What
- In the worktree-create flow, if no hook commands ran at all, the code used to always throw a "hook is configured but did not run" error. Now it first checks whether any
WorktreeCreatehook is actually configured, and silently continues if none is, only throwing when a hook is genuinely configured but failed to run. - The decision of whether to use a
WorktreeCreatehook (instead of falling back to the built-in git worktree flow) now goes through a renamed gate that checks either an existing predicate or the presence of a registeredWorktreeCreatehook. - The existing
WorktreeCreate/WorktreeRemovehook events are now also checked through a broader capability lookup (not just whether hooks are registered for them), used to decide whether the worktree create/remove UI is shown at all.
Why These changes reduce spurious "hook is configured but did not run" errors when no worktree hook is set up, and make the logic for whether worktree create/remove hooks and UI are active more consistent.
Names in the bundleWorktreeCreate
The entry above is what we published on the day. These lines were added later, as Anthropic's own pages caught up, and they sit beside the original rather than replacing it.
Added since
A small documentation edit on Commands touched a line naming WorktreeCreate after this was published.
| `/batch <instruction>` | **[Skill](/docs/en/skills#bundled-skills).** Orchestrate large-scale changes across a codebase in parallel. Researches the codebase, decomposes the work into 5 to 30 independent units, and presents a plan. Once a…commands see the edit
Confirmed since
Anthropic's documentation has since written up WorktreeCreate, on Claude Code changelog.
* Improved `/batch` to run where a WorktreeCreate hook provides the agent worktrees, not only inside a git repositorychangelog see the edit
One source agreesOne thing we can check says the same as this entry.
Anthropic's documentation agrees
Anthropic's documentation has since written up WorktreeCreate, on Claude Code changelog.