use-worktrees

use-worktrees is a skill for Claude Code, Codex from maxedapps/agent-skills. It costs 65 tokens per session (652 once invoked), scanned A, original, MIT.

A guide for using Git worktrees, which are extra working directories connected to separate Git branches. It defines where to create them, how to sync them, and how to complete their work safely.

In plain words
What is it for?
Use it when feature or milestone work needs its own checkout, when a worktree must be updated from its target branch, or when finished work needs merging and cleanup.
Why use it?
It keeps feature work isolated so the main checkout is not changed by parallel tasks. It also clarifies branch syncing, setup, deployment separation, and cleanup responsibilities.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it when feature or milestone work needs its own checkout, when a worktree must be updated from its target branch, or when finished work needs merging and cleanup.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/maxedapps/agent-skills/use-worktrees
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add maxedapps/agent-skills --skill use-worktrees
Clone the repo
git clone --depth 1 https://github.com/maxedapps/agent-skills

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for use-worktrees

README.md
[![agentmods](https://agentmods.dev/badge/skills/maxedapps/agent-skills/use-worktrees/github.svg)](https://agentmods.dev/skills/maxedapps/agent-skills/use-worktrees)
Your own site
<a href="https://agentmods.dev/skills/maxedapps/agent-skills/use-worktrees"><img src="https://agentmods.dev/badge/skills/maxedapps/agent-skills/use-worktrees/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for use-worktrees

Your own site · 80×15
<a href="https://agentmods.dev/skills/maxedapps/agent-skills/use-worktrees"><img src="https://agentmods.dev/badge/skills/maxedapps/agent-skills/use-worktrees.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 65 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 652 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00065 $0.00652
Opus 5.5 $0.00026 $0.00261
Sonnet 5.5 $0.00013 $0.00130
Haiku 4.5 $0.00006 $0.00065

Measured 1mo ago against content hash 027e0acbc7d1, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

use-worktrees scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/use-worktrees/SKILL.md · 61 lines

How it starts

The opening of the file, as written. The whole thing — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Use git worktrees

Create

Worktrees live only in ~/worktrees/<repo-dir>/<branch-slug> (branch name with / → -). Create them with Git:

git worktree add ~/worktrees/<repo-dir>/<branch-slug> -b <branch> <target-branch>

For an existing branch, such as an open PR, run git fetch origin <branch>:<branch> and drop -b. Inside herdr (HERDR_ENV=1), don't use herdr worktree create: it opens a separate workspace. Open the worktree in a new pane of your current workspace instead:

herdr pane split --current --direction right --cwd <worktree-path> --no-focus

Tag the new pane as your worker, with its id from the split's reply (result.pane.pane_id), so agentmanager shows it under you:

herdr pane report-metadata <new-pane-id> --source agentmanager --token parent=$HERDR_PANE_ID

Then set it up: install dependencies and copy ignored config such as .env from the main checkout. Give concurrent worktrees their own deployment stage; never let parallel checkouts mutate the same deployment state.

Sync

  • Merge the target branch into the worktree branch (git merge <target-branch>) at checkpoints and before touching anything parallel work may have changed. Never rebase or rewrite shared branches.
  • Run the full check suite after every sync.
  • Migrations: if the target took a migration number this branch also uses, renumber this branch's migration, even without a textual conflict.
  • Lockfiles: keep both sides' manifest changes and regenerate the lockfile; never hand-edit it.

Finish

Do a final sync, get the checks green, push, and open a pull request. Never merge into the target branch yourself; that happens on the PR, when the user asks. No force-push, no bulk --theirs/--ours resolutions.

Cleanup

Whoever created a worktree removes it; an agent working inside one leaves it alone. Once everything is committed, pushed, and in an open PR, remove it right away; later reviews or fixes use a fresh worktree.

git worktree remove <path>
git branch -d <branch>

Read the full file on GitHub · 61 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 1mo ago Changed · +6 lines 027e0acbc7d1
  2. 1mo ago Changed · -7 lines · +3 tokens per session dc26a273c172
  3. 1mo ago Changed · -2 lines · -11 tokens per session 783059718c56
  4. 2mo ago First seen · 64 lines · 73 tokens per session scan A c1d66d550f8c

Subscribe to this mod's changes

use-worktrees is a skill published in the GitHub repository maxedapps/agent-skills (60 stars, last pushed 1mo ago), licensed MIT. It adds 65 tokens to every session and 652 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.