pskill9/website-downloader

Free

使用 wget 下载整个网站的 MCP 服务器。保留网站结构并将链接转换为本地工作。

FreeFree tier

Inputs: url

Type

Open Source

Visit Website GitHub (40 stars)

About pskill9/website-downloader

Website Downloader MCP Server is an open-source tool that provides a Model Context Protocol (MCP) server for downloading entire websites using wget. It preserves the original website structure, converts links to work locally, and includes all page requisites such as CSS and images. The server offers a 'download_website' tool with configurable parameters including URL (required), output path (optional, defaults to current directory), and recursion depth (optional, defaults to infinite). It restricts downloads to the same domain to avoid external content. Built with JavaScript and requires wget to be installed on the system. Installation is simple via npm and the server can be integrated into MCP-compatible applications.

Key Features

Recursive downloading with configurable depth (default infinite)

Includes all page requisites such as CSS, images, and scripts

Converts links to work locally for offline browsing

Restricts downloads to the same domain to avoid external content

Preserves original website directory structure

Adds appropriate file extensions automatically

Simple MCP tool integration with parameters: url, outputPath, depth

Pros & Cons

Pros

Open source and free to use
Leverages wget, a mature and reliable command-line tool
Preserves website structure and converts links for full offline functionality
Includes all page requisites automatically
Configurable recursion depth gives control over download scope
Restricts to same domain by default, reducing unwanted external content
Easy to install via npm and integrate with MCP-compatible environments

Cons

Requires wget to be installed separately on the system
Not a standalone binary; depends on Node.js runtime and npm
No graphical user interface – command line only
May not handle modern JavaScript-heavy single-page applications or dynamic content that requires a browser render
Limited to wget's capabilities; no built-in support for authentication, cookies, or complex site interactions

Best For

Creating offline copies of documentation or reference websitesArchiving websites for personal or research purposesMirroring static websites for local development or testingDownloading website content for offline reading without internet accessBacking up websites preserving full structure and assets

FAQ

What is the Website Downloader MCP Server?

It is an open-source MCP (Model Context Protocol) server that provides a tool to download entire websites using wget. It preserves the website structure and converts links to work locally.

What are the prerequisites for using this tool?

The server requires wget to be installed on your system. You can install wget via package managers like Homebrew (macOS), apt (Linux), or Chocolatey (Windows), or download the binary from eternallybored.org.

How do I install and use the server?

Clone the repository, run 'npm install' and 'npm run build' to build the server, then add it to your MCP settings by specifying the path to the built index.js file.

What parameters does the download_website tool accept?

It accepts three parameters: 'url' (required) – the website URL, 'outputPath' (optional, defaults to current directory) – where to save the downloaded site, and 'depth' (optional, defaults to infinite) – maximum recursion depth (0 for single page, 1 for direct links, etc.).

Does the downloader include images, CSS, and other assets?

Yes, it downloads all page requisites including CSS, images, and other files, and converts links to work locally.