git-worktrees
Pass
Audited by Gen Agent Trust Hub on Mar 15, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill provides instructions for a PR review workflow that involves fetching code from a remote pull request and immediately executing 'npm install' and 'npm test'. This is a surface for indirect prompt injection where an attacker could provide a malicious pull request containing scripts that execute unauthorized commands when the agent follows the workflow.
- Ingestion points: External code from git pull requests.
- Boundary markers: Absent.
- Capability inventory: The skill utilizes subprocess calls for 'git' and 'npm'.
- Sanitization: No verification of external code is performed before execution.
- [EXTERNAL_DOWNLOADS]: The skill facilitates downloading code from remote sources (Git pull requests) via the 'git fetch' command. While Git and npm are well-known services, the code being fetched is untrusted.
Audit Metadata