cmux Multi-Agent Workstation Setup

What it does

Sets up cmux as a multi-agent coding workstation with documented config, custom terminal appearance, worktree and workspace-rename helper scripts, and per-repo templates. Includes a guided interview that adapts it all to your own repositories.

When to use it

• When you run several AI agents in parallel across different repositories.
• When you want a consistent terminal and workflow across projects.
• When you need to isolate changes per worktree without losing context.

Steps

1. Answer the guided interview about your repos and preferences.
2. The assistant generates the config and custom terminal appearance.
3. Worktree and workspace-rename helper scripts are created.
4. Per-repo templates are built based on what you specified.
5. You review the generated setup before applying it.
6. You decide when to activate the config in your real environment.

The skill

---
name: cmux-multi-agent-setup
description: Turns cmux into a multi-agent coding workstation with appearance, worktree helpers, and per-repo templates.
---

# cmux Multi-Agent Workstation Setup

## 1. What it does

- Sets up cmux as a multi-agent coding workstation with documented config, custom terminal appearance, worktree and workspace-rename helper scripts, and per-repo templates. Includes a guided interview that adapts it all to your own repositories.

## 2. When to use it

- When you run several AI agents in parallel across different repositories.
- When you want a consistent terminal and workflow across projects.
- When you need to isolate changes per worktree without losing context.

## 3. Step by step

- Answer the guided interview about your repos and preferences.
- The assistant generates the config and custom terminal appearance.
- Worktree and workspace-rename helper scripts are created.
- Per-repo templates are built based on what you specified.
- You review the generated setup before applying it.
- You decide when to activate the config in your real environment.

## 4. Safety rules

- Safety rules: everything ships as a draft first; nothing is sent, published, or paid for without your explicit approval; no data, numbers, or results are ever invented; no credentials or secrets are typed into chat.

## 5. Best practices

- Always review the full draft before approving any send, publish, or payment.
- If real information (data, numbers, results) is missing, ask for it to be flagged as pending instead of invented.
- Keep credentials and keys out of chat; set them in environment variables or your secrets manager.

## 6. GoHighLevel integration

- Where relevant, results can be saved as contacts, tags, tasks, or notes inside your GoHighLevel sub-account.
- No automation inside GHL (workflow, campaign, trigger) activates without your review first.

For Claude Code, save it as ~/.claude/skills/cmux-multi-agent-setup/SKILL.md. In Cursor, ChatGPT or another assistant, paste it as project instructions.