Clone live websites into local runnable Vite projects with assets and design tokens.
npx skills add https://github.com/BEIRUX/agent-skills --skill site-replicator
Please help me install this Agent Skill. Skill: site-replicator Source: https://github.com/BEIRUX/agent-skills/tree/main/site-replicator Command: npx skills add https://github.com/BEIRUX/agent-skills --skill site-replicator
Reverse-engineer any live website into a local, runnable Vite project by fetching HTML, assets, and design tokens, then reconstructing CSS/JS with assets ready for development.
Provide a URL, and I will generate a runnable local Vite project that mirrors the site.
High-intent search queries and answers about installing and using this skill.
To clone a live website into a local Vite project, the tool fetches HTML, assets, and design tokens using web_fetch and curl, then reconstructs the files to produce a ready-to-run project via npm run dev.
Yes, you can mirror a public site for offline testing without browser automation by relying on Node.js parsing and cheerio to extract content and download assets directly from the HTML structure.
The best way to extract design tokens and assets from an existing website is to fetch the HTML structure, parse it to identify linked resources, and download the CSS, fonts, and images to reconstruct the site locally.
Site replication with cheerio and Node.js works for static public sites by fetching HTML directly. It does not use browser automation, so dynamically rendered content loaded by JavaScript may not be captured.
Limitations of cloning a live site into a Vite project include missing dynamically rendered content, potential issues with absolute paths, and the inability to replicate server-side logic or private authentication states.
AgentComputerAIFetch GitHub HTML pages and parse repository metadata with cheerio.
elbwalkerGuide walkerOS contributors through npm-based build, test, and lint workflows.
ceslepConvert Stitch project design assets into a structured DESIGN.md file.
eddmannRender dynamic web pages with Playwright to capture HTML, text, and screenshots.
