Worktrunk Git Worktree Manager

What it does

When your agent uses it

• Running several agent sessions in parallel on separate git branches
• Setting up wt hooks or editing wt.toml
• Creating, listing, merging or removing worktrees with wt
• Automating dependency or dev server setup for every new worktree

Example prompts

• “Start three worktrees with wt, one per task, and launch Claude in each.”
• “How do I configure a wt hook that installs dependencies in every new worktree?”
• “List my worktrees with wt and merge the finished feature branch into main.”

Requirements

• Worktrunk (`wt`) installed, for example with Homebrew
• Shell integration enabled with `wt config shell install`
• Git

Workflow steps

4 steps, taken from the step headings in SKILL.md.

1. 1Create and Switch
2. 2Inspect Status
3. 3Parallel Agents Pattern
4. 4Merge or Clean Up

What it can do on your machine

Read from SKILL.md and the folder at commit 28dc323. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Shell commands in SKILL.md call:

  • claude
  • git
  • brew
  • cargo
  • winget
  • gh

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  Links to these hosts (documentation or services it may open):

  • worktrunk.dev

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Worktrunk Git Worktree Manager loads about 1.9k tokens when it runs, and up to ~3.6k if it reads all its reference files. Until then it costs about 151 tokens; SKILL.md has 418 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~151

When it runs · the whole SKILL.md, loaded when a task matches

~1.9k

With references · SKILL.md plus every file in references/, read only if the agent opens them

~3.6k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

• NoteMentions a .env fileSKILL.md:233

  env = "cp {{ primary_worktree_path }}/.env.local .env.local 2>/dev/null || true"

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from jjmartres/opencode at commit 28dc323, republished under its MIT licence (© jjmartres). 418 words, ~1,917 tokens.

name

worktrunk

description

Use this skill when the user asks about Worktrunk (wt), git worktree management, running parallel AI agents with worktrees, setting up wt hooks, configuring wt.toml, using `wt switch`, `wt list`, `wt merge`, `wt remove`, `wt step`, LLM commit messages, or integrating Worktrunk with your AI agent. Also trigger when the user wants to run multiple AI agent sessions in parallel across isolated git branches, or automate dev server / database / dependency setup per worktree. Always use this skill for any question involving the `wt` CLI, worktree lifecycle automation, or the worktrunk plugin.

# macOS/Linux (recommended)
brew install worktrunk && wt config shell install

# Cargo
cargo install worktrunk && wt config shell install

# Windows
winget install max-sixty.worktrunk
git-wt config shell install

wt switch --create feature-auth         # New branch + worktree
wt switch --create feature-auth -x claude  # + launch Claude in it
wt switch existing-branch               # Switch to existing worktree
wt switch -                             # Previous worktree

wt list                    # All worktrees with status, commits, CI
wt list --full --branches  # Include remote branches without worktrees
wt list --format=json      # Machine-readable output

wt switch -x claude -c feature-a -- 'Add user authentication'
wt switch -x claude -c feature-b -- 'Fix the pagination bug'
wt switch -x claude -c feature-c -- 'Write tests for the API'

wt step commit              # Commit staged changes with optional LLM message
gh pr create                # Open PR
wt remove                   # After PR merged

wt merge main
# Commits, rebases onto main, fast-forward merges, removes worktree

[post-start]
copy = "wt step copy-ignored"

[post-start]
server = "npm run dev -- --port {{ branch | hash_port }}"
[pre-remove]
server = "lsof -ti :{{ branch | hash_port }} -sTCP:LISTEN | xargs kill 2>/dev/null || true"

[pre-merge]
test = "npm test"
build = "npm run build"

wt step commit              # Commit staged changes, generate message
wt step commit --all        # Stage all + commit

[commit.generation]
model = "claude-opus-4-5"   # or any supported model
provider = "anthropic"

claude plugin marketplace add max-sixty/worktrunk
claude plugin install worktrunk@worktrunk

{
  "statusLine": {
    "type": "command",
    "command": "wt list statusline --format=claude-code"
  }
}

zellij run -- wt switch --create fix-auth -x claude -- \
  'Fix the session timeout bug in auth module'

# .config/wt.toml
[post-create]
zellij = """
zellij action new-tab --name {{ branch | sanitize }}
zellij action new-pane -- claude
"""

[post-create]
env = "cp {{ primary_worktree_path }}/.env.local .env.local 2>/dev/null || true"

[post-start]
copy = "wt step copy-ignored"

[pre-merge]
lint = "make lint"
test = "make test"

[post-merge]
notify = "echo '✓ Merged {{ branch }} → {{ target }}'"

© jjmartres, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 2 other files (references) in opencode/skills/worktrunk of jjmartres/opencode.

• SKILL.md
• references/commands.md
• references/hooks.md

Open the folder on GitHubat commit 28dc323