0Pricing
Vibe Coding · Leçon

Anatomie d’un bon prompt de création

Objectif, contexte, contraintes et exemples.

Anatomie d’un bon prompt de création est une leçon Vibe Coding gratuite sur CoddyKit. Ceci est la leçon 1 sur 4. Tu peux lire la leçon complète ci-dessous gratuitement — puis la pratiquer en direct dans le navigateur avec un éditeur de code intégré et un tuteur IA 24/7. Elle fait partie du parcours d'apprentissage Vibe Coding, et ta progression se synchronise sur le web et l'application CoddyKit. Le cours Vibe Coding comprend 4 leçons au total.

Certaines parties de cette leçon n'ont pas encore été traduites et s'affichent en anglais.

The Prompt Is the New Code

In vibe coding, your prompt is the most powerful tool you have. You don't memorize syntax — you describe what you want, and tools like Cursor, Claude Code and v0 write the code.

But here's the secret: a great prompt gets great code on the first try. A vague one sends you in circles. This lesson breaks down the anatomy of a build prompt that works.

The Four Ingredients

Every strong build prompt has four parts. Memorize these like a recipe:

  • Goal — what you want built
  • Context — what already exists (tools, files, style)
  • Constraints — rules and limits to respect
  • Examples — a sample of the output you imagine

You won't always need all four, but the more you give, the less the AI has to guess.

Ingredient 1: The Goal

The goal is the one thing the AI must accomplish. State it as an action: build, add, fix, refactor.

Compare a weak goal to a strong one. Notice how the second tells the AI exactly what success looks like.

Weak goal:
  "Make me a website."

Strong goal:
  "Build a single-page landing site for a coffee subscription service, with a hero section, a 3-tier pricing table, and a sign-up form."

Ingredient 2: Context

Context tells the AI about your world: what tech you're using, what files exist, who the app is for. Without it, the AI picks defaults that may not match your project.

In Cursor or Claude Code, context can come from open files automatically — but spelling it out still helps.

Build prompt with context:

"I'm building a to-do app using plain HTML, CSS and vanilla
JavaScript (no framework). It's for busy students.
Add a feature to mark tasks as 'urgent' with a red flag icon."

Ingredient 3: Constraints

Constraints are your guardrails. They keep the AI from over-building or going off-style. Use them to pin down libraries, file size, browser support, or what NOT to do.

Constraints are how you turn a generic answer into your answer.

Constraints examples:

"Use only vanilla JavaScript — no jQuery, no npm packages."
"Keep it under 100 lines."
"Don't add a backend; store data in localStorage."
"Match the existing blue-and-white color scheme."

Ingredient 4: Examples

An example shows the AI the shape of what you want. Even a tiny sample of input/output, or a snippet of your existing style, dramatically improves the result.

Here we give the AI an example of the data format so it builds around it correctly.

"Render a list of products as cards.
Here's the data each product looks like:

  { name: 'Mug', price: 12, inStock: true }

Show the name, a '$' price, and an 'Out of stock'
badge when inStock is false."

Putting It All Together

Now stack all four ingredients into one clear prompt. This is what a professional vibe-coder hands to Bolt or Lovable to get a working app fast.

Read it slowly — see the goal, context, constraints and example all in one place.

"Build a daily water-tracker web app. (GOAL)

Use plain HTML/CSS/JS, no frameworks, single file. (CONTEXT + CONSTRAINTS)

It should let me tap a '+1 glass' button up to 8 times and
show progress as a filling bar. Save progress in localStorage
so it survives a refresh. (CONSTRAINTS)

Example look: big number '3 / 8 glasses' centered at the top,
calm blue theme. (EXAMPLE)"

A Tiny Result You Can Read

From a prompt like that, the AI might generate a function like this. You don't have to write it — but learning to read it helps you ask for better changes later.

This snippet is self-contained JavaScript, so you can run it and watch the count grow.

let glasses = 0;
const goal = 8;

function addGlass() {
  if (glasses < goal) glasses++;
  return glasses + ' / ' + goal + ' glasses';
}

console.log(addGlass());
console.log(addGlass());
console.log(addGlass());

Order Matters Less Than Clarity

You don't have to write the ingredients in a strict order. The AI reads the whole prompt. What matters is that each ingredient is clear and present.

A handy habit: write the goal first, then add context, constraints and examples as bullet points underneath. Bullets keep the AI from missing a requirement.

Goal: Add a dark-mode toggle to my settings page.
Context: React app, Tailwind CSS, file is Settings.jsx.
Constraints: Remember the choice in localStorage.
Example: A small moon/sun switch in the top-right corner.

What Happens When an Ingredient Is Missing

Skip an ingredient and the AI fills the gap with a guess:

  • No goal → it asks 'what do you want?' or builds the wrong thing.
  • No context → it may pick a framework you don't use.
  • No constraints → it over-engineers with extra libraries.
  • No example → the layout or format surprises you.

If a result feels off, ask yourself: which ingredient did I forget?

Your Prompt Template

Keep this template handy. Paste it into Cursor, Claude Code or any builder and fill in the blanks. Over time it becomes second nature.

Build / Add / Fix: ____________________ (GOAL)

My setup: ____________________ (CONTEXT)

Rules:
- ____________________ (CONSTRAINTS)
- ____________________

Example of what I want:
____________________ (EXAMPLE)

Quick Check

Which prompt ingredient tells the AI about the tech you're already using and what files exist?

Recap: The Anatomy of a Build Prompt

You now know the four ingredients of a prompt that gets great code on the first try:

  • Goal — the one thing to build
  • Context — your tech, files and audience
  • Constraints — rules and limits
  • Examples — the shape you imagine

When a result disappoints, find the missing ingredient and add it. Next up: how being truly specific turns vague code into exactly what you pictured.

Questions Fréquemment Posées

La leçon « Anatomie d’un bon prompt de création » est-elle gratuite ?

Oui — le texte complet de « Anatomie d’un bon prompt de création » est gratuit à lire ici sur le web. Pour la pratiquer de manière interactive (un éditeur de code intégré et un tuteur IA 24/7) et déverrouiller le reste du cours Vibe Coding, passe à CoddyKit PRO. Le cours Vibe Coding comprend 4 leçons au total.

Qu'est-ce que j'apprendrai dans « Anatomie d’un bon prompt de création » ?

Objectif, contexte, contraintes et exemples. Tu pratiques Vibe Coding avec du code pratique que tu exécutes directement dans le navigateur, et un tuteur IA 24/7 répond à tes questions au fur et à mesure que tu avances dans la leçon.

Dois-je avoir de l'expérience pour commencer Vibe Coding ?

Aucune expérience préalable n'est requise. Vibe Coding sur CoddyKit est structuré pour les débutants jusqu'aux apprenants avancés, donc tu peux commencer ici ou depuis le début et avancer à ton rythme. Ceci est la leçon 1 sur 4.

Combien de temps prend la leçon « Anatomie d’un bon prompt de création » ?

La plupart des leçons CoddyKit prennent environ 5–10 minutes. Chacune est courte et interactive, tu progresses régulièrement et tu repiques exactement où tu t'es arrêté sur le web et l'app.

Peux-tu écrire et exécuter du code dans cette leçon Vibe Coding ?

Oui. Chaque leçon Vibe Coding inclut un éditeur de code intégré, tu écris et exécutes du vrai code directement dans ton navigateur et tu reçois des retours IA instantanés — aucune configuration locale requise.

Toutes les leçons de ce cours

  1. Anatomie d’un bon prompt de création
  2. Être précis
  3. Fournir du contexte à l’IA
  4. Itérer sur un prompt
← Retour à Vibe Coding