oh-my-claudecode is the zero-config orchestration layer that turns Claude Code into a coordinated team of 19 specialists — architect, executor, reviewer, and more.
Ship features 3-5× faster. Save 30-50% on tokens. Never compromise on quality.
Trusted by 39,000+ developers worldwide
A single agent juggling every role — planning, coding, testing, reviewing — means slower output, more bugs, and wasted tokens on rework. You're leaving 80% of Claude's potential on the table.
Everything Claude Code should have shipped with. Batteries included, zero config.
Architect plans. Executor builds. QA tests. Security reviews. Designer polishes. OMC delegates to the right agent automatically — no prompting hacks required.
Up to 5 workers in parallel. Fullstack features land in minutes.
Smart routing sends trivial tasks to Haiku. Opus only when needed.
Install, run setup. Describe what you want. OMC picks the mode.
Spin up coordinated squads that share task lists, message each other, and persist until done. No more serial loops.
Architect-backed verification loops. Security review runs before commit. Pass/fail gates eliminate the "it mostly works" trap.
Pick manually, or just describe the work — OMC auto-selects.
Describe the feature. Autopilot plans with architect, builds with executor, tests with qa-tester, and reviews with code-reviewer — all automatically. The flagship experience.
$ /autopilot "build a todo app with auth"
Hand Execute a task and it carries it all the way through — implement, verify, fix, repeat — with no silent partials. In v5 it replaces ultrawork, ultrapilot, swarm, and pipeline. Reach for /team when you want coordinated parallel workers, and /verify to cycle a quality gate on its own.
$ /execute "refactor auth to OAuth and get the suite green"
Spawn a team with a shared pool of tasks. Agents claim atomically, hand off via messages, and synthesize results. The canonical multi-agent surface.
$ /team 3:executor "fix all TypeScript errors in /src"
For hairy bugs and high-stakes refactors. Ralph keeps iterating — plan, fix, verify, repeat — until the Architect agent signs off. Maximum persistence.
$ /ralph "refactor auth to JWT with zero downtime"
19 named agents. Teams-first architecture. Specify Haiku, Sonnet, or Opus per invocation.
Every agent accepts a model parameter.
No tier suffixes. Invoke any agent with haiku,
sonnet, or opus.
Pulled from the 100+ articles already published about OMC.
The only agents swarm orchestration you actually need. It makes multi-agent coding simple — even for the least experienced Claude Code users.
Specialized teammates operating in parallel with shared task lists and inter-agent messaging. The orchestration pattern modern AI coding needs.
Zero-config orchestration layer delivering 3-5× speedup and 30-50% token savings through 32 specialized agents. Trended #1 on GitHub.
Already have Claude Code? You're 30 seconds from a full AI team.
Add OMC to Claude Code's plugin marketplace.
$ /plugin marketplace add https://github.com/Yeachan-Heo/oh-my-claudecode
$ /plugin install oh-my-claudecodeOne command. No config files to edit.
$ /omc-setup
✓ 32 agents registered
✓ 35+ skills loaded
✓ Configuration complete.Natural language in. Shipped feature out.
$ autopilot build a REST API for managing tasks
↳ Activating autopilot…
architect · executor · qa-tester · security-reviewerCurated hands-on reviews, deep dives, and tutorials from the OMC community.
Hands-on report on Opus 4.8 dynamic workflows — spawning a 200-agent swarm to parallelize research and lift evaluation scores.
Hands-on review of multi-agent orchestration capabilities with real-world examples.
In-depth Korean analysis of the plugin's architecture and team-like agent collaboration.
Comprehensive guide covering Claude Code ecosystem including plugins like OMC.
Deep dive into Claude Code's native agent coordination and how OMC unlocks it.
First-hand lessons — architecture, best practices, and common pitfalls.
How OMC transforms Claude Code into a multi-agent system with five execution modes.
Enterprise adoption comparison between vanilla Claude Code and OMC plugin ecosystem.
How agent teams enable specialized teammates to operate in parallel with shared task lists.
