using-git-worktrees
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- Command Execution (SAFE): The skill executes git, npm, cargo, pip, and go commands to create worktrees and set up environments. These actions are the core functionality of the skill and are performed on the local repository.
- External Downloads (SAFE): Dependency installation via standard package managers (npm, pip, cargo, go) is used to prepare the isolated workspace. This is expected behavior for development environment setup.
- Indirect Prompt Injection (SAFE): The skill reads from CLAUDE.md and branch content to automate setup. While this represents a theoretical attack surface, it is a necessary part of the development workflow and no malicious exploitation patterns were found. Evidence: 1. Ingestion points: CLAUDE.md and branch files. 2. Boundary markers: Absent. 3. Capability inventory: git worktree, npm/pip/cargo/go install and test tools. 4. Sanitization: Absent. The findings are marked SAFE because the associated risks are inseparable from the intended primary purpose of the skill.
Audit Metadata