{"id":50986704,"url":"https://github.com/Varalix-Digitech-Solutions/clone-team","last_synced_at":"2026-07-07T18:00:30.120Z","repository":{"id":362813626,"uuid":"1259165014","full_name":"Varalix-Digitech-Solutions/clone-team","owner":"Varalix-Digitech-Solutions","description":"🧬 Clone any website with a team of AI agents — pixel-perfect UI clone + reverse-engineered architecture docs. A Claude Code skill built on the new dynamic Workflow engine with an unskippable test gate. ⭐ Star it if it helps!","archived":false,"fork":false,"pushed_at":"2026-06-15T07:52:28.000Z","size":11440,"stargazers_count":11,"open_issues_count":2,"forks_count":2,"subscribers_count":1,"default_branch":"main","last_synced_at":"2026-06-15T09:23:33.125Z","etag":null,"topics":["agentic-ai","ai-agents","anthropic","automation","claude","claude-code","claude-skill","codegen","dynamic-workflow","frontend","multi-agent","opus","web-development","website-clone"],"latest_commit_sha":null,"homepage":"https://claude.com/claude-code","language":"JavaScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Varalix-Digitech-Solutions.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-06-04T08:49:13.000Z","updated_at":"2026-06-15T07:52:33.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/Varalix-Digitech-Solutions/clone-team","commit_stats":null,"previous_names":["varalix-digitech-solutions/clone-team"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/Varalix-Digitech-Solutions/clone-team","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Varalix-Digitech-Solutions%2Fclone-team","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Varalix-Digitech-Solutions%2Fclone-team/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Varalix-Digitech-Solutions%2Fclone-team/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Varalix-Digitech-Solutions%2Fclone-team/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Varalix-Digitech-Solutions","download_url":"https://codeload.github.com/Varalix-Digitech-Solutions/clone-team/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Varalix-Digitech-Solutions%2Fclone-team/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":35237753,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-07-07T02:00:07.222Z","response_time":90,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["agentic-ai","ai-agents","anthropic","automation","claude","claude-code","claude-skill","codegen","dynamic-workflow","frontend","multi-agent","opus","web-development","website-clone"],"created_at":"2026-06-19T20:00:32.723Z","updated_at":"2026-07-07T18:00:30.101Z","avatar_url":"https://github.com/Varalix-Digitech-Solutions.png","language":"JavaScript","funding_links":[],"categories":["Multi-Agent Systems"],"sub_categories":["Orchestration Platforms"],"readme":"\u003cdiv align=\"center\"\u003e\n\n# 🧬 clone-team\n\n### Clone any website with a coordinated team of AI agents — and get the architecture docs to rebuild it.\n\nA [Claude Code](https://claude.com/claude-code) **skill** that orchestrates a **Manager**, **Frontend Developer**, **Backend Architect**, and **Tester** to produce a **pixel-perfect UI clone** *and* a reverse-engineered **`ARCHITECTURE.md`** — in the stack *you* choose.\n\n[![Website](https://img.shields.io/badge/Website-clone--team.varalix.com-2563eb)](https://clone-team.varalix.com)\n[![Claude Code Skill](https://img.shields.io/badge/Claude%20Code-Skill-d97757)](https://claude.com/claude-code)\n[![Built on Dynamic Workflows](https://img.shields.io/badge/Built%20on-Dynamic%20Workflows-7c3aed)](#-the-engine-a-dynamic-workflow)\n[![License: MIT](https://img.shields.io/badge/License-MIT-green.svg)](./LICENSE)\n[![PRs Welcome](https://img.shields.io/badge/PRs-welcome-brightgreen.svg)](#-contributing)\n\n**🌐 Website, demos \u0026 docs → [clone-team.varalix.com](https://clone-team.varalix.com)**\n\n**If this saves you a week of pixel-pushing, drop a ⭐ — it genuinely helps.**\n\n\u003cbr/\u003e\n\n\u003cimg src=\"docs/media/wembi-compare.gif\" alt=\"wembi.ai — original (left) vs clone-team build (right), scroll-synced side by side\" width=\"100%\"\u003e\n\n\u003csub\u003e\u003cb\u003eOriginal (left)\u003c/b\u003e vs. \u003cb\u003eclone-team build (right)\u003c/b\u003e — \u003ca href=\"https://www.wembi.ai/\"\u003ewembi.ai\u003c/a\u003e, driven in lockstep so both stay at the same scroll point. A hard target: Lenis smooth-scroll + scroll-driven 3D + reveal animations. \u003ca href=\"case-study/wembi/\"\u003eFull case study →\u003c/a\u003e\u003c/sub\u003e\n\n\u003c/div\u003e\n\n---\n\n## Why clone-team?