Comprehensive breakdown of the multi-agent paradigm shift in AI-assisted development.
From subagents to Agent Teams to external orchestrators — how to orchestrate Claude Code agents across repos and an entire team.
July 2026 roundup — Claude Code as a terminal-based work layer for teams: bugs, files, commands, and git inside the dev flow.
How to wire parallel review agents and vulnerability diff scans into your review workflow with Claude Code plugins and skills.
Complete guide to the 2026 agentic coding stack — plugins, multi-model orchestration, and AI agent teams working in parallel.
Hands-on Korean comparison of AI coding agent stacks — when Claude Code with orchestration layers is the right pick.
Chapter of a Korean Claude Code onboarding course dedicated to OMC — autopilot, the learner skill, HUD session monitoring, and the ecomode/ralph/ultrawork/team modes.
Why skills fix the "Claude forgets between sessions" problem, and which combinations actually change a daily workflow — a companion read to OMC's built-in skill set.
Field report from 12 real repositories — six working plugin combinations, memory routing that cut tokens 60%, three essential hooks, and four setup pitfalls the author reversed.
Clear walkthrough of swarm mode — how parallel specialists improve context efficiency and make large codebases tractable.
For authoring deterministic multi-agent workflow scripts in .claude/workflows/ — scaffolding, sub-agent orchestration, and running custom workflows.
Reference repo for multi-agent coordination with Claude Code's TeammateTool and Task system — parallel reviews, dependency pipelines, self-organizing queues.
Drop-in skill documenting the full coordination pattern catalog for TeammateTool and the Task system.
Large open collection of skills, agents, and commands spanning engineering through business ops — portable across Claude Code, Codex, Gemini CLI, and more.
Analyst-style profile of OMC's adoption curve, the staged team pipeline, and the tmux-based Codex/Gemini workers that make it multi-model.
Korean walkthrough of agent collaboration, feature development and legacy refactoring workflows, LSP/ast-grep foundations, plus honest limitations.
When to spend a full context window on a subagent and when a skill will do — covering dynamic workflows, performance grading, and hooks.
Delegate mode, spawn-prompt detail, and token-cost math for multi-teammate runs — the failure modes OMC's team pipeline is built to avoid.
Decision guide across the three orchestration primitives — scale, context cost, and coordination overhead compared side by side.
End-to-end guide to scripting fan-out over hundreds of subagents while keeping the lead's context small.
Plain-language explainer of how moving the plan into code lets orchestration scale past what a single conversation can hold.
Reference-grade tour of the extensibility surface OMC builds on: hooks, MCP servers, skills, and slash commands.
Curated plugin roundup — Context7, Frontend Design, Superpowers, Chrome DevTools MCP, and the LSP pack — with what each is actually good at.
What a plugin boundary actually buys you — name, version, manifest, install path — and why that turns "works on my machine" into team-wide tooling.
Compares subagents, agent teams, and third-party frameworks — highlighting OMC's five execution modes as the flexible, minimal-setup option.
17-section Korean reference covering orchestration modes, multi-model routing, the ralplan consensus gate, deep-interview parameters, and deepinit.
Ethan Mollick on the agentic harness — how skills, subagents, and MCP let Claude Code sustain autonomous work, and what that means for knowledge work.
Week-long hands-on report on ralplan, ralph, autopilot, and team — how each mode prevented accidents on risky refactoring and parallel work.
Systematic tour of Claude Code's six extensibility primitives — CLAUDE.md, skills, subagents, slash commands, hooks, and MCP — with 25 worked examples.
Hands-on working notes covering installation, the Team/Autopilot/Ralph modes, and natural-language orchestration across OMC's specialized agents.
Korean deep-dive comparing spec-kit, Everything Claude Code, and OMC — positioning OMC as the zero-learning-curve option with execution modes and smart model routing.
Fresh review covering structured planning, execution modes, and Codex/Gemini CLI workers layered on top of Claude Code.
