code-reviewer
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill is designed to ingest and analyze untrusted user-provided source code, which can contain instructions intended to influence the agent.
- Ingestion points: User code provided during the "code review" or "PR review" workflow as specified in the SKILL.md metadata.
- Boundary markers: Absent. The skill does not define specific delimiters or instructions to prevent the agent from executing commands embedded in the code being reviewed.
- Capability inventory: The skill is purely instructional and regex-based. It does not include active scripts for file system or network operations, but it operates in an environment where the agent may have such capabilities.
- Sanitization: None provided. The skill relies on the LLM to differentiate between the code structure and potential instruction-carrying data within that code.
Audit Metadata