omc

Coordinate multi-agent workflows in Claude Code using omc CLI commands.

40|9|Updated Jan 4, 2026
One-click install
npx skills add https://github.com/akillness/oh-my-skills --skill omc-akillness
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: omc
Source: https://github.com/akillness/oh-my-skills/tree/main/.agent-skills/omc
Command: npx skills add https://github.com/akillness/oh-my-skills --skill omc-akillness

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Orchestrates teams-first, multi-agent workflows inside Claude Code, enabling coordinated planning, execution, and verification, with a shared task list across diverse agents and modes.

Core Features & Use Cases

  • Canonical Team mode with a 5-stage pipeline (plan → prd → exec → verify → fix) for shared task lists.
  • Supports Autopilot, Ralph, Ultrawork, and CCG modes for flexible, multi-model orchestration.
  • Rich agent system catalog and CLI tools to manage and monitor collaborative tasks.
  • Use cases include large-scale feature work, cross-model code reviews, and guaranteed completion with persistent loops.

Quick Start

Install the oh-my-claudecode plugin via Claude Code's marketplace, run omc-setup to configure, and start a Team task with /oh-my-claudecode:team 3:executor "sample task".

Frequently Asked Questions about omc

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

FAQPage Schema
How do I orchestrate multi-agent workflows inside Claude Code?

Multi-agent workflows inside Claude Code are orchestrated using a teams-first approach with a 5-stage pipeline (plan, prd, exec, verify, fix) for shared task lists. You coordinate agents via CLI commands and magic keywords.

What is the canonical Team mode for multi-agent orchestration?

The canonical Team mode is a 5-stage pipeline: plan, prd, exec, verify, and fix. It enables coordinated planning, execution, and verification across diverse agents with a shared task list.

Do I need to enable experimental features to use multi-agent teams in Claude Code?

Yes, you must enable the CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS flag before using multi-agent orchestration. After enabling it, run the setup command to configure the environment and CLI tools.

How do I start a multi-agent task with a specific workflow mode?

You start a multi-agent task by invoking the team command with a mode and executor count, such as /oh-my-claudecode:team 3:executor "sample task". This initiates the workflow using the configured orchestration pipeline.

Can I use different orchestration modes for cross-model code reviews?

Yes, flexible multi-model orchestration is supported through Autopilot, Ralph, Ultrawork, and CCG modes. These modes allow you to manage diverse workflows including cross-model code reviews and large-scale feature work.

When should I use persistent loops for multi-agent code generation?

Persistent loops in multi-agent orchestration are used for guaranteed task completion. You should use them when executing large-scale feature work that requires continuous verification and fixing until the shared task list is fully resolved.