Features
Worktrees
A worktree session runs in an isolated copy of your repository, so parallel agents cannot overwrite each other's files. Worktrees require a git repository, live under ~/.grok/worktrees/<repo>/<name>, and start from your current HEAD, including uncommitted changes. Subagents can also request worktree isolation when the parent delegates parallel work.
Starting one
grok -w
grok --worktree=feat "refactor module X" # = keeps the prompt out of the name
grok -w --ref main "fix the flaky test" # clean checkout of the ref
grok -w -r <session-id> # resume in a fresh worktree
In the TUI: /fork --worktree forks the current session into a worktree, Ctrl+W on the welcome screen opens the New Worktree dialog, and Ctrl+W in the Agent Dashboard dispatches new agents into worktrees. Whether /new and /fork offer a worktree is configurable — see TOML Values.
A worktree is a real git checkout, detached at its base commit; land changes with ordinary git.
Housekeeping
Worktrees persist until you remove them: ending or deleting a session leaves its worktree in place, and gc runs only when you invoke it.
| Command | What it does |
|---|---|
grok worktree list |
List tracked worktrees |
grok worktree show <id> |
Show details for one worktree |
grok worktree rm <ids...> |
Remove worktrees (--dry-run to preview) |
grok worktree gc |
Remove entries whose directory is gone; --max-age 7d also expires idle worktrees not in use by a running process |