AI/TLDR

oh-my-claudecode

Multi-agent orchestration layered on top of Claude Code

Coding-Agent Harnesses & RuntimesOpen source
Language
TypeScript
License
MIT

Overview

oh-my-claudecode — OMC — is a plugin and companion CLI that sits on top of Claude Code and turns it into a multi-agent workflow runner. Its pitch is that you should not have to learn the harness: after a one-time setup you describe a task in one line and OMC decides which specialised agents to run, in what order, and how to hand work between them.

It exposes two distinct surfaces that are easy to confuse. In-session slash commands (`/autopilot`, `/ralph`, `/execute`, `/deep-interview`, `/team`) run inside a Claude Code session and drive the native in-session workflows. The terminal CLI, installed from npm as `oh-my-claude-sisyphus`, adds commands that make sense outside a session — `omc setup`, `omc ask` to consult another provider, `omc team` to launch tmux CLI workers, and `omc lookout scan` as an advisory pre-flight danger scan before an unattended run.

Multiple model providers are first-class: the advisor flow routes to `claude`, `codex`, `gemini`, `antigravity`, `grok` or `cursor`, so a plan written by one model can be reviewed by another. Autopilot supports named stage profiles — ordered sequences such as `plan-build-qa` configured in `.claude/omc.jsonc` — with state, cancel, resume and HUD lifecycle carried across stages. The project also exports TypeScript helpers such as `createOmcSession()` for programs built on the Claude Agent SDK.

What it does

  • `/autopilot` runs a task end to end through staged agents; named stage profiles pin the sequence, with cancel, resume and crash-safe state
  • Two surfaces — in-session slash commands inside Claude Code, and an `omc` terminal CLI for setup, advisors and workers
  • Cross-provider advisors: route a review or second opinion to Codex, Gemini, Grok, Cursor or Antigravity from inside a Claude session
  • Parallel teams — `omc team` launches tmux CLI workers, `/team` runs the in-session native team workflow
  • `omc lookout scan` reads a task briefing and the workspace before an unattended run and reports danger findings with severities and CI-friendly exit codes
  • An SDK surface (`createOmcSession()`, prompt expansion helpers) for local Node.js programs on `@anthropic-ai/claude-agent-sdk`

Getting started

The marketplace install is the recommended path for Claude Code users. The two slash commands must be entered one at a time — pasting both lines together fails.

Install the plugin

Run these as separate Claude Code slash commands.

bashbash
/plugin marketplace add https://github.com/Yeachan-Heo/oh-my-claudecode

/plugin install oh-my-claudecode

Or install the CLI

The npm package gives you the terminal `omc` commands and the SDK helpers.

bashbash
npm i -g oh-my-claude-sisyphus@latest

Set up

Either entry point works; use the one matching how you installed.

bashbash
# inside a Claude Code / OMC session
/omc-setup

# from your terminal
omc setup

Run a task

Autopilot takes the task from plan to finished change. A named stage profile is selected with `--workflow`.

bashbash
/autopilot "build a REST API for managing tasks"

/autopilot --workflow plan-build-qa "build a REST API for managing tasks"

Ask another model, or scan before an unattended run

Advisors and the pre-flight scan are terminal commands.

bashbash
omc ask codex "review this patch"
omc team 2:codex "review auth flow"
omc lookout scan --brief "..." --json --strict

Commands and code are distilled from the project's own documentation — always check the official repo for the latest.

When to use it

  • Hand a whole feature to Claude Code as one sentence and let staged agents plan, build and QA it
  • Get a second opinion on a patch from a different vendor's model without leaving the session
  • Run several coding workers in parallel on independent parts of a task
  • Scan a task briefing for risky operations before starting an unattended run in CI

How oh-my-claudecode compares

oh-my-claudecode alongside other open-source coding-agent harnesses & runtimes tools AI/TLDR tracks, ranked by GitHub stars.

ToolStarsWhat it does
CC Switch★ 133kA cross-platform desktop app that manages provider configuration for eight coding-agent CLIs from one place, with a local failover proxy, MCP and skill management, and spend tracking.
Oh My OpenAgent★ 69.1kOpinionated agent bundle for OpenCode and Codex CLI that ships preconfigured sub-agents, lifecycle hooks and built-in MCP servers behind a single ultrawork command.
oh-my-claudecode★ 39.2kMulti-agent orchestration layered on top of Claude Code
herdr★ 39kA single-binary terminal multiplexer built for coding agents: persistent sessions across local and SSH machines, per-pane working/blocked/idle status, and a socket API agents drive themselves.
AionUi★ 32.9kAn open-source desktop workspace that pairs a built-in agent engine with dozens of external CLI agents, adding scheduled runs, MCP tools and remote access from a WebUI or chat apps.
cmux★ 27.2kA Ghostty-based macOS terminal built around coding agents: vertical tabs that show git branch, PR status and ports, attention notifications when an agent needs you, and a scriptable built-in browser.
cc-connect★ 15.5kA Go daemon that bridges local coding agents — Claude Code, Codex, Cursor Agent, Gemini CLI, OpenCode, Kimi CLI and more — to Feishu, DingTalk, Slack, Telegram, Discord, LINE and WeCom, so you drive them from chat.
Claude Code Haha★ 14.5kA cross-platform desktop workspace for Claude Code: multi-session workspaces, worktree launch, diff review, a built-in browser preview, GUI permission approval, per-session model choice and visual MCP and subagent managers.