outworked

How It Works

1. Hire agents — Give each one a name, role, personality, model, and pixel sprite
2. Describe a goal — Write what you want in plain English; the orchestrator breaks it into subtasks and routes them to the right agents automatically
3. Watch them work — Agents walk to their desks, open Claude Code sessions, edit files, run commands, and talk to each other — all visible in the office
4. Ship code — Review diffs in the built-in git panel, approve PRs, and merge without ever leaving the app

Why Outworked?

Most AI coding tools give you a chat box. Outworked gives you an office.

You don't just prompt one model — you build a team. A frontend agent, a backend agent, a reviewer. They work in parallel, talk to each other, and you can see all of it happening live in a pixel-art world. It's the most fun way to ship code with AI.

Features

• Pixel Office — A Phaser-powered world where your agents walk, sit at desks, and collaborate — in real time
• Build a Team — Give each agent a name, role, personality, model, and sprite. Mix Claude Opus for hard problems with Sonnet for fast ones
• Auto-Orchestration — Describe a goal; the router breaks it into tasks and assigns them to the right agents
• Multi-Agent Collaboration — Agents talk to each other via [ASK:AgentName] and a shared message bus
• Full Claude Code Power — Every agent gets full tool access (Bash, Edit, Read, etc.) with persistent sessions
• Live Chat — Markdown-rendered conversations with syntax-highlighted code blocks and diffs
• Built-in Git — View status, staged changes, branches, and create PRs without leaving the app
• File Browser — Live-updating workspace tree that syncs as agents edit files
• Skills System — Extend agents with plug-in SKILL.md files (GitHub, MCP servers, and more)
• Cost Dashboard — Track tokens and spend per agent, session, and day
• Background Mode — Minimize and let agents keep working. Tray icon keeps you updated
• Parallel Processing — Multiple agents tackle subtasks simultaneously
• Safe by Default — Approval gates, command allowlists, directory restrictions, timeouts, and full audit logging
• Desktop Notifications — Get pinged when tasks finish or agents need approval
• Original Soundtrack — Because every office needs a vibe

Supported Models

| Provider | Models | | --------- | ---------------------------------- | | Anthropic | Claude Opus 4.6, Claude Sonnet 4.6 | | Local | Any model via Claude Code |

Each agent can run a different model — pair a fast model for simple tasks with a powerful one for complex work.

Get Started in 2 Minutes

1. Install

Homebrew (recommended)

brew tap outworked/tap
brew install --cask outworked

Manual download

Grab the latest .dmg from the releases page and drag to Applications.

2. Prerequisites

• Claude Code installed and authenticated (your existing subscription covers usage — no separate API key needed)

3. Launch and Hire Your First Agent

On first launch, the onboarding modal walks you through picking a workspace and creating your first agent. You'll be watching pixel employees write code in under two minutes.

Build from Source

git clone https://github.com/outworked/outworked.git
cd outworked
npm install
npm run electron:dev

Scripts

| Command | Description | | ------------------------ | --------------------------------------------------------------------------- | | npm run dev | Start Vite dev server (browser only, no Electron) | | npm run electron:dev | Build and launch the full Electron app | | npm run electron:build | Package distributable (dmg/zip on macOS, exe on Windows, AppImage on Linux) |

Tech Stack

| Layer | Technology | | -------- | ------------------------------------ | | Desktop | Electron | | Frontend | React 19 + TypeScript + Tailwind CSS | | Build | Vite | | Graphics | Phaser 3 | | AI | Claude Code SDK |

Project Structure

src/
├── components/       # React UI (ChatWindow, OfficeCanvas, GitPanel, etc.)
├── lib/              # Core logic (AI, orchestration, terminal, storage, costs)
├── basic-skills/     # Bundled SKILL.md modules (github, whisper, etc.)
└── phaser/           # Phaser game scene and sprite logic

electron/
├── main.js           # Electron main process (IPC, shell, permissions)
├── preload.js        # Context bridge to renderer
└── sdk-bridge.js     # Claude Code SDK bridge

public/
└── music/            # Background music tracks

Skills

Outworked uses a SKILL.md format — markdown files with YAML frontmatter that define what an agent can do. Bundled skills include:

| Skill | Description | | ------------ | ----------------------------------------------------- | | github | GitHub API access via gh CLI (issues, PRs, CI runs) | | mcporter | MCP server support |

Create custom skills by writing a SKILL.md file and assigning it to any agent.

Safety Model

Outworked takes a defense-in-depth approach:

1. Approval gates — Explicit approval required before dangerous commands (deletes, installs, network changes)
2. Command allowlists — Permit or block specific shell commands per agent
3. Directory restrictions — Agents are confined to their workspace
4. Timeouts — Configurable limits on long-running tasks
5. Audit trail — Full logging of all agent actions
6. Plan-first execution — Agents produce a plan before making changes
7. Permissions dashboard — Review and manage all access grants in one place

Roadmap

• [x] Conversation persistence & session history
• [x] Streaming markdown output with syntax highlighting
• [x] Live file watcher & workspace sync
• [x] Git integration panel
• [x] MCP server support
• [x] Multi-agent collaboration
• [x] Cost & token dashboard
• [x] Desktop notifications
• [x] Background agent mode
• [x] Parallel task processing
• [ ] Agent templates
• [ ] Hooks configuration UI
• [ ] Customizable sprites & backgrounds
• [ ] More model integrations

The Soundtrack

Every office needs a vibe. Outworked ships with an original soundtrack featuring tracks like:

"Don't Ever Turn Me Off" · "She's Just a Coder Agent" · "LLM Love Lying Machine" · "Can Love Be Binary" · "Goodbye Humans" · "How Does Grass Feel" · "Are You Even Reading This"

Toggle music from the player in the bottom-left corner.

Contributing

1. Fork the repo
2. Create a feature branch (git checkout -b feature/my-feature)
3. Commit your changes (git commit -m 'Add my feature')
4. Push to the branch (git push origin feature/my-feature)
5. Open a Pull Request

Please open a discussion first for large changes so we can discuss the approach.

FAQ

Nope. Outworked runs on Claude Code, so your existing subscription covers everything. No extra keys or billing to set up.

Outworked runs locally on your machine. Code is processed through whichever model provider you select for each agent, with the same privacy guarantees as using those APIs directly.

Agents require explicit approval