recreate

Recreate UI from screenshots using multi-model extraction and visual verification.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/primovera12/Master-Kit --skill recreate
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: recreate
Source: https://github.com/primovera12/Master-Kit/tree/main/Master-Starter-Kit/tools/skills/recreate
Command: npx skills add https://github.com/primovera12/Master-Kit --skill recreate

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Recreate UI from screenshots or Dribbble images automatically by detecting multiple screens, performing triple extraction (Gemini + Super Design + Claude Vision), cross-validating results, generating production-ready code, and verifying visually with Playwright.

Core Features & Use Cases

  • Detects and separates multiple screens within artistic showcases (e.g., device mockups with 2-4 screens)
  • Triple extraction: Gemini, Super Design, and Claude Vision for robust design understanding
  • Cross-validation, then code generation and visual verification
  • Generates frontend code and previews suitable for integration into existing projects
  • Use cases: when a user asks to recreate a design from a screenshot or drops a UI image

Quick Start

Provide an image or Dribbble URL and run the design-lab workflow to produce a pixel-accurate recreation.

Frequently Asked Questions about recreate

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I convert a screenshot into UI code?▼

To convert a screenshot into UI code, provide an image or Dribbble URL to trigger a multi-model extraction pipeline that generates production-ready frontend code. This process uses triple extraction with Gemini, Super Design, and Claude Vision to ensure pixel-accurate recreations.

What is the best way to recreate a UI design from a Dribbble image?▼

Recreating a UI design from a Dribbble image is best handled by a cross-validation pipeline that detects multiple screens, performs triple model extraction, and generates code. It visually verifies the output with Playwright to ensure the final recreation matches the original design.

Can I separate multiple screens from a single design showcase screenshot?▼

Yes, you can separate multiple screens from a single design showcase screenshot. The extraction pipeline automatically detects and isolates individual screens within artistic showcases, such as device mockups containing two to four separate UI screens, before generating code.

Does visual verification with Playwright guarantee pixel-accurate UI code generation?▼

Visual verification with Playwright ensures pixel-accurate UI code generation by cross-validating the extracted design elements and visually confirming the rendered output. This step guarantees the generated frontend code matches the original screenshot before integration.

Do I need to provide a design brief to generate frontend code from a screenshot?▼

No, you do not need to provide a design brief to generate frontend code from a screenshot. You can simply drop a UI image or provide a Dribbble URL, and the workflow automatically extracts the design to produce a pixel-accurate recreation suitable for integration.