git-worktree

Creating isolated git worktrees spawns dedicated working directories under ~/.claude-worktrees/<repo>/<branch>, allowing parallel branch development without affecting your main checkout. You run create-worktree.sh with a BRANCH_NAME environment variable to output the new worktree path.

Git worktrees provide isolated working directories for each branch, eliminating disruptive context switching. This skill creates worktrees under a dedicated structure and outputs BRANCH_NAME, WORKTREE_PATH, and DEFAULT_BRANCH as plain text for immediate use in your workflow.

To set a custom base branch when creating a git worktree, provide the BASE_BRANCH environment variable when running the creation script. If BASE_BRANCH is not provided, the skill automatically derives the default base branch from the repository's origin HEAD.

Yes, you can synchronize specific files by listing them in a .worktreeinclude file. When the worktree is created, the skill copies these listed files from the source repository into the new worktree directory to maintain your local configuration.

The skill includes management commands to safely remove a git worktree and automatically clean up any empty parent directories left behind. This keeps your ~/.claude-worktrees structure tidy after finishing or abandoning feature work.

You only need the standard git command-line tool installed, as the skill relies on built-in git worktree commands and simple shell scripts. No additional dependencies or complex environment setup beyond git are required to isolate your feature work.