Hands-On Guide
Real tools. Real repositories. Real results.
Each chapter is an exercise, not an explanation. You'll set up a real development environment, submit a pull request to a live repository, pair-program a web app with an AI agent, deploy it to a real server, pentest it with autonomous AI security agents, build automated pipelines, and ship your own SaaS product — all from a terminal, all without prior coding experience.
What's inside
The Hands-On Guide is the practical companion to the main engineering book. Where the main book explains the ideas, this one puts them into practice.
You don't need to be a programmer. You need to be willing to type commands into a terminal and see what happens. That's how you learn this — not by reading, but by doing.
Terminal, package manager, Git, GitHub CLI, and an AI coding assistant — from scratch, on any OS.
Fork a real repository, write a review, commit and push, and submit a pull request that goes live on GitHub.
Fork the web-presenter project and use an AI agent with a Hugging Face API key to generate custom illustrations, spoken narration, and animated backgrounds — all from prompts.
Build a complete Travel Bucket List web app from scratch — without writing a single line of code yourself. You describe, the agent builds, you iterate.
Deploy your app to a Hetzner VPS — the agent SSHes in, configures the server, and calls the Cloudflare API for DNS. Real self-hosting, agent-operated.
Attack your own app before anyone else does. Donna orchestrates AI agents through recon, vulnerability analysis, exploitation, and reporting — autonomous security scanning.
Give the agent eyes. Write end-to-end tests with Playwright, experience TDD with an AI agent, and set up GitHub Actions CI — so no change ships unchecked.
Build an automated pipeline that reads unstructured input, triages it with AI agents, creates GitHub issues, and generates reports — all running on a cron schedule while you sleep.
The capstone. Build a complete micro-SaaS from scratch — auth, payments, AI features, CI/CD, landing page — and ship it to a live URL you can share with anyone.
Write your own CLAUDE.md, reflect on the journey, and take the helm. You walked in as a passenger — you leave as the captain of your own ship.
Contents
Exercise highlights
Every step uses actual terminal commands — git, gh, python3. No web interfaces, no drag-and-drop.
You'll fork schlunsen/theagenticcrew and schlunsen/web-presenter — real projects with real histories.
Use FLUX.1-schnell via the Hugging Face API to generate ten custom images for your presentation — from plain English descriptions.
Turn your slide text into spoken audio using text-to-speech models on Hugging Face — no recording required.
Ask an AI agent to change the Three.js network animation to match your presentation's colour palette — in seconds.
Build a Travel Bucket List app from scratch — describe what you want, review what the agent builds, iterate until it's right. Full CRUD with localStorage.
Your agent SSHes into a Hetzner VPS, configures Caddy with auto-HTTPS, and calls the Cloudflare API for DNS — real self-hosting, agent-operated.
Donna orchestrates five parallel agents to scan for SQL injection, XSS, auth bypasses, and more — then produces a professional security report with CVSS scores.
Describe behaviours in plain English, let the agent write Playwright tests, then experience the red-green-refactor loop — breaking things on purpose and watching the agent fix them.
Chain multiple AI agents into a shell script pipeline. Unstructured input goes in, categorised GitHub issues and summary reports come out — running on a schedule while you sleep.
Build a complete micro-SaaS from scratch — auth, Stripe payments, AI features, CI/CD, and a landing page. Deployed to a live URL you can text to a friend.
Every exercise ends with a commit and push. Your work lives on GitHub — shareable, versioned, real.
Free download
Pick your operating system. Available in English and Spanish.
The lab companion to the engineering and crew books.
The Agentic Crew