ctf-reverse
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to analyze external, untrusted data such as binary files, APKs, and bytecode (e.g., in SKILL.md, references/tools.md, and references/field-notes.md). This exposes a surface for indirect prompt injection where content within the analyzed binaries could attempt to influence the agent's instructions.\n
- Ingestion points: Binary files and bytecode processed using GDB, Ghidra, and angr.\n
- Boundary markers: None. The skill does not explicitly warn the agent about ignoring instructions embedded within the data being analyzed.\n
- Capability inventory: The skill facilitates the use of powerful analysis tools (gdb, frida, angr, python3) which are used to inspect and execute target code.\n
- Sanitization: Not applicable, as the goal is a complete analysis of the provided data.
Audit Metadata