using-bones-swarm

Create and manage bones swarm sessions with worktrees and hub branches.

4|Updated Jan 29, 2026
One-click install
npx skills add https://github.com/danmestas/wardrobe --skill using-bones-swarm
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: using-bones-swarm
Source: https://github.com/danmestas/wardrobe/tree/main/skills/using-bones-swarm
Command: npx skills add https://github.com/danmestas/wardrobe --skill using-bones-swarm

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Helps developers isolate feature work by creating and managing a bones swarm session that bundles a worktree, a claimed task, and an open hub branch, ensuring isolated, reproducible work contexts.

Core Features & Use Cases

  • Isolated work context: Each swarm session provides a dedicated leaf, task claim, and worktree, isolating work from the main workspace.
  • Atomic lifecycle: Follows a 5-step lifecycle (join, cwd, work, commit, close) to manage lifecycle consistently.
  • Recovery & safety: Supports recovery commands to rejoin or recover from stale sessions and slot-based navigation.
  • Reusability: Enables reuse of workspaces across related tasks by switching leaves when appropriate.

Quick Start

Join a swarm for a slot and task, then navigate to the worktree to begin isolated work.

Frequently Asked Questions about using-bones-swarm

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I isolate feature work in a separate worktree and branch?

To isolate feature work, you can use a bones swarm session that bundles a dedicated worktree, a claimed task, and an open hub branch together. This ensures your work context remains separate and reproducible from the main workspace.

What is the best way to manage parallel feature branches without conflicts?

Managing parallel feature branches is best handled by creating isolated swarm sessions for each task. Each session claims a dedicated leaf and worktree, allowing you to run multiple isolated experiments simultaneously without overlapping changes.

How do I start an isolated worktree session for a new task?

Starting an isolated worktree session involves joining a swarm for a specific slot and task. You then navigate to the newly created worktree directory to begin working through the 5-step lifecycle: join, cwd, work, commit, and close.

Can I recover a stale worktree session if my feature branch work is interrupted?

Yes, you can recover a stale worktree session using built-in recovery commands. The system supports rejoining interrupted sessions and provides slot-based navigation to help you safely resume your isolated work context.

When should I use an isolated worktree for feature development?

You should use an isolated worktree when starting feature work that requires strict isolation or when you need to run parallel experimentation within a bones workspace. It keeps your main workspace clean and your changes reproducible.