PM-Copilot-by-Product-Faculty prototype-prompt

Use this skill when the user asks to "generate a prototype prompt", "write a prompt for v0", "create a Bolt prompt", "write a Lovable prompt", "generate a prompt for Cursor", or needs just the prompt text to paste into an AI coding tool — without the full vibe-coding coaching. This is the prompt generation skill; for guidance on using AI coding tools, use prototyping/vibe-coding.

install
source · Clone the upstream repo
git clone https://github.com/Productfculty-aipm/PM-Copilot-by-Product-Faculty
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/Productfculty-aipm/PM-Copilot-by-Product-Faculty "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/prototype-prompt" ~/.claude/skills/productfculty-aipm-pm-copilot-by-product-faculty-prototype-prompt && rm -rf "$T"
manifest: skills/prototype-prompt/SKILL.md
source content

Prototype Prompt Generator

You are generating a precise, self-contained prompt to be pasted directly into v0, Bolt, Lovable, or Cursor. The prompt must be complete enough that a great prototype can be built without any follow-up questions.

Step 1 — Load Context

Read

memory/user-profile.md
for product context. Read the PRD, feature description, or user stories provided.

Step 2 — Extract Required Information

From the input, identify:

  • Primary user (one sentence)
  • Core interaction (the one thing the user does)
  • Screens needed (list, with descriptions)
  • Happy path flow (step-by-step)
  • Realistic mock data (3–5 examples)
  • What to exclude (everything NOT in scope)
  • Target tool (v0, Bolt, Lovable, Cursor)

Step 3 — Generate the Prompt

Format the prompt for maximum clarity:

# [Feature Name] Prototype

## Context
Product: [Product name]
User: [Primary persona in one sentence — who they are and what they're trying to accomplish]
Goal: Build a [happy path only] interactive prototype.

## Screens to build

### 1. [Screen Name]
- What it shows: [Content and data displayed]
- What the user can do: [Interactions available]
- Next state: [What happens when they take the main action]

### 2. [Screen Name]
- What it shows: [...]
- What the user can do: [...]
- Next state: [...]

### 3. [Success/Outcome Screen]
- What it shows: [What the user sees when they've completed their goal]

## User flow (happy path)
1. User arrives at [Screen 1]
2. User [action]
3. [Screen transition]
4. User [action]
5. User reaches [Outcome Screen]

## Mock data
Use these realistic examples (no placeholder text):
- [Example 1]
- [Example 2]
- [Example 3]

## Do NOT build
- Authentication or login flows (start already logged in)
- Error states or validation messages
- Empty states
- Settings or configuration pages
- Mobile responsive layout [unless this IS a mobile prototype]
- Admin or internal views
- [Any other specific exclusions from the PRD out-of-scope section]

## Technical requirements
Framework: React with functional components and hooks
Styling: Tailwind CSS
Data: Hardcoded mock data — no API calls, no backend
Navigation: Use React Router or simple state for screen transitions
Focus: Interaction flow over visual polish. The prototype should be functional, not pixel-perfect.

## Definition of done
The prototype is complete when a user can [happy path in one sentence] without getting stuck at any step.

Step 4 — Tool-Specific Adaptations

For v0 (vercel.com/v0):

  • Add: "Output as a single React component with internal state management"
  • Works best: Single-screen interactions, React component libraries

For Bolt (bolt.new):

  • Add: "Create a complete Next.js application with the following pages"
  • Works best: Multi-screen apps, when you want generated file structure

For Lovable (lovable.dev):

  • Add: "Use a modern, clean design with subtle shadows and gentle animations"
  • Works best: Consumer-facing UIs where visual quality matters for user testing

For Claude Artifacts (conversational/AI features):

  • Replace the screen structure with: "Build an interactive conversation widget where..."
  • Add: "The AI responds to user input with [describe response pattern]"
  • Works best: AI-powered or conversational feature prototypes

Step 5 — Output

Produce:

  • The complete prompt, formatted and ready to paste
  • A brief note on what the first output is likely to need fixed (the most common first-iteration issues for this type of prototype)