kwin-mcp

Overview

kwin-mcp is a comprehensive Model Context Protocol server that enables AI agents to interact with Linux desktop applications running on KDE Plasma 6 Wayland. It supports both isolated virtual sessions that don't affect the user's desktop and live session connections for collaborative workflows. The server provides 30 MCP tools covering mouse, keyboard, touch, clipboard, accessibility tree inspection, screenshot capture, and window management, making it suitable for end-to-end GUI testing and desktop automation.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• session_startapp_command? str, screen_width? int (1920), screen_height? int (1080), enable_clipboard? bool (false), keep_screenshots? bool (false), isolate_home? bool (false), keep_home? bool (false), env? dict

  Start an isolated KWin Wayland session, optionally launching an app

  Try:Start a KWin session with FirefoxTry:Launch kcalc in an isolated session with screen size 1280x720
• session_connectdbus_address? str, wayland_display? str, keep_screenshots? bool (false)

  Connect to an existing KWin session (real desktop or container)

  Try:Connect to my existing KDE Plasma desktopTry:Connect to the KWin session in systemd-nspawn container
• screenshotinclude_cursor? bool (false)

  Capture a screenshot of the virtual display (saved as PNG)

  Try:Take a screenshot of the current desktopTry:Capture screenshot with cursor included
• accessibility_treeapp_name? str, max_depth? int (15), role? str

  Get the AT-SPI2 widget tree with roles, names, states, and coordinates

  Try:Get the accessibility tree for FirefoxTry:Show all buttons in the current application
• find_ui_elementsquery str, app_name? str, states? list[str]

  Search for UI elements by name, role, or description with optional state filtering

  Try:Find the calculate button in kcalcTry:Search for all visible text input fields
• mouse_clickx int, y int, button? str ('left'), double? bool, triple? bool, modifiers? list[str], hold_ms? int (0), screenshot_after_ms? list[int]

  Click at coordinates with various options for button, modifiers, and click behavior

  Try:Click at position 100,200Try:Double-click the open button with Ctrl key held
• keyboard_typetext str, screenshot_after_ms? list[int]

  Type a string of text character by character (US QWERTY layout)

  Try:Type 'Hello World'Try:Enter the username 'admin' in the login field
• launch_appcommand str, env? dict

  Launch an application inside the running session

  Try:Launch Firefox browserTry:Start the calculator application
• list_windows

  List all accessible application windows with titles and focused state

  Try:Show all open windowsTry:List applications with their window titles
• wait_for_elementquery str, app_name? str, timeout_ms? int (5000), poll_interval_ms? int (200), expected_states? list[str]

  Poll the accessibility tree until an element matching query and/or states appears

  Try:Wait for the login button to appearTry:Wait for the progress bar to become active
• mouse_dragfrom_x int, from_y int, to_x int, to_y int, button? str ('left'), modifiers? list[str], waypoints? list[[x,y,dwell_ms]], screenshot_after_ms? list[int]

  Drag from one point to another with smooth interpolation and optional waypoints

  Try:Drag file from position 100,200 to 500,600Try:Drag with waypoints to draw a custom shape
• focus_windowapp_name str

  Focus a window by application name

  Try:Focus the Firefox windowTry:Bring the terminal application to front

Comparable tools

Installation

uv tool install kwin-mcp

pip install kwin-mcp

{
  "mcpServers": {
    "kwin-mcp": {
      "command": "uvx",
      "args": ["kwin-mcp"]
    }
  }
}

FAQ

What desktop environments does kwin-mcp support?

kwin-mcp is specifically designed for KDE Plasma 6 on Wayland. It works with any Wayland application including Qt, GTK, and Electron apps running on KDE Plasma.

How does kwin-mcp handle input without triggering system authorization prompts?

kwin-mcp uses KWin's private EIS (Emulated Input Server) D-Bus interface directly, bypassing the XDG RemoteDesktop portal, which eliminates the need for user confirmation dialogs.

On Hacker News

Recent discussion from the developer community.

• Show HN: Kwin-MCP – MCP server for AI-driven Linux GUI automation via KWin▲ 1
  Story by isac322 · 2026-02-23

Compare kwin-mcp with