Skip to content
Agentic Levels

Everything starts here.

GuestLocal progress only
PreferencesSign in
01Start with one taskBest first move for beginners.02Check your LevelMeasure where you are.03Score an AI resultFind the habit to practice first.04Return to Your WorkScores, links, and checkpoints.
Start here

Begin

HomeThe main entry point.New to AIStart with one useful task.
Know where you are

Measure

Check your LevelUse this after you have tried AI.Fluency ScoreScore an AI result you can review.
Build the habit

Learn

LevelsLessonsTracks
Find the reference

Library

PromptsReferenceResourcesCompare Tools
Turn it into work

Apply

Your Next MoveChoose what AI should change next.Tool SetupGet the tools ready.
Come back later

Return

Your WorkScores, links, and checkpoints.My PathContinue from your level.Updates
Site

Site

PricingAboutFAQ & FeedbackPreferences

© 2026 Fuentes Studio

Privacy·Terms
yourCouncil
Ready to help
✦

What do you want to understand?

Ask anything about what you're learning.

← Resources

How to use Oh My ClaudeCode (OMC)

Multi-agent orchestration framework for Claude Code: 19 specialized agents, intelligent Haiku/Opus model routing, and an autonomous plan-to-ship pipeline with 30-50% token cost reduction.

Running every Claude Code task through the same general session means paying Opus prices for tasks Haiku handles fine, losing context on long builds, and manually managing multi-step workflows. OMC installs 19 specialized agents with intelligent model routing. Routing simple tasks to Haiku and complex architectural work to Opus. Its plan→design→execute→verify→fix pipeline runs autonomously with persistent loops and auto-retry. Autopilot mode handles well-defined tasks end-to-end without staged coordination. The result is 30-50% token cost reduction plus consistent quality gates on every phase.

L4L5L6L7
Before you startClaude Code must be installed. OMC v4.7.7+ is required. Earlier versions shipped modes (swarm, ultrapilot, ecomode) that are now removed. If you see references to those modes in any guide, that guide is outdated.

Steps

01

Install OMC and confirm the version

Install via the Claude plugin marketplace. After install, restart Claude Code and check the version. OMC v4.7.7 removed swarm, ultrapilot, ultrawork, pipeline, and ecomode. The current primitives are ralph (iterative), team (parallel), and ccg (split-model).

Install and verify
claude plugin install oh-my-claudecode # Restart Claude Code, then: /omc --version
02

Use ralph for iterative PRD-driven development

ralph works through a PRD story by story, pausing at each for verification before advancing. It is the primary agent for sustained feature development. Between iterations, ralph polls the peer task queue so high-priority work can interrupt without manual intervention.

Start a ralph session
/ralph [PRD_FILE_PATH] # ralph reads the PRD, identifies the first incomplete story, works it, verifies, then advances.
[PRD_FILE_PATH]Path to your PRD file e.g. .planning/feature-name/prd.json
03

Use team for 3+ genuinely independent parallel workstreams

team spins up multiple tmux-based Claude Code sessions with explicitly defined file ownership per workstream. File ownership prevents conflicts. A session that does not own a file cannot write to it. Use team when you have three or more independent tasks (different files, different concerns) each taking 30+ minutes.

Start a team session
/team # OMC asks you to define each workstream: task, files owned, expected output. # It spins up one tmux pane per workstream.
04

Use ccg for backend + frontend splits

ccg routes backend work to Codex and frontend work to Gemini. Both run on their own token budgets, not your Anthropic quota. Use ccg when a feature requires API changes and UI changes that can happen in parallel and you have a clear contract between them before starting.

Start a ccg session
/ccg # Define the backend task (Codex) and frontend task (Gemini) separately. # ccg coordinates the handoff point.
05

Wire code-review as a phase gate

OMC's code-review agent fires automatically after any phase marked complete. It checks for logic errors, security issues, empty event handlers, and unresolved TODOs in a consistent format. A phase with an unresolved finding should not advance. Invoke it manually any time with the command below.

Manual code review
/omc:code-review [FILE_OR_DIFF]
[FILE_OR_DIFF]File path or 'diff' for the current git diff e.g. src/app/api/auth/route.ts

How do you know it worked?

