Skip to content
SkillModel-led677 tokens in use

Clone Site

Reverse-engineer a website and rebuild it section by section as a Next.js, Tailwind and shadcn/ui clone, with parallel builder agents and a visual QA pass.

3,862 stars on the repository MIT Licenseericosiu/ai-marketing-skills · clone-site
npx skills add ericosiu/ai-marketing-skills -s clone-site
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing Clone Site costs

28tokens idle
677tokens in use
7,089tokens, up to

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Up to adds the files its instructions link. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

shelldetectednetworkdetected

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Good
  • Frontmatter8/15
  • Structure25/25
  • Examples12/15
  • Output10/15
  • Size15/15
  • Resources7/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent
  • Runs shell commands−5

    does this on the machine, and the frontmatter does not say so

    SKILL.md, line 34
  • Reaches the network−10

    does this on the machine, and the frontmatter does not say so

    SKILL.md, line 35

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Good
  • Development activity88
  • Maintenance44
  • Community56
  • Maturity28
  • Momentum40

Clone Site takes a URL and rebuilds the page as a Next.js project. It drives Chrome through the Chrome MCP to screenshot the target at desktop, tablet and mobile widths, extracts fonts, colors and every asset, and records scroll, click and hover behavior before anything is built.

For each section it writes a specification file with exact computed CSS values, per-state content and responsive rules, then hands that spec to a builder agent working in its own git worktree. Sections are merged as they finish, the page is assembled, and a side-by-side visual comparison against the original closes the job.

What You Get

  • A Next.js site that mirrors the target, runnable with npm run dev
  • Downloaded images, videos, fonts and extracted SVG icon components under public and src
  • One specification file per component in docs/research/components
  • Design reference screenshots and behavior and page topology notes in docs/research and docs/design-references
  • A completion report with sections built, assets downloaded, build status and known gaps

Common Use Cases

  • Replicate a competitor or inspiration landing page as a starting point for your own
  • Rebuild a legacy marketing site on a modern Next.js stack
  • Turn a URL into a component library with exact spacing, type and color values
  • Recreate scroll-driven and hover-driven interactions you want to study

What Makes It Unique

Most cloning prompts screenshot a page and guess. This one forces extraction first: exact getComputedStyle values, every state of tabs and scroll triggers, the interaction model (click, scroll, hover or time) before building, and a spec file as the contract for every builder. It caps builder prompts at about 150 lines of spec so sections stay small, and it requires a passing build after each merge.

Under The Hood

It runs on your machine through your agent. It controls Chrome through Chrome MCP to load the target site, take screenshots and run JavaScript that reads computed styles. It then writes and runs a Node.js script that downloads the site’s images, videos and fonts into public, writes many files in your project (specs, notes, components), runs npm install and npm run build, and dispatches builder agents in git worktrees that are merged back into main. It fetches web pages and assets from the site you point it at, so use it only on sites you are allowed to copy.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers