OpenClawFactory

A multi-agent AI factory floor powered by Open Interpreter, with a real-time Next.js Mission Control dashboard, WebSocket pipeline updates, and an optional Telegram bot interface.

  ██████╗ ██████╗ ███████╗███╗   ██╗ ██████╗██╗      █████╗ ██╗    ██╗
  ██╔══██╗██╔══██╗██╔════╝████╗  ██║██╔════╝██║     ██╔══██╗██║    ██║
  ██║   ██║██████╔╝█████╗  ██╔██╗ ██║██║     ██║     ███████║██║ █╗ ██║
  ██║   ██║██╔═══╝ ██╔══╝  ██║╚██╗██║██║     ██║     ██╔══██║██║███╗██║
  ╚██████╔╝██║     ███████╗██║ ╚████║╚██████╗███████╗██║  ██║╚███╔███╔╝
   ╚═════╝ ╚═╝     ╚══════╝╚═╝  ╚═══╝ ╚═════╝╚══════╝╚═╝  ╚═╝ ╚══╝╚══╝

What it does

• Multi-agent factory: Four named AI agents (Henry, Charlie, Sarah, Max) each backed by Open Interpreter, moving through pipeline stages: Break Room → Build → QA → Review → Ship
• Mission Control dashboard: Live Next.js UI showing agents as pixel-art sprites, animated between pipeline stages via WebSocket
• Telegram bot: Send tasks and check status from anywhere — /assign, /status, /tasks
• PM2 + Nginx: Production-grade process management on a DigitalOcean Droplet (or any Ubuntu server)

Architecture

Browser / Phone
      │
   Nginx :80
    ├── /        → Next.js frontend  :3000
    ├── /api/*   → FastAPI backend   :8000
    └── /ws      → WebSocket         :8000
           │
    AgentManager
    ├── Henry  (Open Interpreter)
    ├── Charlie (Open Interpreter)
    ├── Sarah  (Open Interpreter)
    └── Max    (Open Interpreter)
           │
    SQLite (data/factory.db)
           │
    TelegramBot (optional)

Quick start (DigitalOcean)

One-time deploy from your local machine:

git clone https://github.com/westland/OpenClawFactory
cd OpenClawFactory
bash deploy.sh root@YOUR_DROPLET_IP

The installer will prompt for:

• Anthropic API key (required)
• OpenAI API key (optional)
• Dashboard password
• Telegram bot token + chat ID (optional)

Access:

• Dashboard: http://YOUR_DROPLET_IP
• API: http://YOUR_DROPLET_IP/api/agents

Local development

# Backend
cd backend
python -m venv venv && source venv/bin/activate   # or venv\Scripts\activate on Windows
pip install -r requirements.txt
cp ../.env.example ../.env  # fill in your keys
uvicorn main:app --reload --port 8000

# Frontend (separate terminal)
cd frontend
npm install
npm run dev   # http://localhost:3000

Configuration

Edit .env to change agent names, roles, and models:

AGENT_1_NAME=Henry
AGENT_1_ROLE=Researcher
AGENT_1_MODEL=claude-sonnet-4-6

AGENT_2_NAME=Charlie
AGENT_2_ROLE=Builder
AGENT_2_MODEL=claude-sonnet-4-6

Add up to 8 agents by continuing the AGENT_N_* pattern.

PM2 commands (on server)

pm2 status          # view all processes
pm2 logs            # tail all logs
pm2 restart all     # restart everything
pm2 stop all        # stop everything

Tech stack

Layer	Technology
AI agents	Open Interpreter + Claude
Backend	FastAPI + uvicorn
Frontend	Next.js 14, React, Tailwind CSS, Framer Motion
Real-time	WebSocket
Database	SQLite
Process mgr	PM2
Proxy	Nginx
Notifications	python-telegram-bot

License

MIT