using-git-worktrees
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill automatically executes package managers including npm install, pip install, poetry install, and go mod download based on the presence of project manifest files. This behavior is standard for developer setup tools but involves fetching code from external package registries.
- [COMMAND_EXECUTION]: Shell scripts interpolate branch names and repository paths into commands like git worktree add. While standard for this workflow, the skill relies on the agent to handle user-provided branch names safely to avoid shell injection.
- [DYNAMIC_EXECUTION]: The skill executes test suites using commands such as npm test, cargo test, and pytest. Running tests involves executing code found within the repository to verify the environment baseline.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests configuration from CLAUDE.md and repository manifest files to determine the worktree directory and setup commands.
- Ingestion points: Reads from CLAUDE.md and files like package.json or Cargo.toml.
- Boundary markers: None present; the skill treats the content of these files as authoritative for configuration.
- Capability inventory: Includes file writing (.gitignore modification), repository commits, dependency installation, and test execution.
- Sanitization: No specific sanitization or validation of the branch names or config paths is documented in the instructions.
Audit Metadata