MCP Desktop Tools

Provides desktop automation capabilities for Claude, including browser control, screenshots, and input actions via an MCP server.

typescript

0

GitHub Stars

typescript

Language

4 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "k1ta141k-mcp-desktop-tools": {
      "command": "node",
      "args": [
        "C:/Users/<you>/mcp-desktop-tools/dist/index.js"
      ]
    }
  }
}

MCP Desktop Tools provides a local MCP server that enables Claude to automate desktop tasks, control browsers, capture screenshots, simulate mouse and keyboard input, manage windows, and access the clipboard. This makes it possible to automate both browser-based workflows and native desktop actions from Claude.

How to use

You run a local MCP server and connect Claude Code to it as an MCP endpoint. The server exposes a set of tools that let you automate browser actions, take screenshots, and control desktop windows and input. Use these tools by sending requests through the MCP client in Claude Code, composing actions in a sequence to automate complex tasks.

How to install

Prerequisites you need before starting are Node.js 18 or later and a Windows environment if you plan to use native window, mouse, and keyboard operations.

Install dependencies and build the server, then install the required browser binaries.

npm install
npm run build
npx playwright install chromium

Claude Code configuration

Configure Claude Code to connect to the MCP server by adding a server entry that points to the local runtime. Use the following example configuration path and snippet.

{
  "mcpServers": {
    "desktop-tools": {
      "command": "node",
      "args": ["C:/Users/<you>/mcp-desktop-tools/dist/index.js"]
    }
  }
}

Available tools

browser_open

Launch Chromium and navigate to a URL.

browser_navigate

Navigate to a URL with configurable wait conditions.

browser_click

Click elements by CSS selector.

browser_type

Type into input fields, optionally clear or press Enter.

browser_read

Read page content (text, HTML, title, URL, or specific elements).

browser_screenshot

Capture viewport or full-page screenshots.

browser_close

Close the browser.

screenshot_fullscreen

Capture the entire screen across multiple monitors.

screenshot_region

Capture a rectangular region by coordinates.

screenshot_window

Capture a specific window by title (partial match).

desktop_mouse_click

Click at screen coordinates.

desktop_mouse_move

Move the cursor, either instantly or with smooth animation.

desktop_keyboard_type

Type text using simulated keystrokes.

desktop_keyboard_hotkey

Press keyboard shortcuts like ctrl+c or alt+tab.

desktop_window_list

List all visible windows with positions and sizes.

desktop_window_focus

Focus a window by its title.

desktop_window_resize

Move and resize a window.

desktop_app_launch

Launch applications by path, name, or URI.

desktop_clipboard_read

Read text from the clipboard.

desktop_clipboard_write

Write text to the clipboard.