Get device model, Android version, screen, battery, IP, and serial.

Get battery level, charging status, and temperature.

Get clipboard text (Android 10+).

Tap screen at pixel coordinates (x, y).

Swipe from (x1,y1) to (x2,y2) with optional duration (ms).

Type text into focused input. ASCII via key injection, Unicode via clipboard paste (Android 10+).

Send a safe key event (HOME, BACK, ENTER, etc.). Allowed: navigation, editing, media, alphanumeric keys.

Take screenshot, returns file path or base64.

Get screen resolution in pixels.

List installed packages, filterable by keyword and type (user/system/all).

Get version, SDK, install date, permissions for an installed app.

Launch an app by package name. Optionally specify an activity.

Force-stop an app by package name.

Get the package name and activity of the foreground app.

Execute an Android shell command. ⚠️ Requires confirm=true.

Execute a command in Termux bash. ⚠️ Requires confirm=true.

Copy a file from device to local computer.

Get UI element tree (accessibility dump) with bounding boxes and positions.

Take screenshot (base64) + UI tree in one call. Recommended for "looking" at the screen.

Find UI element by text/ID/content-desc and tap its center. Regex partial match.