Allows controlling an AwesomeWM multi-monitor X11 desktop, including window listing, moving, resizing, fullscreen toggling, screenshots, typing, and clicking.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@computer-use move the browser window to the second monitor"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
AwesomeWM + X11 computer-use MCP for Ada. Beginning of a desktop-control surface: window geometry on a multi-monitor Awesome setup, plus screenshot, type, and click.
Coordinates are global (one system across all screens). screen is
Awesome's 1-based index so Ada can place a window on a specific monitor.
Fullscreen uses Awesome client.fullscreen, not a fake maximize.
Typing uses xdotool type (XTEST), the same injection path as
/workspace/whisper (pynput XTEST / vk keysyms). GTK4 apps discard XSendEvent.
Tool | Notes |
| screens + windows ( |
| one window's x,y,w,h,screen,fullscreen |
| global x,y or |
| width/height |
|
|
| image of that X window |
| current pointer x,y (global) + window under cursor |
| type text (max 4000 chars) |
| click x,y relative to the window |
Related MCP server: linux-computer-use
cd /workspace/mcp-computer-use && bun install
bun server.coffeeAda loads this as MCP computer-use (prefix: false). Requires awesome-client,
xdotool, scrot, and a working DISPLAY.
This server cannot be deployed
Eyes and hands on real Windows PCs — observe, click, type via Glasswarp API.
Operate Linux, macOS and Windows from your LLM. Every action runs through an auditable allowlist.
Automate cloud Chrome—navigate, click, type, screenshot, run code, record screen video
Stealth web automation for AI agents. Login, signup, navigate, screenshot.
