xmap-analysis
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted linker map files to extract symbols and layout information for binary analysis.\n
- Ingestion points: Linker map files provided by the user (e.g., arm9.o.xMAP) as indicated in the SKILL.md workflow.\n
- Boundary markers: Absent; the instructions do not specify delimiters or warnings for the agent to ignore potentially malicious instructions embedded within the map files.\n
- Capability inventory: The skill executes local Python and Ghidra scripts (parse_xmap.py, export_ghidra_program.py, correlate_xmap.py, analyze_function.py) and utilizes Ghidra's analyzeHeadless utility to perform file system operations and binary analysis.\n
- Sanitization: There are no mentioned sanitization steps for the symbol names or other metadata extracted from the XMAP files before they are passed to the analysis scripts.
Audit Metadata