using-git-worktrees

Git worktrees create isolated working directories that share the same repository, letting you develop on different branches without switching. Each worktree has its own working tree and index, so you can compile and test multiple features in parallel without disrupting your main workspace.

Local worktrees (.worktrees or worktrees directories) live inside your repository and move with it; global worktrees sit outside. Local worktrees are safer for project-specific isolation and respect .gitignore rules, while global worktrees persist across repository clones.

Git worktrees share the repository's objects but have separate working trees. Checking .gitignore ensures project-local build artifacts, dependencies, and temporary files won't pollute your repository when the worktree is created and populated.

Yes. After creating a worktree, automated setup detects your project type—Node.js, Rust, Python, or Go—and runs the appropriate initialization commands so your feature branch is ready to work in immediately.

The worktree creation workflow prioritizes existing local directories (.worktrees first), honors CLAUDE.md preferences for configuration, and prompts you when choices are needed, ensuring the worktree aligns with your project's workflow conventions.

After creation and automated setup, baseline tests run to verify the worktree is functional and ready. You can then begin feature development in the isolated environment without affecting your main branch or other worktrees.