\n\nMost \"AI clone this site\" tools hand you a rough, one-shot approximation and call it done. clone-team is different in two ways:\n\n1. **It can't skip the quality gate.** The build/test loop is a *deterministic program*, not a model's good intentions. A **Tester** runs a full visual + behavioral regression every round, and a section isn't \"done\" until the Tester says so — because \"done\" is literally a loop condition in code (see [the engine](#-the-engine-a-dynamic-workflow)).\n2. **You get understanding, not just pixels.** A **Backend Architect** documents the target's routes, API/network surface, inferred data model, auth/session flow, and end-to-end user journeys — so you can build your *own* product on top, not just stare at a copy.\n\n## ✨ Two deliverables, every run\n\n| | What you get |\n|---|---|\n| 🎨 **An exact UI clone** | Pixel-perfect, behavior-accurate, in **your** stack (React, Next.js, Vue, Svelte, Angular, plain HTML/CSS…) — independent of the source site's stack. |\n| 📐 **`ARCHITECTURE.md`** | Routes, observed API/network surface, inferred data model \u0026 entities, auth/session flow, state \u0026 navigation, and step-by-step user journeys. |\n\n## ⚙️ The engine: a **dynamic Workflow**\n\n\u003e clone-team is built on **Claude Code's new [dynamic Workflow](https://claude.com/claude-code) engine** — and it's the whole reason the quality guarantee holds.\n\nA *dynamic Workflow* is a deterministic JavaScript program that spawns and coordinates a fleet of subagents — where the **loop count, fan-out width, and exit conditions are all computed at runtime** from real data, not hardcoded. clone-team uses it so the process can't drift:\n\n- **The Tester gate is unskippable.** Per page: `extract → spec → build → full-regression-test → fix → re-test …` runs as a `while` loop whose exit condition *is* the Tester's `OK`. No human, no model, and no \"looks fine under deadline pressure\" can bypass it.\n- **It adapts to the work.** One-page site → one builder. Nine-page site → nine, running concurrently. The Workflow reads the page list and fans out accordingly.\n- **It sizes the agent fleet to your PC.** Before launching, a capacity probe reads your machine's free memory and computes how many builder agents to run at once — many in parallel on a big box, a few at a time on a laptop — so it goes as fast as your hardware allows without ever OOM-crashing it.\n- **It feeds failure back as input.** A Tester `NG` isn't just a fail; its structured issue list becomes the *next* Developer round's fix list.\n- **It's pausable \u0026 resumable** — even across sessions and usage-limit cutoffs — because progress is reconciled against a durable `state.json` + on-disk artifacts.\n\n```\nuser ⇄ MANAGER (main thread)         requirements, creds, recon, foundation,\n        │                            final regression, pause/resume\n        │ launches a DYNAMIC WORKFLOW that ENFORCES the loop\n        ▼\n   per page, in parallel:   extract → spec → DEVELOPER builds →\n                            TESTER full regression → (NG) fix → re-test → OK ✅\n   in parallel:             BACKEND ARCHITECT writes ARCHITECTURE.md\n   then:                    assemble → final regression → fix\n```\n\nTwo gates protect quality — the **Tester** (inside the loop) and the **Manager** (final sign-off) — and nothing ships past both.\n\n## 🚀 Install\n\n**Recommended — install as a plugin** (one marketplace, one install, auto-updates). In Claude Code:\n\n```\n/plugin marketplace add Varalix-Digitech-Solutions/clone-team\n/plugin install clone-team@clone-team\n```\n\nThat's it — the skill and the `/clone-status`, `/clone-pause`, `/clone-resume`, `/clone-report`, and `/clone-update` commands are registered automatically. Plugin installs **auto-update** when a new version ships; run `/clone-update` anytime to check what you're on vs. the latest. (Restart Claude Code if the commands don't show up immediately.)\n\n**You don't install the companion skills yourself.** The first time you run a clone, clone-team **bootstraps its own toolchain** — it runs a bundled, idempotent installer that fetches the `agent-browser` CLI and the companion skills (`ui-pack` and friends, `ui-animation`) **project-local** (into `./.claude/skills`, so your global skills stay clean), skipping anything already present. Just install the plugin and ask it to clone a site; it sets itself up and then gets to work. See [Dependencies](#-dependencies).