In-depth Chinese-language guide to OMC skills — ralplan, Ralph, Autopilot — with the "conductor, not performer" delegation workflow.
MIT-licensed companion interface built on top of oh-my-claudecode, surfacing execution modes, agents, and skills in a UI layer.
Korean walkthrough covering installation, execution modes, magic keywords, and practical OMC workflows.
Korean introduction to OMC's architecture, agent lineup, and everyday usage patterns.
Practical guide focusing on the features that actually get used daily - ralplan, autopilot, ultraqa, and honest cost analysis.
Comprehensive guide covering the Claude Code ecosystem, from autonomous coding to enterprise deployment strategies.
How three specialized plugins work together to enforce architecture, testing, and quality in every Claude Code session.
Accessible introduction to Claude Code for non-engineers, covering practical workflows and plugin ecosystem.
Comprehensive review covering features, pricing, performance benchmarks, and real-world value assessment.
One of the most influential product newsletters makes the case for Claude Code as essential developer tooling.
O'Reilly Radar explores automated code review patterns with Claude Code and multi-agent quality assurance workflows.
Curated ranking of the best Claude Code plugins including OMC, with feature comparisons and use-case recommendations.
First-hand account of how Claude Code and its plugin ecosystem transformed a developer's approach to AI-assisted coding.
How community tools like OMC drove platform evolution, from multi-agent orchestration to official Agent Teams.
How coordinated agent teams are reshaping software development workflows with parallel execution and shared task management.
Synthesizes insights from creator Boris Cherny and power users - covering plan mode, parallel sessions, and agent teams.
Anthropic's engineering team demonstrates building a working C compiler using Claude Code's multi-agent capabilities.
Comprehensive review of the Claude Code plugin ecosystem with top 10 rankings, security guidance, and custom plugin development tips.
Building plugins for multi-agent workflows using SubagentStart hooks to inject session context into parallel agent teams.
How OMC's five execution modes and 32 specialized agents enable a single developer to achieve the output of a small team.
Step-by-step setup guide for running parallel AI coding agents with shared task lists and inter-agent coordination.
Getting started with agent teams - how teammates self-organize via shared task lists and direct messaging.
Educational breakdown of agent teams architecture - when to use subagents vs teams, cost tradeoffs, and scaling patterns.
Official Anthropic blog post on the Claude Code plugin ecosystem - architecture, marketplace, and building custom plugins.
Popular Laravel developer shares his complete Claude Code configuration, plugins, and daily workflow tips.
Comprehensive walkthrough of OMC's architecture, 28 specialized agents, 26+ skills, and multi-tier memory system.
Hands-on experience running Claude Code's Agent Teams with parallel workers and inter-agent coordination.
First-hand experience report on Claude Code Teams - setup, coordination patterns, and lessons learned.
Practical guide on mastering LLM-powered coding tools - why learning to use AI effectively is the essential 2026 skill.
Deep dive into parallel orchestration patterns - how multi-agent swarms coordinate like a Zerg rush on complex codebases.
Official launch of multi-agent Code Review - a fleet of specialized agents examine PRs for logic errors, security vulnerabilities, and regressions.
Coverage of Anthropic's multi-agent code review launch - how parallel AI agents catch bugs that human reviewers miss, with <1% false positive rate.
Technical deep-dive into Claude Code Review's multi-agent architecture - 54% PR coverage vs 16% before, with specialized agents for different review dimensions.
Analysis of how Claude Code Review's multi-agent approach dispatches specialized agents to examine code from different perspectives before aggregating findings.
Comprehensive breakdown of Claude Code Review's multi-agent architecture, $15-25 per review pricing, and step-by-step enterprise setup guide.
Curated selection of the best Claude Code plugins for extending AI coding capabilities - from orchestration to code review automation.
Deep technical analysis of multi-agent orchestration frameworks for Claude Code - architecture patterns, distributed coordination, and real-world benchmarks.
Curated directory of essential plugins, skills, and MCP servers for Claude Code.
