pr-reviewer
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external, potentially untrusted content from pull request descriptions, commit messages, and code diffs.
- Ingestion points: Uses
gh pr view,gh pr diff, andgit diffto pull external data into the agent's context (SKILL.md). - Boundary markers: The skill includes an explicit 'Untrusted input' section that defines a clear boundary, instructing the agent to treat PR content as data only and specifically naming patterns like 'ignore previous rules' as malicious directives to be ignored.
- Capability inventory: The skill utilizes GitHub CLI (
gh) and Git commands for data retrieval. It does not possess broad file-write or arbitrary network-send capabilities outside of standard version control operations. - Sanitization: Relies on specific behavioral instructions that require the agent to actively flag injection attempts discovered during the review process.
Audit Metadata