\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cb\u003eAlternative — manual skill install (no plugins)\u003c/b\u003e\u003c/summary\u003e\n\n```bash\ngit clone https://github.com/Varalix-Digitech-Solutions/clone-team.git\ncp -r clone-team/skills/clone-team ~/.claude/skills/clone-team\n# optional: the slash commands\ncp clone-team/commands/clone-*.md ~/.claude/commands/\n# bootstrap the companion skills + agent-browser CLI (idempotent)\nbash ~/.claude/skills/clone-team/scripts/install-deps.sh\n```\n\u003c/details\u003e\n\n**Requirements:** [Claude Code](https://claude.com/claude-code), Node.js + npm (for the durable state CLI and to install `agent-browser`), and a Chromium browser (for real-browser verification via `agent-browser`). Everything else installs automatically — see [Dependencies](#-dependencies).\n\n## 🧑‍💻 How to use it\n\nJust ask Claude Code to clone a site — natural language triggers the skill:\n\n```\nClone https://example.com — the marketing homepage — in Next.js,\nand document how the site is put together.\n```\n\nThe **Manager** then walks you through a short setup (you stay in control):\n\n- **Stack** — what to build the clone in (React / Next.js / Vue / Svelte / plain HTML…).\n- **Scope** — whole site, or specific pages.\n- **Login** — if the target needs auth, you provide credentials; they're stored in a **gitignored local file** and never printed, committed, or pasted into prompts.\n- **Model tier** — `max-fidelity` (Opus, default), `cost-optimized` (Sonnet + Haiku), or `ultra-cheap`.\n- **Autonomy** — fully autonomous, or **checkpoint at each Tester verdict** for sign-off.\n- **Backend depth** — `none` / `flows` / `inferred` / `deep`.\n\nThen it runs the dynamic Workflow in the background and reports back with the clone + `ARCHITECTURE.md`.\n\n### Pause, resume \u0026 recovery\n\nLong jobs need an off switch. State lives in a durable `.clone-team/state.json` plus the built artifacts on disk:\n\n| Command | What it does |\n|---|---|\n| `/clone-status` | See what's done / in-flight / pending / flagged |\n| `/clone-pause` | Stop cleanly — nothing approved is lost |\n| `/clone-resume` | Pick up exactly where it stopped — **even in a new session or after a usage-limit reset** — without redoing finished pages |\n\n## 📦 Dependencies\n\n**These install automatically** — clone-team ships `scripts/install-deps.sh`, an\nidempotent installer it runs on first use (and which you can run yourself any time:\n`bash scripts/install-deps.sh`, or `--check` for a dry run). By default it installs\nthe companion skills **project-local** — into `./.claude/skills` of the folder\nyou're cloning in — so it **never pollutes your global `~/.claude/skills`** (pass\n`--global` if you do want them installed globally). It skips whatever's already\npresent:\n\n- **`ui-pack`** — design/frontend skill **bundle** (the single entry point the agents load); **vendored with this plugin**, so it's always available. It in turn loads its constituents below.\n- **`clone-website`** — extraction + builder dispatch\n- **`ui-ux-pro-max`**, **`impeccable`**, **`emil-design-eng`** — design intelligence + polish\n- **`ui-animation`** — motion craft (transitions/keyframes/springs, easing, clip-path reveals, gestures, performance) for the two motion specialists\n- **`agent-browser`** — real-browser automation (npm CLI) for building *and* verifying; the one **hard** dependency (needs Node/npm on PATH)\n\nEvery piece degrades gracefully if absent except `agent-browser`, which is required for real-browser verification.\n\n## 🛡️ Responsible use\n\nOnly clone sites you're **authorized** to access and replicate, and respect each site's terms of service. Credentials stay in a gitignored local file — never committed, printed, or embedded in prompts. The Backend Architect **documents observed behavior** — it does not attack, fuzz, or exfiltrate. This is a tool for prototyping, design study, and migration of sites you own or have permission to work on.\n\n## 🤝 Contributing\n\nIssues and PRs welcome — better extraction scripts, new stack templates, sharper Tester heuristics, and capacity tuning are all great places to start. If clone-team helped you, a ⭐ is the easiest way to say thanks and helps others find it.\n\n## License\n\n[MIT](./LICENSE).\n\n## 🙏 Thanks\n\nBuilt on the shoulders of `ui-pack`, `clone-website`, `agent-browser`, `ui-ux-pro-max`, `impeccable`, `emil-design-eng`, and [`ui-animation`](https://awesomeskill.ai/skill/mblode-agent-skills-ui-animation) — and scaffolded with `skill-creator`. Grateful to every author; this skill wouldn't exist without yours. 🫶\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FVaralix-Digitech-Solutions%2Fclone-team","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FVaralix-Digitech-Solutions%2Fclone-team","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FVaralix-Digitech-Solutions%2Fclone-team/lists"}