Coverage of Anthropic's enterprise Code Review launch - multi-agent PR analysis for Teams and Enterprise customers.
Nine days of real-world testing distilled into actionable tips.
Curated list of essential SKILL.md files every developer should install.
Plain-language explainer of Claude Code concepts - skills, hooks, subagents, plugins, MCP servers, and agent teams demystified.
HN discussion on the oh-my-claude-sisyphus npm package and the story behind the naming.
Stack-specific plugin recommendations with OMC featured as the top orchestration layer.
Practical guide to coordinating multiple Claude Code agent teams for parallel development workflows.
Coverage of OMC as a structured orchestration layer for collaborative development environments with bilingual English/Korean documentation.
Deep technical paper on team setup, task coordination, git worktree isolation, and prompt engineering strategies.
Hands-on walkthrough of building a Claude Orchestrator multi-agent system for complex development tasks.
How Claude Code evolved from an AI assistant into a full development workspace.
Exploring Claude Code's new Code Review feature and how it scales development workflows.
Analysis of what Claude Code's temporary OpenCode block revealed about competition and strategy in AI coding tools.
Real-world case study: 18 parallel agents fixing 175 TypeScript errors across 33 files.
Practical breakdown of role-based agent specialization.
Layered markdown memory architecture for multi-agent systems.
The evolution from single subagents to true multi-agent teams with direct agent-to-agent messaging.
Practical strategies for agent teams: domain separation, teammate sizing, parallel exploration patterns.
Deep dive into TeammateTool's 13 operations and swarm mode capabilities with Opus 4.6 integration.
Comprehensive roundup of March 2026 features.
Trending #1 on GitHub with 858 stars in 24 hours.
Compares multi-agent orchestration tools for Claude Code.
Comprehensive 2026 guide to the best Claude Code skills and plugins.
Step-by-step guide to Claude Code's swarm mode.
Real-world tested breakdown of the plugins enabling a complete workflow.
Rigorous evaluation of 11 Claude Code plugins across content, business, and coding workflows.
Based on testing 200+ skills - which ones actually improve output vs. hurt it.
Joe Njenga uncovers Claude Code's hidden swarm capabilities.
First-hand test of Claude Code Agent Teams.
Real-world account of handing a codebase to Claude Code's multi-agent swarm.
Coverage of OMC's explosive GitHub growth.
Beginner-friendly introduction to OMC.
Hands-on field notes from running OMC on real projects.
In-depth review highlighting how OMC eliminates babysitting Claude Code.
Comprehensive Q1 2026 breakdown covering major features.
Five-step workflow guide featuring OMC as "an incredible leap".
Comprehensive 50-tip guide covering Plan Mode, CLAUDE.md, hooks, and more.
Community hub dedicated to oh-my-claudecode tips and workflows.
Overview of how OMC transforms a single instance into a coordinated 32-agent team.
Comprehensive handbook on building with AI-assisted development.
Honest firsthand account of learning Claude Code as a staff engineer.
Live cohort course teaching production Claude Code workflows.
First-hand experience report on OMC's local swarm orchestration.
How top developers use Claude Code as their daily OS.
Curated plugin ranking advising developers to run 2-3 plugins max.
Interactive help skill on the MCP Market.
Comprehensive resource compilation covering plugins, skills, MCP servers, and orchestration tools.
Deep-dive into OMC's team-plan pipeline and staged coordination patterns.
Complete setup guide for Agent Teams with shared tasks and inter-agent messaging.
Technical walkthrough showing how to orchestrate multiple Claude Code sessions as a coordinated team.
Hands-on multi-agent development guide covering parallel task execution.
Auto-tracked changelog of every Claude Code release.
Reference gist on TeammateTool, Task system, and orchestration patterns.
Curated plugin ranking for DevOps workflows.
End-to-end marketplace guide covering plugin discovery and evaluation.
Hands-on deployment guide for OMC.
Economist Paul Goldsmith-Pinkham shares his research-oriented Claude Code setup.
Security research team breaks down how orchestration frameworks catch vulnerabilities.
