swarm

Orchestrate parallel AI agent implementation across isolated Git worktrees.

6|Updated Feb 25, 2026
One-click install
npx skills add https://github.com/l3wi/agents-workflow --skill swarm-l3wi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: swarm
Source: https://github.com/l3wi/agents-workflow/tree/main/.pi/skills/swarm
Command: npx skills add https://github.com/l3wi/agents-workflow --skill swarm-l3wi

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill automates the complex process of implementing software features by coordinating multiple AI agents to work in parallel across different code branches, ensuring efficient and structured development.

Core Features & Use Cases

  • Parallel Agent Execution: Spawns and manages multiple worker agents simultaneously in dedicated Git worktrees.
  • Automated Merging: Integrates completed work from phase branches back into the main feature branch.
  • Progress Monitoring & Resumption: Tracks agent progress and allows for resuming interrupted orchestrations.
  • Use Case: When developing a new user authentication system, this Skill can assign different agents to work on the frontend UI, backend API, and database schema in parallel, merging their contributions seamlessly.

Quick Start

Execute the implementation plan for the 'user-avatar' feature.

Frequently Asked Questions about swarm

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

FAQPage Schema
How do I execute parallel agent implementation for software features across isolated Git worktrees?

Parallel agent implementation is executed by spawning and managing multiple AI worker agents simultaneously across dedicated Git worktrees, automating the execution of development phases and handling worktree merging.

What is the best way to orchestrate AI agents to work on different code branches in parallel?

The best way to orchestrate AI agents on different code branches is by coordinating worker agents across isolated Git worktrees, allowing them to implement frontend UI, backend API, and database schema in parallel with automated merging.

Do I need a pre-defined plan before starting workflow automation with parallel execution?

Yes, parallel execution workflow automation requires a pre-defined plan from the `/skill:plan` phase to provide the structured development phases that the worker agents will execute across the Git worktrees.

Can I resume interrupted workflow orchestration if an AI agent stops working midway?

Yes, you can resume interrupted workflow orchestration because the system tracks agent progress continuously, allowing you to pick up and resume the orchestration seamlessly from where it was interrupted.

How does automated merging handle completed work from phase branches back into the main feature branch?

Automated merging integrates completed work by taking the contributions from individual phase branches and seamlessly merging them back into the main feature branch once the parallel agents finish their implementation tasks.

What are the limitations of parallel agent orchestration for complex software development?

Parallel agent orchestration is limited by its dependency on a pre-defined plan and isolated Git worktrees, meaning it cannot function without an existing implementation strategy or outside version-controlled environments.