Yeachan-Heo/oh-my-claudecode

English | 한국어 | 中文 | 日本語 | Español | Tiếng Việt | Português

For Codex users: Check out oh-my-codex — the same orchestration experience for OpenAI Codex CLI.

Liked OmC but found it a bit overkill? Try gajae-code. Keeps Claude OAuth as-is while being faster, cheaper, simpler, and more powerful — with an SDK-based integration path built for OpenClaw, Hermes, Grokbot, and similar agent runtimes.

Multi-agent orchestration for Claude Code. Zero learning curve.

Don't learn Claude Code. Just use OMC.

Get Started • Documentation • CLI Reference • Workflows • Migration Guide • Discord

Step 1: Install

Marketplace/plugin install (recommended for most Claude Code users). These are Claude Code slash commands — enter them one at a time (pasting both lines at once will fail):

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

Then:

/plugin install oh-my-claudecode

If you prefer the npm CLI/runtime path instead of the marketplace flow:

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

Known npm warning: npm may print deprecated prebuild-install@7.1.3 during the CLI install. This currently comes from the upstream better-sqlite3 native-addon dependency (better-sqlite3 -> prebuild-install); prebuild-install@7.1.3 is still the latest published version, so there is no safe repo-side dependency bump or override to remove the warning yet. The warning is tracked in #2913 and does not by itself mean the OMC CLI install failed.

Step 2: Setup

# Inside a Claude Code / OMC session
/omc-setup

# From your terminal
omc setup

If you run OMC via omc --plugin-dir <path> or claude --plugin-dir <path>, add --plugin-dir-mode to omc setup (or export OMC_PLUGIN_ROOT before running it) so the installer doesn't duplicate skills/agents that the plugin already provides at runtime. See the Plugin directory flags section in REFERENCE.md for a complete decision matrix and all available flags.

Step 3: Build something

# Inside a Claude Code / OMC session
/autopilot "build a REST API for managing tasks"

# Natural-language in-session shortcut
autopilot: build a REST API for managing tasks

Select a configured stage profile only through /autopilot --workflow <name> <task>:

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

Profiles are configured under autopilot.workflows in .claude/omc.jsonc (project) or ~/.config/claude-omc/config.jsonc (user). A v1 profile contains only version: 1 and stages:

{
  "autopilot": {
    "workflows": {
      "plan-build-qa": {
        "version": 1,
        "stages": ["ralplan", "execution", "qa"]
      }
    }
  }
}

The admitted sequences are [ralplan, execution], [ralplan, execution, ralph], [ralplan, execution, qa], and [ralplan, execution, ralph, qa]. A project profile of the same name wholly replaces the user profile; different names coexist. Environment variables cannot define profiles. Profiles remain within autopilot's existing state, cancel, resume, Stop, and HUD lifecycle; legacy invocations without --workflow remain compatible.

Named profiles currently require Linux with the flock utility because their transcript evidence boundary uses Linux no-follow file-descriptor traversal and their recoverable mutation lock uses kernel advisory locking. Unsupported environments reject explicit --workflow invocation before creating or changing autopilot state; legacy autopilot remains available.

V1 intentionally excludes model fields or routing (stageModels), inline execution, dynamic commands/modes/state, arbitrary stages or plugins, and the separate custom-skill frontmatter parser mismatch. See Named Autopilot Stage Profiles ADR and Reference.

That's it. Everything else is automatic.

OMC exposes two different surfaces:

• Terminal CLI commands: run omc ... from your shell after installing the npm/runtime path (npm i -g oh-my-claude-sisyphus@latest) or from a local checkout.
• In-session skills: run /... inside a Claude Code session after installing the plugin/setup flow.

• VS Code / IDE extension: OMC does not ship a VS Code extension and does not document extension-specific install or automation flows. Use the Claude Code plugin or terminal CLI surfaces above; IDE integrations are only an optional way to access Claude Code itself.
• Agent SDK / programmatic usage: the npm package exports TypeScript helpers such as createOmcSession() and prompt expansion utilities for local Node.js programs using @anthropic-ai/claude-agent-sdk. This is a library surface, not a replacement for the Claude Code plugin UI.
• CI/CD and headless automation: prefer deterministic terminal commands (omc setup, omc ask, omc session search, repository scripts such as npm run sync-metadata:verify) and set ANTHROPIC_API_KEY or provider-specific CLI auth in the runner environment. Do not rely on interactive slash commands (/autopilot, /ralph, /execute, /team) in CI; they require an active Claude Code session.

