code-review
Pass
Audited by Gen Agent Trust Hub on Mar 25, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes standard command-line utilities including
git,gh(GitHub CLI),glab(GitLab CLI), andjqto determine branch history and extract code changes. These tools are used as intended for repository management and code review workflows. - [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it ingests data from external sources such as pull request metadata and git diffs which could contain adversarial instructions.
- Ingestion points: Branch names, pull request titles/descriptions (via
ghandglab), and the contents of code diffs referenced in SKILL.md. - Boundary markers: The instructions recommend isolating code changes using
git diffranges, which serves as a logical boundary, though no explicit text delimiters are defined. - Capability inventory: The agent is authorized to execute git and platform-specific CLI commands to inspect the repository state.
- Sanitization: The skill does not explicitly specify sanitization or escaping of branch names or PR content before they are used in command-line operations.
Audit Metadata