reversing-unity-il2cpp

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns, deceptive metadata, or security risks were detected. The skill provides legitimate procedural guidance for mobile application security assessments.
  • [COMMAND_EXECUTION]: The skill instructs the use of local command-line tools (unzip, rg, strings) and specialized binaries (Il2CppDumper, ilspycmd, frida) to process application files. This behavior is expected and appropriate for the skill's stated purpose of reverse engineering.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it processes untrusted binary data from external files. 1. Ingestion points: target.apk, libil2cpp.so, and global-metadata.dat mentioned in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: Local file extraction, regex searching, binary metadata dumping, and dynamic process instrumentation. 4. Sanitization: Absent. This risk is inherent to the skill's primary function and is considered safe in this context as it processes binary artifacts rather than natural language instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 01:16 AM
Security Audit — agent-trust-hub — reversing-unity-il2cpp