attach-review-to-pr
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and process data from external GitHub pull requests, including file paths, code diffs, and potentially existing comments. This external content could contain malicious instructions aimed at influencing the agent's review or comment generation.
- Ingestion points: Data fetched via
gh apicalls inSKILL.md(e.g.,gh pr view,gh api repos/{owner}/{repo}/pulls/{pr_number}/files). - Boundary markers: None specified in the instructions to separate PR content from agent instructions.
- Capability inventory: The skill utilizes
gh apito write data back to GitHub (creating comments and reviews). - Sanitization: No specific sanitization or validation steps are outlined for the content being processed or generated.
- [COMMAND_EXECUTION]: The skill relies on executing shell commands via the GitHub CLI (
gh). While these are standard operations for repository management, the instructions involve interpolating PR metadata (like commit SHAs and file paths) directly into shell command arguments.
Audit Metadata