Getting Started
Get up and running with DevOS in under 2 minutes. This guide covers installation, configuration, and your first AI-powered project.
Installation
DevOS requires Node.js 18 or later. Install it globally via npm:
npm install -g devos-ai
Verify the installation:
devos --version
Quick Start (30 Seconds)
# 1. Set up an AI provider (choose one)
devos config set OPENAI_API_KEY=sk-your-key-here
# OR use Ollama for free local models:
devos model add ollama
# 2. Create a project
devos init "Express API with user authentication"
# 3. Run it (with auto-debugging)
devos run
# 4. Deploy it
devos deploy
Configuration
DevOS stores configuration in your home directory. Manage it with:
# Set a value
devos config set KEY=value
# View all config
devos config list
# Reset to defaults
devos config reset
API Keys
Configure AI providers by setting their API keys:
# OpenAI
devos config set OPENAI_API_KEY=sk-...
# Anthropic
devos config set ANTHROPIC_API_KEY=sk-ant-...
# Google (Gemini)
devos config set GOOGLE_API_KEY=AIza...
# OpenRouter (25+ models)
devos config set OPENROUTER_API_KEY=sk-or-...
# Ollama (local, no key needed)
devos model add ollama
Model Selection
# Use a specific model
devos model use gpt-4o
# Let DevOS choose the best model per task
devos model auto
# See all available models
devos model list
Your First Project
Let's build a blog with React and Tailwind CSS:
# Create the project
devos init "React blog with Tailwind CSS, dark mode, and markdown support"
# The AI will:
# ✓ Create folder structure
# ✓ Generate package.json
# ✓ Write React components
# ✓ Set up Tailwind CSS
# ✓ Install dependencies
Or use Goal Mode for the full autopilot experience:
# This single command will plan, code, test, debug, AND deploy
devos goal "Build a React blog with auth, markdown, and deploy to Vercel"
AI Provider Setup
OpenAI (Recommended for beginners)
- Get an API key from platform.openai.com
- Run:
devos config set OPENAI_API_KEY=sk-your-key - Select model:
devos model use gpt-4o
Ollama (Free, Local)
- Install Ollama from ollama.com
- Pull a model:
ollama pull llama3.3 - Enable in DevOS:
devos model add ollama - Select model:
devos model use llama3.3
💡 Pro Tip
Enable Smart Router to automatically select the best model for each task:
devos model auto
Environment Variables
You can also set API keys as environment variables instead of using devos config:
export OPENAI_API_KEY=sk-your-key
export ANTHROPIC_API_KEY=sk-ant-your-key
export GOOGLE_API_KEY=AIza-your-key