using-git-worktrees
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTION
Full Analysis
- COMMAND_EXECUTION (LOW): The skill automatically executes build and test commands such as
npm install,cargo build, andnpm testto prepare the isolated workspace. While this is the primary purpose of the skill, it presents a risk if the repository contains malicious configuration files or test suites. - REMOTE_CODE_EXECUTION (LOW): The use of package managers (npm, pip, cargo, go) triggers the download and execution of external dependencies. This behavior is expected for environment setup but carries inherent risks associated with untrusted third-party code.
- PROMPT_INJECTION (LOW): The skill reads from
CLAUDE.mdto identify directory preferences. This constitutes an indirect prompt injection surface where a malicious file could influence the agent's behavior. Evidence Chain: 1. Ingestion points:CLAUDE.md(via grep). 2. Boundary markers: Absent. 3. Capability inventory: Shell command execution (git, npm, pip, cargo, go, poetry). 4. Sanitization: None.
Audit Metadata