oh-my-copilot

Orchestrate multi-agent execution through Sisyphus, Hephaestus, Prometheus, and Ralph Loop patterns.

10|Updated Mar 22, 2026
One-click install
npx skills add https://github.com/Lee-SiHyeon/oh-my-copilot --skill oh-my-copilot
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: oh-my-copilot
Source: https://github.com/Lee-SiHyeon/oh-my-copilot/tree/main
Command: npx skills add https://github.com/Lee-SiHyeon/oh-my-copilot --skill oh-my-copilot

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill solves the problem of coordinating complex, multi-step software work by delegating tasks across a structured agent team instead of relying on a single, error-prone response.

Core Features & Use Cases

  • Multi-agent orchestration: Coordinates Sisyphus, Hephaestus, Prometheus, and Ralph Loop patterns for end-to-end execution.
  • One-command full workflow: Use /ultrawork to trigger a full pipeline that covers planning, parallelization, deep implementation, and verification.
  • Task-driven entry points: Use dedicated triggers like /prometheus (planning) and /ralph-loop (self-correcting loop) to match your task’s needs.
  • Primary best-practice fit: Best for requests framed as "ultrawork", "orchestrate", "multi-agent", or "deep work", including “REST API auth system” style implementations.

Quick Start

Run /ultrawork "REST API 인증 시스템 구현해줘" to start the full multi-agent orchestration pipeline.

Frequently Asked Questions about oh-my-copilot

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

FAQPage Schema
How does multi-agent orchestration work for Copilot CLI?

Multi-agent orchestration routes complex software work through structured patterns like Sisyphus, Hephaestus, Prometheus, and Ralph Loop. It delegates tasks across an agent team for planning, deep implementation, and self-correcting verification instead of relying on a single response.

How do I start a full multi-agent workflow for deep implementation tasks?

To start a full multi-agent workflow, use the one-command /ultrawork trigger. This automatically enables the intended agent set, covering the entire pipeline from planning and parallelization to deep implementation and verification.

Can I use specific agents for planning and self-correction instead of the full pipeline?

Yes, you can use task-driven entry points instead of the full pipeline. Use the /prometheus trigger for dedicated planning and the /ralph-loop trigger to initiate a self-correcting loop tailored to your specific task needs.

When should I use multi-agent execution for software engineering?

Use multi-agent execution for complex, multi-step software work like implementing a REST API auth system. It is best for requests framed as ultrawork, orchestration, multi-agent, or deep work where a single error-prone response is insufficient.

Does multi-agent orchestration include safety hooks for execution?

Yes, orchestration supports safe, trigger-based activation with safety hooks. This ensures structured coordination across Sisyphus, Hephaestus, Prometheus, and Ralph Loop patterns during end-to-end execution.