If you're uncertain about requirements, have a vague idea, or want to micromanage the design:

/deep-interview "I want to build a task management app"

The deep interview uses Socratic questioning to clarify your thinking before any code is written. It exposes hidden assumptions and measures clarity across weighted dimensions, ensuring you know exactly what to build before execution begins.

Starting in v4.1.7, Team is the canonical orchestration surface in OMC. The legacy swarm keyword/skill has been removed; use team directly.

/team 3:executor "fix all TypeScript errors"

Use /team ... when you want Claude Code's in-session native team workflow. Use omc team ... when you want terminal-launched tmux CLI workers (claude / codex / gemini panes).

Team runs as a staged pipeline:

team-plan → team-prd → team-exec → team-verify → team-fix (loop)

Enable Claude Code native teams in ~/.claude/settings.json:

{
  "env": {
    "CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS": "1"
  }
}

If teams are disabled, OMC will warn you and fall back to non-team execution where possible.

v4.4.0 removes the Codex/Gemini MCP servers (x, g providers). Use the CLI-first Team runtime (omc team ...) to spawn real tmux worker panes:

omc team 2:codex "review auth module for security issues"
omc team 2:gemini "redesign UI components for accessibility"
omc team 2:antigravity "redesign UI components for accessibility"
omc team 1:claude "implement the payment flow"
omc team 1:cursor "implement the payment flow"
omc team status auth-review
omc team shutdown auth-review

For mixed Codex + Antigravity work in one command, run /ask codex and /ask antigravity and have Claude synthesize the results (Gemini remains available as an enterprise/API-key fallback):

/ask codex "Review this PR — architecture"
/ask antigravity "Review this PR — UI components"

Workers spawn on-demand and die when their task completes — no idle resource usage. Requires the selected CLI (codex, gemini, agy (antigravity), grok, or cursor-agent) installed/authenticated and an active tmux session.

Autopilot can prefer Cursor executor workers during team execution via .claude/omc.jsonc:

{
  "autopilot": {
    "execution": "team",
    "team": { "agentTypes": ["cursor"] }
  }
}

This config makes the autopilot execution stage use omc team 1:cursor "..." or /team 1:cursor "..." for implementation work. Cursor also supports reviewer-style roles (critic, code-reviewer, security-reviewer, test-engineer): those workers emit the structured verdict file the team leader consumes to transition the task, and final approval stays a lead-session responsibility. Cursor requires an installed/authenticated cursor-agent.

Pin a Cursor model with the OMC_EXTERNAL_MODELS_DEFAULT_CURSOR_MODEL environment variable, or per role with team.roleRouting.<role>.model. externalModels.defaults.cursorModel applies to workers routed through team.roleRouting. Ids come from cursor-agent --list-models, for example cursor-grok-4.6-high or composer-2.5. Left unset, cursor-agent chooses its own model.

Native team worker worktrees are being added behind an opt-in/config gate. See Native Team Worktree Mode for the workspace contract, canonical state-root rules, dirty-worktree preservation policy, and verification checklist.

Note: Package naming — The project is branded as oh-my-claudecode (repo, plugin, commands), but the npm package is published as oh-my-claude-sisyphus. If you install or upgrade the CLI tools via npm/bun, use npm i -g oh-my-claude-sisyphus@latest; the package installs both oh-my-claudecode and the short omc command aliases.

If you installed OMC via npm, upgrade with the published package name:

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

Package naming note: the repo, plugin, and commands are branded oh-my-claudecode, but the published npm package name remains oh-my-claude-sisyphus. npm installs expose both oh-my-claudecode and omc; examples prefer omc for brevity.

If you installed OMC via the Claude Code marketplace/plugin flow, update with:

# 1. Update the marketplace clone
/plugin marketplace update omc

# 2. Re-run setup to refresh configuration
/omc-setup

If you are developing from a local checkout or git worktree, update the checkout first, then re-run setup from that worktree so the active runtime matches the code you are testing.

Note: If marketplace auto-update is not enabled, you must manually run /plugin marketplace update omc to sync the latest version before running setup.

