review-external-pr
Pass
Audited by Gen Agent Trust Hub on Jul 10, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- Command Execution: The skill automates several Git and GitHub CLI operations, such as checking out branches, merging PRs, and creating staging branches. These actions are standard for repository management and are structured around user-confirmed paths.
- Data Sanitization: For the creation of staging branches, the skill implements a robust sanitization routine for PR titles (e.g., converting to lowercase, replacing non-alphanumeric characters with hyphens, and truncation). This is a security-positive pattern that helps ensure that dynamic content from PRs does not cause unexpected behavior in shell environments.
- Indirect Input Handling: The skill processes external metadata such as PR titles and commit messages. While these are sources of untrusted data, the skill uses these inputs primarily for display or within the aforementioned sanitization logic, which minimizes the risk of indirect prompt injection affecting the agent's behavior.
Audit Metadata