patch-diff-analyzer
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection (LOW): The skill is designed to ingest and analyze untrusted code from decompiled binaries, creating a surface for indirect prompt injection where malicious instructions could be hidden in code comments or strings. Ingestion points: Binary files processed via scripts/decompile-jar.sh and scripts/decompile-dll.sh, and the resulting patch-analysis.diff analyzed by the agent as per SKILL.md. Boundary markers: Absent. The instructions in SKILL.md direct the agent to read the raw diff output without using delimiters or warnings to ignore embedded instructions. Capability inventory: The skill possesses file-system write access and the ability to execute local commands (git, jadx, ilspycmd) via its shell scripts. Sanitization: Absent. There is no pre-processing or filtering of the decompiled source code before it is presented to the agent for reasoning.
Audit Metadata