If you experience issues after updating, clear the old plugin cache:

/omc-doctor

• Zero configuration required - Works out of the box with intelligent defaults
• Team-first orchestration - Team is the canonical multi-agent surface
• Natural language interface - No commands to memorize, just describe what you want
• Automatic parallelization - Complex tasks distributed across specialized agents
• Persistent execution - Won't give up until the job is verified complete
• Cost optimization - Smart model routing saves 30-50% on tokens
• Learn from experience - Automatically extracts and reuses problem-solving patterns
• Real-time visibility - HUD statusline shows what's happening under the hood

Multiple strategies for different use cases — from Team-backed orchestration to token-efficient refactoring. Learn more →

Use only one primary loop authority in a session. Claude Code /goal is useful for a native cross-turn completion condition, while Execute owns single-agent verified completion, Team owns parallel staged execution, and Verify owns repeated quality-gate cycling. Artifact-only Ultragoal is the safe fallback when you need durable goal artifacts and evidence without starting another loop.

For /goal behavior, rely on Claude Code/Anthropic sources: the Claude Code /goal docs and Anthropic Claude Code changelog. Do not claim the /goal evaluator independently runs commands or reads files; surface test output, diffs, and review evidence in the conversation before treating a goal as proven.

• 19 specialized agents (with tier variants) for architecture, research, design, testing, data analysis
• Smart model routing - Haiku for simple tasks, Opus for complex reasoning
• Automatic delegation - Right agent for the job, every time
• Model × Agent Compatibility Matrix - Which model to pair with each agent, with premium/balanced/budget presets

• Prompt triggers - ralph, ralplan; Team stays explicit via /team
• HUD statusline - Real-time orchestration metrics in your status bar
  • If you launch Claude Code directly with claude --plugin-dir <path> (bypassing the omc shim), export OMC_PLUGIN_ROOT=<path> in your shell so the HUD bundle resolves to the same checkout as the plugin loader. See the Plugin directory flags section in REFERENCE.md for details.
• Skill learning - Extract reusable patterns from your sessions
• Analytics & cost tracking - Understand token usage across all sessions

Want to contribute to OMC? See CONTRIBUTING.md for the full developer guide, including how to fork, set up a local checkout, link it as your active plugin, run tests, and submit PRs.

Learn once, reuse forever. OMC extracts hard-won debugging knowledge into portable skill files that auto-inject when relevant.

# .omc/skills/fix-proxy-crash.md
---
name: Fix Proxy Crash
description: aiohttp proxy crashes on ClientDisconnectedError
triggers: ["proxy", "aiohttp", "disconnected"]
source: extracted
---
Wrap handler at server.py:42 in try/except ClientDisconnectedError...

Manage skills: /skill list | add | remove | edit | search Skillify: /skillify extracts reusable patterns with strict quality gates Auto-inject: Matching skills load into context automatically — no manual recall needed

Project-scoped OMC-authored skills are stored in .omc/skills/ and are intended to be committed when you want them shared. During slash/skill execution OMC also reads Claude Code workspace skills from .claude/skills/ and compatibility skills from .agents/skills/, so existing workspace-local SKILL.md packages remain callable without copying them into user-global skills. If you create project-local skills inside a linked git worktree and do not commit them, they disappear when that worktree is removed.

OMC writes runtime state, session data, plans, logs, handoffs, research notes, and local artifacts under .omc/ by default. The repository .gitignore keeps that runtime data local with one intentional exception: .omc/skills/** remains committable for project-scoped skills you want to share with the team. Treat everything else under .omc/ as local operational state that may contain prompts, transcripts, or machine-specific paths.

For linked git worktrees, the default .omc/ directory lives inside that worktree, so deleting the worktree deletes its local OMC state. Set OMC_STATE_DIR if you want state to survive worktree deletion, or add a .omc-workspace marker when several independent repos should share one parent-level state root. See OMC state, gitignore, worktree, and workspace contract.

Outside a git repository, OMC uses one canonical safe state root at ~/.omc/ (or $OMC_STATE_DIR/non-git when centralized state is configured); it does not create a new state root for every cwd or write state into sensitive directories such as ~/.ssh, ~/Downloads, or descendants of the system temp root. Legacy cwd-local state is untouched until explicitly migrated with state_migrate_non_git. State MCP tools honor an explicit workingDirectory while retaining repository-boundary checks for git-backed sessions.

Full feature list →

When several independent git repos share a parent directory, drop a .omc-workspace marker at the parent so all sub-repos share one .omc/ state root:

cd /path/to/parent-dir-with-many-repos
echo '{"id":"my-workspace"}' > .omc-workspace
# Sessions inside any sub-repo now share /path/.omc/
# For parallel ultragoal runs:
cd repo-A && omc ultragoal create-goals --auto-plan-id --brief "..."
cd ../repo-B && omc ultragoal create-goals --auto-plan-id --brief "..."