Addy Osmani explores three multi-agent patterns.
Official Anthropic engineering deep-dive on Managed Agents architecture.
Deep analysis of the Claude Code source leak phenomenon with extensive OMC coverage.
Video walkthrough showing install-through-autopilot workflow and real project demo.
Drop-in ultrawork prompts that optimize Claude Code performance with maximum parallel execution.
Ralph-style autonomous AI development loop for Claude Code with intelligent exit detection.
Multi-agent orchestration with Ultrawork mode for Claude Code, inspired by oh-my-opencode.
Walks through how OMC structures AI collaboration via staged pipelines (team-plan → exec → verify) and tmux workers.
In-depth review of OMC's multi-agent layer with practical workflow examples.
Analysis of how OMC fits into the Claude multi-agent landscape alongside Anthropic's Agent Teams and Ruflo.
Engineer-curated ranking of must-have Claude Code extensions with OMC featured prominently.
Production-ready plugin shortlist with installation walkthroughs and team workflow tips.
Hands-on review of the top Claude Code plugins for orchestration, code review, and TDD.
Official guidance on Opus 4.7's literal instruction following, xhigh effort level, and multi-session memory — directly relevant to OMC's upgraded agents.
In-depth review of Opus 4.7 for builders — what changed, what works, and how to adapt your prompts and orchestration workflows.
Practical guide showing why detailed plans outperform vague prompts on Opus 4.7 — pairs perfectly with OMC's planner-first workflows.
Live coverage of Anthropic's Code w/ Claude 2026 conference — announcements, demos, and ecosystem updates including multi-agent orchestration.
Deep dive on scripting deterministic fan-out → reduce → synthesize agent pipelines — the same control-flow model behind OMC's staged team workflows.
Technical guide to delegation, parallel fan-out, and custom agent definitions — foundational concepts that OMC automates with zero config.
Roundup of May 2026 Claude Code developments for startups — from helpers to production tools, with multi-agent orchestration front and center.
June 2026 roundup covering Claude Code 2.1, auto mode, plan mode, and Routines — scheduled cloud agents that run without your machine staying on.
Official Anthropic post on dynamic workflows — Claude Code writing and orchestrating its own multi-agent harness on the fly, the native counterpart to OMC's staged pipelines.
InfoQ news analysis of Claude Code's dynamic workflows — deterministic orchestration scripts, subtask fan-out, and result validation at scale.
Practitioner walkthrough of a real production agentic coding setup — how multi-agent orchestration actually fits into day-to-day engineering work.
Hands-on practical guide to writing and using Claude Code skills effectively — useful companion to OMC's 36+ built-in skills.
Example-driven visual guide to Claude Code with copy-paste templates — from basic concepts to advanced agents.
Community-maintained documentation hub for OMC — installation, agents, skills, and orchestration workflows.
Korean-language technical deep-dive on OMC's architecture, Team Mode pipeline, and orchestration mode comparisons.
Community fork emphasizing zero-learning-curve multi-agent orchestration for Claude Code projects.
Beginner-friendly explainer of OMC's multi-agent orchestration model, execution modes, and how it fits the broader MCP ecosystem.
Beginner-to-power-user Claude Code guide with production-ready templates, agentic workflow walkthroughs, quizzes, and a cheatsheet.
Step-by-step beginner video tutorial covering Claude Code installation, configuration, and first project — useful onboarding companion to OMC.
Long-form beginner tutorial covering installation, daily workflows, and best practices for AI coding with Claude Code.
Comprehensive walkthrough of Claude Code — installation, core workflows, CLAUDE.md configuration, MCP integration, and subagent delegation for complex tasks.
Beginner tutorial covering installation, authentication, the CLAUDE.md project memory system, and essential slash commands — with a hands-on first project build.
Curated evaluation of top Claude Code plugins across development, testing, and DevOps — with installation instructions and a security evaluation framework for building your plugin stack.
