oh-my-claudecode

A weapon, not a tool

Multi-AI orchestration plugin for Claude Code. Coordinate Claude, Gemini, and Codex with 19 specialized agents, 28 skills, and MCP-powered tools.

--- stars
--- /mo
---
Get Started View on GitHub
Oh My ClaudeCode character
NEW v4.8.0

What's New in 4.8.0

Tracer Agent for evidence-driven causal analysis, 21 security vulnerability patches, HUD token tracking, and team governance hardening.

Tracer Agent & Trace Skill

New evidence-driven causal tracing agent with competing hypotheses, evidence ranking, and discriminating probe recommendations. Enables systematic root-cause analysis with scientific rigor.

Security Hardening

Patched 21 security vulnerabilities including SSRF bypass via IPv6-mapped addresses, command injection in tmux launch, prototype pollution in deepMerge, and shell injection vectors.

HUD Token Usage Tracking

Real-time token usage display in HUD with optional transcript token totals for better cost visibility. Monitor your AI usage as you work.

OMX Team Governance Backport

Hardened team runtime with leader nudge guidance, governance enforcement, and improved pane stall heuristics ported from OMX production runtime.

Unified MCP Registry

Synchronized MCP registry now syncs to Codex config for consistent server management across Claude Code, Codex CLI, and other tools.

Reliability Improvements

23+ PRs merged including Windows hook path preservation, ExitPlanMode unblocking, Codex worker lifecycle sync, and 15 additional bug fixes.

View Release Notes Full Changelog

Multi-AI Orchestration

Claude, Codex, and Gemini CLI workers running in parallel tmux panes — each doing what it does best.

Claude

The Conductor

Gemini Design & Vision
Codex Analysis & Review

Gemini CLI Worker

Spawns a real gemini CLI process in a tmux pane. 1M token context for design review and UI work across your entire codebase.

  • UI/UX design review
  • Documentation writing
  • Large-context file analysis

Codex CLI Worker

Spawns a real codex CLI process in a tmux pane. Deep code analysis, architecture validation, and security review.

  • Architecture review
  • Security analysis
  • Code review validation

Claude Orchestrates

Claude decomposes tasks, fans out work to tmux workers via /omc-teams or /ccg, and synthesizes results — no MCP servers required.

  • On-demand worker spawn
  • Graceful fallbacks
  • Result synthesis

Get Started in Seconds

Install via Claude Code plugin marketplace, or use the npm package directly: oh-my-claude-sisyphus@4.8.0.

1

Install the plugin

# In Claude Code, run:
/plugin marketplace add https://github.com/Yeachan-Heo/oh-my-claudecode
/plugin install oh-my-claudecode
Alt

Or install the npm package directly

# CLI install via npm:
npm install -g oh-my-claude-sisyphus@4.8.0
2

Run setup

/oh-my-claudecode:omc-setup
# That's it! You're ready.

Powerful Execution Modes

Choose how you want to work. From full autopilot to token-efficient parallel execution.

Autopilot

Full autonomous execution from high-level idea to working, tested code. No manual intervention required.

Ralph

Self-referential loop until completion with architect verification. The boulder never stops rolling.

Ultrawork

Maximum parallelism with aggressive agent delegation. Process multiple tasks simultaneously.

Deep Interview

Socratic questioning to clarify vague ideas before execution. Exposes hidden assumptions and measures clarity.

Team

N coordinated agents on a shared task list using Claude Code native teams. Real-time messaging and task coordination.

Planning

Strategic interview-based planning workflow. Gather requirements before diving into implementation.

Powerful MCP Tools

Language server integration, structural code analysis, and persistent state - all built in.

LSP Integration

Hover info, go-to-definition, find references, and project-wide type checking with any language server.

AST Grep

Structural code search and replace. Find patterns by syntax, not just text matching.

Python REPL

Persistent Python environment with pandas, numpy, and matplotlib for data analysis.

State & Memory

Notepad, project memory, and session state that persists across conversations.

Magic Keywords

Just say the keyword and the right mode activates automatically.

Keyword Effect Example
autopilot Full autonomous execution "autopilot: build a todo app"
ralph Persistence mode until completion "ralph: refactor auth system"
ulw Maximum parallelism (ultrawork) "ulw fix all errors"
plan Planning interview mode "plan the new API"
ralplan Iterative planning consensus "ralplan this feature"
deep-interview Socratic requirements clarification "/deep-interview 'vague idea'"
team Coordinated native teams (replaces swarm/ultrapilot) "team 5:executor fix all errors"
swarm Deprecated Use team instead "team 5:executor fix all errors"

19 Specialized Agents

Unified agents organized by domain. Use the model parameter to match complexity.

BUILD Analysis

Build & Analysis

explore analyst planner architect debugger executor code-simplifier

Core workflow from exploration to verification

REVIEW Quality

Review Lane

security-reviewer code-reviewer critic

Multi-dimensional code quality checks

DOMAIN Specialists

Domain Specialists

document-specialist test-engineer designer writer qa-tester scientist git-master tracer

Deep expertise for specialized tasks

All agents use model-optimized routing (Opus for architecture/analysis, Sonnet for standard work, Haiku for quick lookups).

Zero Learning Curve

No docs needed. No commands to memorize. Just say what you want.

Just Talk

"Build me a REST API" - that's it. Autopilot detects your intent and orchestrates 19 agents automatically.

Smart Detection

Say "fast" and it activates parallelism. Say "don't stop" and it persists. Every intent maps to the right mode.

Self-Orchestrating

Delegates to specialists, routes to the right model tier, verifies with architect. You focus on the what, not the how.

Plugin Install

Install from Claude's plugin marketplace, run /oh-my-claudecode:omc-setup - done. Everything else is automatic.