fix-pr-review
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses shell commands to interact with the environment and GitHub API. It executes
gh(GitHub CLI) for repository viewing, PR management, and GraphQL API interactions. It executesgitcommands for branch management, checking out PRs, and committing changes. It utilizesrg(ripgrep) for searching within the codebase.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from GitHub PR comments which could be used to influence agent behavior.\n - Ingestion points: Pulls
bodyanddiffHunkdata from PR review threads usinggh api graphqlinSKILL.md.\n - Boundary markers: The instructions lack specific markers to differentiate between data and potentially malicious instructions within PR comments.\n
- Capability inventory: The skill grants the agent the ability to modify source code, commit changes, and perform API mutations (replying to and resolving threads) in
SKILL.md.\n - Sanitization: No mechanisms are described to sanitize or filter instructions embedded within the fetched PR comment text.
Audit Metadata