code-review
Pass
Audited by Gen Agent Trust Hub on Mar 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the GitHub CLI (
gh repo view) to retrieve the default branch name of the repository. This is a standard and expected operation for a tool designed to interact with Git repositories. - [PROMPT_INJECTION]: An indirect prompt injection surface is present as the skill processes external data in the form of pull request comments and review outputs.
- Ingestion points: The skill explicitly mentions incorporating 'PR comments passed in by the caller' into the evaluation phase in SKILL.md.
- Boundary markers: No specific delimiters or safety instructions are defined to separate user-provided comments from the system's logic.
- Capability inventory: The skill triggers the
/peer-reviewand/evaluate-findingssub-skills, which process the ingested content. - Sanitization: No explicit sanitization or filtering of external comments is mentioned in the instructions.
Audit Metadata