Quick start · Screenshots · How it works · Build · Report an issue

Website Cloner turns a website into a folder you can browse, inspect, and keep. Capture browser-rendered pages and their assets with CloakBrowser, preview the saved copy over local HTTP, and open the files in your own editor.

✨ What you can do

• 🌐 Capture rendered pages. Save HTML plus captured CSS, scripts, images, and fonts through a Chromium browser context.
• 🎚️ Control the crawl. Set the link depth and page limit, from a single page to a multi-page capture.
• 🔌 Configure a proxy. Choose HTTP, HTTPS, SOCKS4, or SOCKS5 with separate host, port, username, and password fields.
• 📁 Keep editable files. Choose an output folder and access the saved source directly.
• 🖥️ Preview over HTTP. Open the saved entry page in your default browser using a local preview server.
• 🕘 Revisit captures. Desktop history lists runs newest first, with timestamps, preview, open-folder, and delete actions.
• 📸 See the result. Follow progress and view the screenshot associated with a completed capture.

📸 The full desktop interface

A real example.com capture with Website Cloner — from capture settings to the finished result and download history.

View the example.com page screenshot saved by the app

🚀 Quick start

Install Node.js 20 or newer and npm, then run:

git clone https://github.com/kalimeraa/website-cloner.git
cd website-cloner
npm ci
npm start

npm run dev launches the same desktop app. The capture browser is separate from Electron; CloakBrowser downloads its Chromium runtime when needed, so allow time and an internet connection on the first capture.

For a quick first run, enter https://example.com, set Depth to 0 and Max pages to 1, then select Start cloning.

Before capturing, Website Cloner displays Checking connection… and checks the target through your selected connection. If the target cannot be reached, it checks example.com and microsoft.com to distinguish a website problem from an unavailable connection. With a proxy selected, all checks go through that proxy; there is no direct fallback.

Connection errors appear in English, including No internet connection, Unable to reach this website, Proxy authentication failed, and The proxy connection timed out. Fix the connection or proxy settings and select Start cloning again. Proxy passwords are not included in these messages. Connectivity checks are best-effort: a firewall blocking all checks can also appear as an unavailable connection. Saved local previews do not require this check.

🧭 How it works

1. Enter a URL. Use an http:// or https:// address. Add a proxy if needed.
2. Choose the scope. Depth 0 captures the starting page; 1 follows its links; 2 also follows links from those pages. Max pages caps the crawl. Assets are captured independently of link depth.
3. Capture. CloakBrowser loads each page, lets it render, and collects available resources. Saved references are rewritten to local paths.
4. Preview and use the files. Select Open preview to browse the local copy or Open folder to inspect its source.

Each capture is a snapshot of browser-visible content. Server-side application code, databases, and live backend services are not part of a website download; features that depend on them may still need the original service. A successfully loaded page is not proof of bypassing an anti-bot challenge.

🔌 Proxy settings

Select a protocol, then fill in the host and port. Username and password are optional. The UI assembles a URL such as:

http://username:password@proxy.example.com:8080
socks5://username:password@proxy.example.com:1080

Use No proxy for a direct connection. Credentials are encoded when the proxy URL is assembled. Avoid sharing your local job-history files: they can contain the proxy configuration used for a run.

📂 What gets saved

The default destination groups captures by domain and timestamp:

Documents/Website Cloner/
└── example.com/
    └── <timestamp>/
        ├── example.com/
        │   └── index.html
        ├── <asset-host>/       # When external assets are captured
        ├── screenshot.png     # Best-effort page screenshot
        └── mirror-manifest.json

The manifest records the entry page and capture details. Open folder gives you ordinary files to edit, copy, or archive. Keep the app running while using its local HTTP previews.

🌍 HTTP mode

An experimental browser entry point is also included:

npm run serve

It listens at http://127.0.0.1:4173. The desktop app is the main interface for persistent history and native folder actions.

📦 Build a desktop package

npm run build

Electron Builder writes artifacts to dist/. Native build targets are configured for macOS (DMG), Windows (NSIS), and Linux (AppImage/deb). Build on the matching operating system. Signing and notarization need your own certificates; a local package build does not create a signed release automatically.

🛠️ Under the hood

Electron provides the desktop shell, CloakBrowser + Playwright drive browser capture, and Node.js handles local storage and HTTP previews.

desktop/          Electron main process, bridge, UI, and local services
vendor/           Browser adapter and capture engine
build/            Application icon and packaging assets
docs/             Product artwork and real app screenshots
test/             Node.js tests
serve.js          Experimental HTTP entry point

Run the tests with npm test. For an end-to-end desktop check, run node test/desktop.smoke.cjs with a graphical desktop and internet access: it tests offline/proxy errors and then captures example.com using an isolated temporary profile and output folder. Contributions are welcome: open an issue with reproduction steps, or send a focused pull request describing the change and how you tested it.

💜 Open source

Website Cloner is released under the MIT license. Use, modify, distribute, and sell the software while preserving its license notice. The software license does not relicense content captured from other websites.

Built with Electron, CloakBrowser, Playwright, proxy-chain, and JSZip.