Directory profile tracking OMC's agent roster, skills catalog, and adoption — a quick at-a-glance snapshot of what ships in the plugin.
Korean-language development guide walking through OMC setup, execution modes, and how to build on top of the agent layer.
When to reach for subagents vs. agent teams vs. MCP servers — a decision framework for structuring multi-agent work without burning tokens.
How shared task lists, teammate messaging, and a coordinating lead turn a task queue into a working swarm.
Why context-window bloat is the real bottleneck, and how scoped subagents with their own tools and permissions raise the delegation ceiling.
Anatomy of a subagent — custom system prompt, scoped tool list, independent permissions — and the patterns that keep them useful.
The description is the trigger — how to write, scope, and test skills, and when a skill should graduate into a plugin.
Categorized directory of official and community skills for design, testing, docs, and memory.
Hands-on tour of the skills ecosystem since Agent Skills became an open standard — plus a walkthrough of authoring your own.
Git worktree isolation for concurrent agents — the practical ceiling teams hit and how to avoid merge chaos.
Zero-to-first-project walkthrough covering install, project setup, and the core commands worth learning first.
End-to-end reference covering setup, CLAUDE.md, hooks, plugins, and multi-agent workflows in one place.
Hands-on walkthrough of driving several Claude instances across one repo from tmux panes and isolated worktrees — the same substrate OMC's CLI workers use.
Six orchestration patterns across subagents, agent teams, and background agents — with cost analysis and guidance on when native Claude Code beats an external framework.
Give every agent its own directory over one shared Git history — no stashing, no branch collisions, no context switching between parallel runs.
Fan-out, pipeline, debate, supervisor, and swarm — separated properly, with the failure mode and real cost of each. Supervisor comes out as the production default.
Fan-out/fan-in, hierarchical agent trees, and shared-memory message passing — plus why over-spawning agents costs you more than it saves.
Plan first, define shared contracts, split by directory ownership, merge sequentially — the discipline that makes file-ownership partitioning actually hold.
Running agents against a groomed backlog moves your job upstream: you shape the work before it starts and judge it when it lands. Spec becomes the deliverable.
Running parallel sessions without collisions — how worktree isolation removes the class of bug where one session's edits stomp another's.
A .NET-flavored but stack-agnostic setup guide for wiring worktrees into a day-to-day parallel agent workflow.
The month's releases were less about raw capability and more about operability — the workflow shell becoming the product.
Companion guide to the 345-skill open collection — which plugins are worth installing first and how they compose.
Latest in the monthly roundup: session stability, permissions, and background workflows, framed for lean teams shipping on Claude Code.
Korean-language wiki collecting OMC usage reports and tips from across the web, each with its source and observed effect.
The canonical reference for the skill format OMC's 40+ built-in skills are written in: SKILL.md frontmatter, progressive disclosure, and the description-as-trigger contract.
"The description is not metadata — it is the trigger." Careful treatment of when a skill beats a hook or a subagent, plus how to test activation reliably.
All thirty hook events documented. CLAUDE.md relies on model compliance; hooks are enforced by the harness — the distinction that matters for autonomous runs.
Anthropic cut 80%+ of Claude Code's system prompt for Claude 5 with no performance loss. Six shifts follow — contextual judgment over rigid rules, progressive disclosure over upfront loading.
Peer messaging and file locking explained end to end, with agent teams weighed against plain subagents and against Kimi's own swarm approach.
Skill architecture, planning, file layout, and distribution across Claude.ai, Claude Code, and the API — condensed from Anthropic's own authoring guidance.
Nine skill categories from library references to infra ops, and why a Gotchas section carries more weight than anything else you write.
Distilled from reading Anthropic's official skill repo: progressive disclosure, descriptions that actually fire, and explicit anti-pattern avoidance.
Structure, writing technique, testing strategy, and sharing — aimed at skills that trigger reliably instead of sitting unused.
Ten skills worth installing, grouped by category, plus a framework for rolling your own when nothing off the shelf fits.