See Multi-repo workspaces in REFERENCE.md for resolution order, OMC_STATE_DIR, and workspace identifier options.

These shortcuts run inside a Claude Code / OMC session, not as terminal CLI commands. For shell commands, use the omc ... forms shown above. Team mode is explicit: use /team ... in-session or omc team ... from your shell rather than expecting a bare team keyword trigger.

Notes:

• Parallel work uses Team or executor delegation: choose /team for coordinated lanes or delegate implementation tasks to executors; use Ralph when persistence until verified completion is the priority.
• swarm compatibility alias has been removed; migrate existing prompts to /team syntax.
• plan this / plan the keyword triggers were removed; use ralplan or explicit /oh-my-claudecode:plan.

Run local provider CLIs and save a markdown artifact under .omc/artifacts/ask/.

# Terminal CLI
omc ask claude "review this migration plan"
omc ask codex --prompt "identify architecture risks"
omc ask gemini --prompt "propose UI polish ideas"
omc ask antigravity --prompt "propose UI polish ideas"
omc ask grok --prompt "cross-check this code review"
omc ask cursor --prompt "apply this implementation plan"
omc ask muse --agent-prompt critic --prompt "review this plan"
omc ask claude --agent-prompt executor --prompt "draft implementation steps"

# Inside a Claude Code / OMC session
/ask claude "review this migration plan"
/ask codex "identify architecture risks"
/ask antigravity "propose UI polish ideas"
/ask cursor "apply this implementation plan"
/ask muse --agent-prompt critic "review this plan"

Canonical env vars:

• OMC_ASK_ADVISOR_SCRIPT
• OMC_ASK_ORIGINAL_TASK

Phase-1 aliases OMX_ASK_ADVISOR_SCRIPT and OMX_ASK_ORIGINAL_TASK are accepted with deprecation warnings.

omc autoresearch is now a hard-deprecated shim. The authoritative workflow is:

/deep-interview --autoresearch improve startup performance
/oh-my-claudecode:autoresearch

• deep-interview --autoresearch generates/sets up the mission and evaluator
• autoresearch runs the bounded, single-mission stateful loop
• each iteration records evaluation JSON plus markdown decision logs
• non-passing iterations continue
• strict stopping is controlled by an explicit max-runtime ceiling

Auto-resume Claude Code sessions when rate limits reset.

omc wait          # Check status, get guidance
omc wait --start  # Enable auto-resume daemon
omc wait --stop   # Disable daemon

Requires: tmux (for session detection)

Use the HUD for live observability and the current session/replay artifacts for post-session inspection:

• HUD preset: /oh-my-claudecode:hud setup then use a supported preset such as "omcHud": { "preset": "focused" }
• Session summaries: .omc/sessions/*.json
• Replay logs: .omc/state/agent-replay-*.jsonl
• Live HUD rendering: omc hud
• Local friction reports: omc session friction report --since 24h summarizes context-bloat and operator-friction signals from local session artifacts without printing raw prompts or tool output; add --json for automation.

Jev brings calibrated decision-making to OMC's orchestration. Enable it with environment variables:

export TYPESAFE_API_KEY="sk-..."              # Jev API key (required)
export OMC_JEV="all:active"                    # All judgment points active
export OMC_JEV_TIMEOUT_MS="2000"               # Per-call timeout (default: 2000ms)

Modes:

• shadow (default): Jev is queried but heuristics decide; answers logged for analysis
• active: Jev's answer overrides the heuristic; adds latency (bounded by OMC_JEV_TIMEOUT_MS)
• off: No Jev queries sent (zero egress)

Jev decides skill triggering, model routing, loop continuation, context pruning, task sizing, and completion verdicts. On timeout, HTTP error, or parse failure, active mode degrades to the heuristic without blocking. Defaults to zero egress: without configuration, nothing is sent to the API. See Jev Judgment Points in HOOKS.md for the full configuration reference and point registry.

You can configure who gets tagged when stop callbacks send session summaries.

# Set/replace tag list
omc config-stop-callback telegram --enable --token <bot_token> --chat <chat_id> --tag-list "@alice,bob"
omc config-stop-callback discord --enable --webhook <url> --tag-list "@here,123456789012345678,role:987654321098765432"
omc config-stop-callback slack --enable --webhook <url> --tag-list "<!here>,<@U1234567890>"

# Incremental updates
omc config-stop-callback telegram --add-tag charlie
omc config-stop-callback discord --remove-tag @here
omc config-stop-callback discord --clear-tags

Tag behavior:

• Telegram: alice becomes @alice
• Discord: supports @here, @everyone, numeric user IDs, and role:<id>
• Slack: supports <@MEMBER_ID>, <!channel>, <!here>, <!everyone>, <!subteam^GROUP_ID>
• file callbacks ignore tag options

Forward Claude Code session events to an OpenClaw gateway to enable automated responses and workflows via your OpenClaw agent.

