code-reviewer
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external data in the form of code diffs and files. When operating in 'Implementation' mode, the agent is instructed to modify files based on its analysis. This capability creates a surface for indirect prompt injection where malicious instructions embedded in the analyzed code could attempt to influence the agent's modifications or behavior.
- Ingestion points: Code diffs and source files provided to the agent for review (SKILL.md).
- Boundary markers: The skill lacks explicit delimiters or warnings to ignore instructions embedded within the code being reviewed.
- Capability inventory: The skill description includes an 'Implementation' mode where the agent is explicitly tasked to 'make the changes' to files (SKILL.md).
- Sanitization: No specific sanitization or validation of the input code is defined.
Audit Metadata