CLAUDE.md supplies the what and why; hooks enforce the how. Ready-to-paste examples for linting, testing, secret scanning, and notifications.
Seven production-ready hook configurations for file protection, auto-formatting, and workflow automation — hooks fire every time, prompts don't.
Agent workflows guide plus an orchestration framework: dynamic team assembly and cost-aware task routing, with a skills-vs-teams cost analysis.
The same orchestration experience for OpenAI Codex CLI, from OMC's creator: hooks, agent teams, a HUD, and the shared five-stage Team pipeline.
The lighter sibling for when OMC feels like overkill: SDK-based, runs on the coding plan you already pay for, and plugs into OpenClaw, Hermes, and Grokbot runtimes.
Japanese walkthrough of the deep-interview → ralplan → autopilot pipeline, ending in a 13-screen, 40-API full-stack MVP generated in about five hours.
Why the same author made different design calls: OMC leans on the Claude Agent SDK and 19 agents with model routing, while OMX trims to three npm deps and Rust hot paths.
Ruflo, Maestro, Claude Octopus, and Codex Peer Review put through real tasks — how specialist division of labor and cross-model consensus review each catch bugs.
Skills vs plugins explained, marketplace installs done safely, and how to author your own — the packaging layer OMC ships through.
The /design skill for terminal mockups, a Concise output style, and auto-mode permissions written as plain-language rules — the host changes OMC v5 lands on.
Japanese introduction to OMC's 19 agents and its parallel full-stack workflow — a Go API and a Next.js frontend built side by side — plus the 30-50% token savings from model routing.
Recorded live session (2h 22m) on the Oh My Claude Code plugin, streamed on the Korea Institute of Science and Technology channel — a Korean-language deep dive from the community that built it.
11-minute walkthrough that shows OMC's agents fanning out from a single prompt — watch the orchestration happen instead of reading about it.
Plugin anatomy, creation workflow, marketplace distribution, organization rollout, versioning, and security review — the packaging format OMC itself ships as.
Folder-based, on-demand instruction packs explained against hooks and subagents, with install steps and a production deployment example.
Skills, plugins, and MCP servers solve different problems — a clear map of when to reach for each without bloating your setup.
Enabling agent teams, choosing display modes, and practical patterns like parallel code reviews and competing-hypothesis debugging.
Monthly roundup for founders: what changed in Claude Code this month and how to ship faster, cut risk, and validate ideas with safer AI coding.
Korean walkthrough of installing OMC and running Team mode, with the common install failures explained and the terminal CLI vs in-session slash command split made clear.
Japanese primer on OMC's agents organized into build, review, specialist, and coordination lanes, and the model routing that picks Haiku, Sonnet, or Opus per task.
Two plugins, two philosophies: superpowers systematizes the process with skill-driven workflows, OMC maximizes agent autonomy through multi-AI orchestration — pick by what you value.
Japanese how-to covering installation, the operating modes, magic keywords, and the model routing that cuts costs 30-50% — aimed at readers going from beginner to pro.
Eighteen design plugins and skills evaluated across five mechanism types — tokens, aesthetic anchors, review gates — plus a copy-paste skill file to start with.
The three extensibility layers side by side — hooks for lifecycle automation, subagents for parallel delegation, skills for reusable expertise — and when to reach for each.
Full reference, changelog, release history, and the v4 → v5 migration table.
Everything you need to know before installing.
autopilot and ralph are optional shortcuts for power users.
/autopilot "..."). Execute — carries an approved task through implement → verify → fix loops until it's truly done; the v5 replacement for ultrawork, ultrapilot, swarm, and pipeline (/execute "..."). Team — N coordinated agents on a shared task list (/team 3:executor "..."). Ralph — relentless, won't quit until architect-verified. Or just describe the work — OMC picks the right mode.
security-reviewer agent, mandatory Architect verification before completion, and verification-before-done protocols. Ralph mode never marks a task complete until it actually works.
Join 39,000+ developers who turned Claude Code into a coordinated team of specialists. Install in 30 seconds. Free forever.