Mach-O binaries, dylibs, and app bundles processed in
SKILL.mdandreferences/macho-triage.md. 2. Boundary markers: No explicit delimiters or instructions are provided to the agent to distinguish between tool output and potential malicious payloads in binary metadata. 3. Capability inventory: The skill primarily performs read-only diagnostic operations using system tools. 4. Sanitization: No sanitization of the output generated by the disassembly or inspection tools is mentioned.
Audit Metadata