Resources
Prompts, templates, and reference material you can use during and after the bootcamp. Copy anything you need.
Prompts Toolkit
These prompts are organized by the phase of building you're in. Copy them, paste them into Cursor or Claude, and customize the bracketed parts for your project. Use the Example tab to see each prompt filled in for a fictional product called HabitFlow (a daily habit tracker).
Ideation
Use this when you have a rough idea and want to sharpen it.
Planning
Use this to turn your refined idea into an implementation plan.
Scaffolding
Use this to set up your project from scratch.
UI/UX Design
Use this to get AI to help you design clean interfaces.
Debugging
Use this when something breaks.
Deployment
Use this when you're ready to go live.
Code Review / Refinement
Use this to improve code you've already written.
Project Kickstart Template
This is the master prompt. Give it a single paragraph describing your idea and it generates everything you need to start building. Copy it, fill in the bracket, and go.
Common Commands Cheat Sheet
You probably won't need these at all. Cursor and AI handle most of this for you automatically. But just in case, here's a quick reference to keep handy. Think of it like a glossary you can look up if something comes up.
Git
| Command | What it does |
|---|---|
git init | Start tracking a project with Git |
git add . | Stage all changes for commit |
git commit -m "message" | Save your changes with a description |
git push | Upload your changes to GitHub |
git pull | Download the latest changes |
git status | See what files have changed |
npm / npx
| Command | What it does |
|---|---|
npx create-next-app@latest | Create a new Next.js project |
npm install | Install all project dependencies |
npm run dev | Start the development server |
npm install [package] | Add a new library to your project |
Vercel CLI
| Command | What it does |
|---|---|
npm i -g vercel | Install the Vercel CLI globally |
vercel | Deploy your project |
vercel --prod | Deploy to production |
vercel env pull | Download environment variables locally |
Supabase CLI
| Command | What it does |
|---|---|
npx supabase init | Initialize Supabase in your project |
npx supabase start | Start a local Supabase instance |
npx supabase db push | Push local schema changes to production |
npx supabase migration new [name] | Create a new database migration |
Good vs. Bad Prompts
The difference between a useful AI response and a useless one usually comes down to how you ask. Here are some real examples.
Example 1: Building a feature
Make a login page
Build a login page using Next.js App Router and Supabase Auth. Include: - Email and password fields - A "Sign up" link below the form - Error messages for invalid credentials - Redirect to /dashboard on success - Use Tailwind CSS, minimal design - Mobile responsive
Why this works: The specific prompt tells AI exactly what tech to use, what the page should include, and how it should behave. The vague one leaves everything up to interpretation, so you'll spend more time fixing it.
Example 2: Fixing a bug
My app isn't working. Fix it.
I'm getting this error when I try to log in: "TypeError: Cannot read properties of undefined (reading 'email')" The error is in src/app/login/page.tsx on line 23. Here's the relevant code: [paste code] I expected the user object to have an email field after calling supabase.auth.signInWithPassword().
Why this works: You're giving the AI the exact error, where it happens, the code, and what you expected. It can now diagnose instead of guess.
Example 3: Designing a UI
Make it look better
This dashboard page feels cluttered. Improve the design by: - Adding more whitespace between sections - Using a clear visual hierarchy (larger headings, smaller labels) - Aligning the cards in a clean grid - Using a consistent color palette (grays + one accent color) - Making the primary action button more prominent Keep using Tailwind CSS. Show me the updated component.
Why this works: Instead of "make it look better" (which is subjective), you're telling AI exactly what "better" means to you: more whitespace, hierarchy, alignment, consistency, and prominence. It can execute on specifics.
The pattern
Good prompts share a pattern: context + specifics + constraints. Tell AI what you're working on, what you need, and what the boundaries are. The more specific you are, the less time you spend going back and forth.