do-git-pr-review-triage
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes text from external contributors (reviewers) via pull request comments.
- Ingestion points: Fetches content using
gh pr viewand the GitHub API. - Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions when processing the fetched comment bodies.
- Capability inventory: The skill performs classification and drafting of responses based on the external input. While it is instructed not to apply changes or post responses automatically, instructions hidden within PR comments could attempt to influence the agent's logic, labels, or draft contents.
- Sanitization: No explicit sanitization or filtering of the external text is defined before it is analyzed by the agent.
Audit Metadata