  • `/omc --version` confirms v4.7.7 or higher.
  • ralph worked through at least one PRD story end-to-end without manual intervention.
  • team ran at least two tmux workstreams with defined file ownership and no conflicts.
  • code-review produced at least one structured finding on a real file.
  • No deprecated modes (swarm, ultrapilot, ecomode) appear in the installed version.

If something looks off

Using team for fewer than 3 independent workstreams
team has real overhead: tmux setup, file ownership definitions, coordination. For 1-2 tasks, run them sequentially with gsd:quick. The coordination cost only pays off at 3+ genuinely independent tasks each taking 30+ minutes.
Running ccg when the API contract is not yet defined
ccg works when you can define the backend/frontend contract before either starts. If the contract is still being figured out, run the backend first, lock in the API shape, then start the frontend.
ralph looping 3+ iterations without progress
If ralph runs three iterations on the same story without completing it, stop and replan. A stuck ralph loop means the story is under-specified or blocked by something the PRD did not anticipate.
References to deprecated OMC modes in old instructions
swarm, ultrapilot, ultrawork, pipeline, ecomode. All removed in v4.7.7. Any guide referencing these modes is outdated. The current primitives are ralph, team, and ccg.

Skills & Commands Reference

OMC uses magic keywords typed inline. No slash-command prefix needed. Every keyword triggers a different orchestration strategy with different model routing. 19 specialized agents auto-activate; pick by task shape.

Command / SkillWhat it doesWhen to use it
teambasicCanonical staged pipeline: plan → PRD → execute → verify → fix across coordinated Claude agentsAny feature that needs design, implementation, and verification, this is the default
omc team N:codex/gemini/claudebasicSpawn N real tmux CLI workers (actual Codex or Gemini processes in split panes)When you need Codex for security/architecture review or Gemini for large-context UI work alongside Claude
ccgbasicTri-model advisor: runs ask-codex + ask-gemini then Claude synthesizes both perspectivesMixed backend+UI decisions where you want multiple model opinions before acting
ralphbasicPersistent execution with automatic verify/fix loops, won't stop until the task is confirmed completeTasks that must finish fully, no silent partials. Use for any critical feature or bug fix
autopilotbasicSingle lead agent, fully autonomous end-to-end execution, minimal ceremonyWell-defined tasks where you don't need staged coordination, faster than team for simple features
ulwadvancedMaximum parallelism, burst parallel fixes/refactors without Team overheadMany independent small fixes (lint errors, type errors, rename operations) that can run concurrently
ralplanadvancedIterative planning consensus, multiple planning passes until a high-confidence plan emergesAmbiguous or high-stakes tasks where you need the plan to be solid before any implementation starts
deep-interviewbasicSocratic requirements clarification, Claude asks structured questions to extract real requirementsVague ideas or underspecified requests before starting any implementation
deepsearchbasicRoutes to codebase-focused search, finds files, symbols, patterns across the repoYou know something exists in the codebase but don't know where it is
ultrathinkadvancedDeep extended reasoning mode for architectural and design problemsComplex architecture decisions, hard trade-offs, or when previous approaches have failed
omc ask claude/codex/geminiadvancedRun local provider CLIs and save a markdown artifact under .omc/artifacts/ask/Asynchronous second opinions or analysis you want saved for later reference
omc waitbasicAuto-resume daemon that waits for rate limit reset and continues your sessionWhen you hit rate limits mid-session and want Claude to pick up automatically
omc cost daily/weekly/monthlybasicUsage and cost reports across all OMC sessionsMonitoring token spend or understanding which workflows cost the most
/omc-setupbasicInstall or refresh all OMC configuration, agents, and skillsFirst install, after updates, or when OMC behavior seems off
/omc-doctorbasicDiagnose OMC configuration issues and report what's wrongSomething isn't working as expected, run this before filing a bug report

See also

  • Get Shit Done: gsd:debug and gsd:quick pair with OMC's ralph and team
  • Superpowers: writing-plans produces the PRDs ralph consumes

Was this helpful?

← Back to ResourcesView on GitHub