The Laboratory: Environment Setup

What you will get in this chapter

• A repeatable environment setup workflow
• A minimum viable laboratory (MVS) and acceptance criteria
• Fast troubleshooting for common issues

One-sentence definition

Laboratory = a repeatable dev environment that lets the system run at any time.

Minimum viable laboratory (MVS)

Qualified signal: the Hello World script runs and prints output.

Core metrics (must track)

Definition (default):

• Time window: unless stated otherwise, use the last 7 days rolling.
• Data source: use one trusted source (GA4/GSC/platform console/logs) and keep it consistent.
• Scope: only the current product/channel, exclude self-tests and bots.

Why this chapter?

Many people read the theory, feel excited, then open their laptop and realize they do not know what to open.

This chapter solves that. We will set up everything so you can:

• Use AI to help write code (Cursor)
• Deploy a site in one click (Vercel)
• Store data (Supabase)
• Call AI APIs (OpenAI / DeepSeek)

Estimated time: 1-2 hours

Prerequisites

Before you start, make sure you have:

Step 1: Install Node.js

Node.js is the runtime for JavaScript/TypeScript. Most of our code needs it.

Download and install

1. Visit nodejs.org
2. Download the LTS version (more stable)
3. Install it

Verify installation

Open a terminal (Windows: PowerShell, Mac: Terminal), run:

node --version

If you see something like v20.10.0, it worked.

Install pnpm (recommended)

pnpm is a faster package manager than npm. Install it:

npm install -g pnpm

Verify:

pnpm --version

Step 2: Install Cursor (AI code editor)

Cursor is an editor with built-in AI. It can write, explain, and fix code for you.

Download and install

1. Visit cursor.com
2. Download the right version for your OS
3. Install and open it

First-time setup

1. Log in (GitHub login works)
2. Choose your keymap (if you used VS Code, choose VS Code)
3. Cursor free plan includes monthly AI calls, enough for learning

Your first AI chat

In Cursor, press Cmd + L (Mac) or Ctrl + L (Windows) to open the AI chat.

Try:

Write a Hello World program in Node.js

AI will generate code. Click "Apply" to insert it into the editor.

Step 3: Set up Vercel (deployment)

Vercel lets you deploy websites in one click, free for small projects.

Register

1. Visit vercel.com
2. Click "Sign Up"
3. Choose "Continue with GitHub"
4. Authorize Vercel to access your GitHub

Verify

You should see the Vercel dashboard. No further setup needed now.

Step 4: Set up Supabase (database)

Supabase is an open-source Firebase alternative providing Postgres, auth, storage, and more.

Register

1. Visit supabase.com
2. Click "Start your project"
3. Sign in with GitHub

Create your first project

1. Click "New Project"
2. Choose an organization (or create one)
3. Fill in:
   • Name: my-first-asgs (or any name)
   • Database Password: set a strong password (save it)
   • Region: choose the closest region
4. Click "Create new project"

After creation, open the project settings and find API:

• Save Project URL
• Save the anon public key

You will use these later.

Step 5: Get an AI API key

OpenAI API

1. Visit platform.openai.com
2. Sign in or register
3. Open the "API Keys" page
4. Click "Create new secret key"
5. Copy and save it (shown only once)

Note: OpenAI is paid, but new users often get free credits.

DeepSeek API (cheaper alternative)

If you want to save money, DeepSeek is a good alternative:

1. Visit platform.deepseek.com
2. Register
3. Get an API key

DeepSeek is about 1/10 the price of OpenAI, with solid quality.

Step 6: Create your first project

Now connect everything and create a real project.

Create a project folder

Open a terminal and run:

mkdir my-first-asgs
cd my-first-asgs
pnpm init

Create a config file

Create .env at the project root:

# .env - store your API keys
# Do not commit this file to Git!

OPENAI_API_KEY=sk-your-openai-key-here
DEEPSEEK_API_KEY=sk-your-deepseek-key-here
SUPABASE_URL=https://your-project.supabase.co
SUPABASE_ANON_KEY=your-anon-key-here

Create .gitignore

# .gitignore - tell Git to ignore these
.env
node_modules/
.DS_Store

Hello World script

Create hello.js:

// hello.js - your first ASGS script
const fs = require('fs');

// Current time
const now = new Date().toISOString();

// Content to write
const content = `
# My first ASGS log

Generated at: ${now}

This file is auto-generated.
If you can read this line, your environment setup worked!

## Next steps

1. Read Ch01: What is an AI self-growth system
2. Modify this script to do more
3. Start building your own self-growth system
`;

// Write file
fs.writeFileSync('output.md', content);

console.log('OK: file generated: output.md');
console.log('Done! Your environment setup is complete.');

Run it:

node hello.js

If you see the success message and output.md exists, you are ready.

Common commands cheat sheet

Common errors

Error 1: node command not found

Cause: Node.js is not in PATH.

Fix: Restart the terminal, or reinstall Node.js and check "Add to PATH".

Error 2: pnpm install fails

Fix: Use npm instead:

npm install

Error 3: API key invalid

Check:

1. No extra spaces in the key
2. Account has balance
3. Key is not expired

Acceptance checklist

Summary

Congrats! Your laboratory is ready. Now read Ch01 to understand what an AI self-growth system is, then start building your first system.