Go library

import windowctl "github.com/muthuishere/windowctl"

ws, err := windowctl.ListWindows(windowctl.Filter{App: "Google Chrome"})
ms, err := windowctl.ListMonitors()

err = windowctl.MoveZone(windowctl.Target{App: "Google Chrome"}, 2, "2B")
err = windowctl.MoveCoords(windowctl.Target{Title: "Terminal"}, 0, 0, 960, 1080)
err = windowctl.Resize(windowctl.Target{App: "Code"}, 900, 700)
err = windowctl.Focus(windowctl.Target{Title: "jira"})

ok, err := windowctl.CheckAccessibility()    // never prompts
err = windowctl.RequestAccessibility()       // prompts

Everything the CLI does, the library does — the command is argument parsing and output formatting over these calls, nothing more. If you are writing Go, import it; do not shell out to a binary you then have to ship and locate.

Where the logic lives

The per-OS layer implements six primitives: list windows, list monitors, move, focus, and the two permission calls. Everything expressible as a composition of those — zone maths, N:M splits, majority-overlap monitor resolution, filter matching, the relative-vs-absolute coordinate rule — lives in the portable package. So the semantics are identical on macOS, Windows and Linux, and they are unit tested once rather than three times.

Module

Go 1.24. Dependencies: golang.org/x/sys, and gorilla/websocket for the device layer. macOS builds use CGO against the system frameworks; Linux and Windows builds are CGO-free.