Run any Hyprland dispatcher using Lua expression syntax for Hyprland 0.55+. Use path+args or raw_expression for dispatchers not covered by dedicated tools.
An MCP server for Hyprland desktop automation that allows AI assistants to see the screen, control mouse and keyboard, and manage windows using native Wayland tools. It integrates OCR for text-based interaction and supports complex multi-monitor setups with pixel-accurate coordinate mapping.
Get a semantic snapshot of the Hyprland desktop, listing monitors, workspaces, windows with geometry and addresses, active window, cursor, and layers. Call this first to obtain targets for actions.
Run a configured Hyprland keybind by its combo to execute the bound action directly; synthetic keypresses cannot trigger compositor binds, so use this for launchers or layout shortcuts.
Capture a native-resolution screen area around a point on a Hyprland desktop, then refine target coordinates on the image so an agent can click accurately.
Blocks until a real Hyprland compositor event occurs, such as a window opening, closing, title change, or layer opening. Filter by match and timeout to return event data without polling.
Open a desktop app by running a Hyprland command, optionally placing its window on a chosen workspace or in the background while waiting for it to appear.
Locate text on your screen using OCR to find matching coordinates for automation tasks. Specify target text and optional search areas like monitors, windows, or regions to get precise screen positions.
Search real-world Quickshell shells (Caelestia, Noctalia) for implementation examples: bars, control centers, IPC, multi-monitor, widgets, services, animations, and more. Filter by shell to compare approaches.
Run commands in Hyprland, place them on a specific workspace even for single-instance apps, and wait for the window. Returns window address, class, title, and workspace, or a timeout note.
Control the mouse cursor on Hyprland: move, click, drag, or scroll in global coordinates. Optionally capture desktop, screenshot, or focused window elements after the action.