Quick setup (recommended):

/oh-my-claudecode:configure-notifications
# → When prompted, type "openclaw" → choose "OpenClaw Gateway"

Manual setup: create ~/.claude/omc_config.openclaw.json:

{
  "enabled": true,
  "gateways": {
    "my-gateway": {
      "url": "https://your-gateway.example.com/wake",
      "headers": { "Authorization": "Bearer YOUR_TOKEN" },
      "method": "POST",
      "timeout": 10000
    }
  },
  "hooks": {
    "session-start": {
      "gateway": "my-gateway",
      "instruction": "Session started for {{projectName}}",
      "enabled": true
    },
    "stop": {
      "gateway": "my-gateway",
      "instruction": "Session stopping for {{projectName}}",
      "enabled": true
    }
  }
}

Environment variables:

Supported hook events (6 active in bridge.ts):

Reply channel environment variables:

See scripts/openclaw-gateway-demo.mjs for a reference gateway that relays OpenClaw payloads to a custom HTTPS automation endpoint.

• Full Reference - Complete feature documentation
• CLI Reference - All omc commands, flags, and tools
• Notifications Guide - Discord, Telegram, Slack, and webhook setup
• Recommended Workflows - Battle-tested skill chains for common tasks
• Release Notes - What's new in each version
• Website - Interactive guides and examples
• Migration Guide - Upgrade from v2.x
• Architecture - How it works under the hood
• Performance Monitoring - Agent tracking, debugging, and optimization
• Model × Agent Compatibility Matrix - Which model to pair with each agent (premium / balanced / budget presets)
• Security Guide - Enterprise deployment and hardening

• Claude Code CLI
• Claude Max/Pro subscription OR Anthropic API key

OMC features like omc team and rate-limit detection require tmux:

Windows users: psmux provides a native tmux binary for Windows with 76 tmux-compatible commands. No WSL required.

OMC can optionally orchestrate external AI providers for cross-validation and design consistency. These are not required — OMC works fully without them.

Migrating from Gemini CLI: Per Google's announcement, the Gemini CLI is being superseded by the Antigravity CLI (agy); see the official Antigravity docs. Use omc team N:antigravity and omc ask antigravity wherever you previously used gemini. Windows headless support for agy is unknown/untested — report issues upstream.

Cost: 3 Pro plans (Claude + Antigravity/Gemini + ChatGPT) cover everything for ~$60/month.

MIT

Top personal non-fork, non-archived repos from all-time OMC contributors (100+ GitHub stars).

• @Yeachan-Heo — oh-my-claudecode (⭐ 40k)
• @junhoyeo — tokscale (⭐ 5.7k)
• @psmux — psmux (⭐ 3.6k)
• @devswha — herdr-web-ui (⭐ 827)
• @MeroZemory — ida-multi-mcp (⭐ 451)
• @GeiserX — awesome-spain (⭐ 357)
• @BowTiedSwan — buildflow (⭐ 296)
• @J-Pster — Psters_AI_Workflow (⭐ 289)
• @alohays — awesome-visual-representation-learning-with-transformers (⭐ 272)
• @HaD0Yun — Doyunha-Gopeak (⭐ 267)
• @jcwleo — random-network-distillation-pytorch (⭐ 264)
• @changeroa — StyleGallery (⭐ 260)
• @HwangTaehyun — github-repository-contribution-stats (⭐ 241)
• @shaun0927 — openchrome (⭐ 238)
• @emgeee — mean-tutorial (⭐ 199)
• @anduinnn — HiFiNi-Auto-CheckIn (⭐ 173)
• @Znuff — consolas-powerline (⭐ 145)

If Oh-My-ClaudeCode helps your workflow, consider sponsoring:

• Keep development active
• Priority support for sponsors
• Influence roadmap & features
• Help maintain free & open source

• ⭐ Star the repo
• 🐛 Report bugs
• 💡 Suggest features
• 📝 Contribute code

OmC includes a geobench product spec for measuring LLM hit rate, MRR, share of voice, and citations.

• Spec: geobench/oh-my-claudecode.yaml
• Runbook: docs/geobench.md