implement
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes Git (
git branch,git commit) and the GitHub CLI (gh issue view,gh pr create,gh pr comment) to manage code branches and pull requests. These actions are performed within the scope of the skill's intended development workflow. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external GitHub issues which could contain malicious instructions.
- Ingestion points: Reads issue content and comments using
gh issue view <n> --commentsinSKILL.md. - Boundary markers: The instructions do not define specific delimiters for the issue content.
- Capability inventory: The skill can modify the local filesystem, execute Git commands, and interact with GitHub repositories.
- Sanitization: The skill incorporates several verification layers, including automated mutation testing and a mandatory adversarial review by a separate subagent before the process is considered complete. This design limits the risk of automatic obedience to injected instructions.
Audit Metadata