using-git-worktrees

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses shell and Git commands to manage the development environment and perform setup tasks.
  • Detects repository status using git rev-parse and git branch.
  • Creates isolated worktrees using git worktree add.
  • Executes project-specific test runners such as npm test, cargo test, pytest, and go test.
  • [EXTERNAL_DOWNLOADS]: The skill invokes standard package managers to install dependencies defined within the project.
  • Detects project configuration files and executes npm install, pip install, cargo build, poetry install, and go mod download as needed.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from the local repository environment to determine its execution flow.
  • Ingestion points: Local project metadata files (e.g., package.json, requirements.txt) and Git branch information in SKILL.md.
  • Boundary markers: No specific delimiters are used to separate repository data from instructions.
  • Capability inventory: The skill has the ability to execute arbitrary commands via package scripts and test runners defined in the local repository.
  • Sanitization: No sanitization is applied to project-defined scripts before they are executed by the environment management tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 09:32 AM
Security Audit — agent-trust-hub — using-git-worktrees