code-review
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill performs code reviews by ingesting external file data, creating a risk for indirect prompt injection where malicious instructions embedded in source code could be executed by the agent.
- Ingestion points: The skill instructs the agent to use 'git diff' and file reading tools to ingest project code into the context (SKILL.md).
- Boundary markers: There are no instructions or delimiters provided to help the agent distinguish between actual code and potential instructions hidden within that code.
- Capability inventory: The agent utilizes terminal commands for file listing and reading to facilitate the analysis.
- Sanitization: The skill does not perform any validation, filtering, or escaping of the code content before it is processed.
Audit Metadata