git-worktree

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands to automate git operations, including git worktree add and directory management via symlinks.
  • [EXTERNAL_DOWNLOADS]: Project dependencies are installed using standard package managers such as pnpm, npm, yarn, pip, poetry, cargo, and go, which fetch code from official registries.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill identifies and copies sensitive environment files (e.g., .env, .env.local) to ensure local worktree configuration consistency; no evidence of remote exfiltration was detected.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied branch names as arguments for shell commands.
  • Ingestion points: The $ARGUMENTS variable in SKILL.md is used as the branch name.
  • Boundary markers: No explicit boundary markers are used in the instruction text.
  • Capability inventory: Includes shell access for git, filesystem manipulation, and package manager execution.
  • Sanitization: Employs input validation using a regular expression (^[a-zA-Z0-9/_-]+$) to verify branch names, preventing common command injection patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 07:38 AM
Security Audit — agent-trust-hub — git-worktree