Setting Up Your Environment

Every cloud tool needs an API key, and every local tool needs a model runner. Set these up once.

1. Get API keys

• Anthropic (Claude) — console.anthropic.com → API Keys.
• OpenAI (Codex/GPT) — platform.openai.com → API Keys.
• Google (Gemini) — aistudio.google.com → Get API key.

Treat keys like passwords: store them in a .env file that is git-ignored, never in source code.

# .env
ANTHROPIC_API_KEY=sk-ant-...
OPENAI_API_KEY=sk-...
GEMINI_API_KEY=AIza...

2. Install a modern terminal

• macOS/Linux — your built-in terminal is fine.
• Windows — install Windows Terminal + WSL2 for the best experience, or use Git Bash.

3. Check your toolchain

Most agents work best with a clean, version-controlled project:

git --version
node --version   # for JavaScript/Python MCP servers
python3 --version

4. Install a model runner (optional, for local)

# Ollama — the easiest way to run local models
curl -fsSL https://ollama.com/install.sh | sh
ollama run llama3.1

You're now ready to follow any platform lesson in this course.