using-git-worktrees

Pass

Audited by Gen Agent Trust Hub on Feb 20, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • COMMAND_EXECUTION (LOW): The skill executes multiple shell commands and package management tools to set up worktrees and prepare environments.
  • Evidence: Commands such as git worktree add, npm install, cargo build, pip install, and various test runners (npm test, pytest) are explicitly defined in the workflow.
  • Context: While these are intended primary functions for a development-oriented skill, executing these commands on an untrusted repository could trigger malicious scripts (e.g., npm postinstall hooks).
  • INDIRECT_PROMPT_INJECTION (LOW): The skill reads from CLAUDE.md to determine configuration preferences, creating an ingestion surface for untrusted data.
  • Ingestion points: CLAUDE.md is parsed using grep to extract directory preferences.
  • Boundary markers: Absent; the skill does not use specific delimiters to separate user data from instructions.
  • Capability inventory: The skill possesses significant capabilities, including file system writes (.gitignore modification), repository modification (git commit), and arbitrary command execution via package managers.
  • Sanitization: None; the extracted preference is used directly in path construction.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 20, 2026, 06:05 AM