Install
$ agentstack add skill-ondrejdrapalik-worktree-management-skills-worktree-management-skills ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ● Environment & secrets Used
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
New Worktree
Resolve the branch name first
- If
$ARGUMENTScontains a branch name, use it exactly. - If it does not, infer one from the current task or the user's most recent
request, and keep any flags from $ARGUMENTS.
Infer it yourself. No subagent, no extra model call. You already have the context that makes the name good. Use 2-5 lowercase kebab-case words with a prefix that matches intent:
fix/bug fixesfeat/new behaviourchore/maintenancerefactor/restructuring without behaviour change
A bare name with no / gets feat/ prepended by the script, so don't hand-prepend it twice.
Run it
wt new [-c|--clean] [-b|--branch ]
Display the output. Done. The script owns everything else:
- checkout at
~/_wt//, or reattach if it already exists - surface: a single-pane herdr workspace inside herdr, a tmux window inside
tmux, neither outside both. Detected, not chosen by you. Split the pane afterwards if the work wants a shell or a server; there is no template.
- the new workspace nests under the herdr workspace you ran from, not under
whichever workspace happens to hold the repo
- uncommitted work carried over by default;
-cfor a clean checkout .env*symlinked from the repo root and from the subdirectory you invoked
from, .env.example skipped, symlink typechanges hidden with skip-worktree
- the checkout's
mise.tomltrusted before installing - deps installed from the nearest lockfile (pnpm/bun/yarn/npm)
- the working path copied to the clipboard (macOS
pbcopy)
Invoking from a subdirectory is meaningful: a package-level lockfile there wins over the repo root, and the tmux window opens at the matching subdirectory. herdr always opens its pane at the checkout root, so cd in from there.
When something is wrong
wt: command not found means scripts/ is not on PATH. Run ./install.sh from this repo rather than reimplementing the flow inline. If the worktree is created but the surface is not, check HERDR_ENV / $TMUX; the checkout is still valid and wt open will render it.
mise ERROR ... are not trusted inside a worktree means the trust step did not run. mise trust in the checkout clears it.
Other commands
wt is the CLI. Prefer wt new from this skill. Focused skills for the other verbs live in skills/ (wt, wt-remove, wt-clear, wt-cleanup-loop). If those are loaded, use them. If not, run the CLI from here:
wt open
wt remove [branch|slug|path] [--force]
wt clear [--merged-only]
wt cleanup-loop [tick] [duration]
wt list
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: OndrejDrapalik
- Source: OndrejDrapalik/worktree-management-skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.