code-review
Pass
Audited by Gen Agent Trust Hub on May 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of instructional text and code snippets demonstrating best practices for code reviews. It does not include executable code, external scripts, or requests for sensitive information. All external references target Sentry's official developer documentation, which is a well-known service.
- [PROMPT_INJECTION]: As the skill is intended to review untrusted code changes from pull requests, it possesses an inherent surface for indirect prompt injection if the code being reviewed contains instructions intended to influence the agent's behavior.
- Ingestion points: Pull request diffs and code files provided to the agent for review (SKILL.md).
- Boundary markers: None identified within the skill instructions.
- Capability inventory: The skill uses default agent tools for reading files and providing feedback; no dangerous internal scripts are provided.
- Sanitization: The skill does not define specific sanitization or validation logic for the code it processes.
Audit Metadata