plugins-copilot/git-worktree/skills/worktree-create/SKILL.md
Create a new linked git worktree on its own branch under `.github/worktrees/<branch-slug>` for parallel agentic work. Trigger when the user says "create a worktree", "new worktree for X", "make a worktree on branch Y", "spin up a parallel checkout", or "/worktree-create". Supports an optional base branch via `--from <branch>`.
npx skillsauth add st0nefish/claude-toolkit worktree-createInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Create a new linked git worktree. Keeping worktrees inside the repo
(.github/worktrees/, auto-gitignored on first use) keeps operations
against them inside the session's CWD scope.
Use direct git worktree commands rather than plugin helper-script paths;
the plugin-root environment variable is not guaranteed to exist inside Bash
tool invocations.
Resolve the repo root with git rev-parse --show-toplevel, then set the
base directory to ${WORKTREE_BASE_DIR:-<repo-root>/.github/worktrees}.
On first use, create the directory and ensure it is gitignored (skip the
gitignore step if WORKTREE_BASE_DIR is overridden).
Convert the requested branch name into a directory slug by:
/ with -[A-Za-z0-9._-] characters with ---Set the worktree path to <base-dir>/<slug> and stop with a clear error
if that directory already exists.
If the branch already exists locally, run git worktree add <path> <branch>.
Otherwise create it with git worktree add <path> -b <branch> [<base-branch>],
defaulting the base branch to the current branch when the user did not specify one.
Report the worktree path and branch back to the user.
Remind them to run /worktree-remove <name> when finished.
.github/worktrees/) is created automatically and
added to .gitignore on first use.WORKTREE_BASE_DIR (skips the
auto-gitignore step).development
Start work from your description — explore the codebase and plan
data-ai
Multi-phase, multi-agent feature workflow: spec → plan → refine → divide → execute → review. Invoke when the user escalates a session-start/session-issue flow to orchestration, or asks to run a non-trivial feature (multiple files, design ambiguity, cross-cutting concerns, correctness-critical paths) through the full multi-agent workflow. For small fixes, prefer session-start.
tools
Browse open issues, pick one, and start work on it
tools
Review, clean up, and open a PR to finalize the work