review
Pass
Audited by Gen Agent Trust Hub on Aug 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements an isolation contract for independent reviews, ensuring that the reviewer sub-agent starts with a fresh context and is limited to specific inputs like diffs and acceptance criteria.
- [COMMAND_EXECUTION]: The skill interacts with version control systems (Git) to analyze code changes and suggests the use of CI tools to verify tests and linting status.
- [SAFE]: The skill maintains task context by reading from and writing to the ".oh/" directory, which is used by the author's tools to store session data and review verdicts.
- [SAFE]: The skill processes untrusted data (git diffs and issue descriptions) as part of its core review function; the risk is managed by using a structured review framework with specific alignment questions.
- Ingestion points: Git diffs, session files in ".oh/", and external issue criteria.
- Boundary markers: None explicitly defined for the diff content.
- Capability inventory: Local file system access and Git command execution.
- Sanitization: Standard processing within the review framework.
Audit Metadata