code-reviewer
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external, untrusted data (source code, PRs, configurations), which creates a surface for indirect prompt injection attacks.\n
- Ingestion points: Instructions for reviewing microservice APIs, database migrations, React components, and Kubernetes configurations (SKILL.md).\n
- Boundary markers: The instructions lack explicit delimiters or guidance for the agent to ignore instructions embedded within the code being analyzed.\n
- Capability inventory: The skill references integration with CLI tools, IDEs, CI/CD pipelines, and communication tools which could be targeted by successful injections (SKILL.md).\n
- Sanitization: No procedures are defined for sanitizing or validating the code inputs before they are processed by the agent.
Audit Metadata