website-cloner

Clone websites into React components using orchestrated screenshot, extraction, implementation, and QA sub-agents.

Updated Jul 27, 2026
One-click install
npx skills add https://github.com/imMamdouhaboammar/Mimera --skill website-cloner-immamdouhaboammar
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: website-cloner
Source: https://github.com/imMamdouhaboammar/Mimera/tree/main/.agents/skills/website-cloner
Command: npx skills add https://github.com/imMamdouhaboammar/Mimera --skill website-cloner-immamdouhaboammar

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires motion, @anthropic-ai/mcp-playwright, and includes references (resource) and assets (resource) components.

What problem does it solve? Recreating an existing website's design by hand requires tedious manual inspection of colors, fonts, spacing, and assets, followed by error-prone visual comparison. This Skill automates the entire process through a multi-agent workflow that captures screenshots, extracts styles and assets, implements the clone, and iterates on QA feedback. ## Core Features & Use Cases - Orchestrated Multi-Agent Workflow: A /clone-website <url> slash command coordinates four specialized sub-agents (screenshotter, extractor, cloner, qa-reviewer) through a five-phase pipeline. - Pixel-Level QA Iteration Loop: A QA reviewer compares the clone against the original at desktop, tablet, and mobile viewports, classifies issues as Critical/Major/Minor, and loops fixes until PERFECT or a 5-iteration cap. - Framework-Aware Output: Auto-detects Next.js, TanStack Start, or Vite and emits a single React component styled with Tailwind CSS arbitrary values and motion animations, with assets organized in public/. - Use Case: Point the command at a marketing landing page you admire; the workflow downloads its images and icons, extracts its exact color palette and typography, and produces a working /clone route in your project. ## Quick Start Ask the agent to run /clone-website followed by the target URL, for example: clone the website at https://example.com into my project.

Frequently Asked Questions about website-cloner

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

FAQPage Schema
How do I clone a website into a React component?

Run the /clone-website slash command with the target URL. The orchestrator captures screenshots, extracts assets and styles into context.md, implements a single React component with Tailwind CSS and motion, then iterates with a QA reviewer until the clone matches.

What sub-agents does the website cloning workflow use?

It uses four sub-agents: website-screenshotter captures full-page and section screenshots at three viewports, website-extractor downloads assets and extracts styles, website-cloner implements the React component, and website-qa-reviewer compares the clone against the original.

Does the cloner support Next.js, Vite, and TanStack Start?

Yes, the cloner agent detects the framework from package.json and outputs to the correct location: app/clone/page.tsx for Next.js App Router, pages/clone.tsx for Pages Router, src/routes/clone.tsx for TanStack Start, and src/pages/Clone.tsx for Vite.

Why use Tailwind CSS and motion instead of plain CSS or framer-motion?

Tailwind arbitrary values like bg-[#1a2b3c] enable exact color and spacing matching without config changes. The motion package is the lighter, modern successor to framer-motion and is imported from motion/react for animations.

What happens if the clone never reaches a perfect QA score?

The workflow caps iteration at five rounds to prevent infinite loops. If the status is still NEEDS_WORK after that, it stops and reports the remaining issues from review-notes.md along with the current functional output.

Why does the cloning workflow fail on some websites?

Sites requiring authentication or protected by bot detection cannot be captured automatically, and the workflow stops with a warning. Playwright MCP must also be configured correctly, or browser operations will fail.