The agent-device server enables AI agents to automate real mobile, TV, and desktop apps through inspection, interaction, evidence capture, and workflow replay across iOS, Android, macOS, tvOS, and Linux platforms.

Device & App Management

• List devices (devices), boot devices (boot), list/install/reinstall apps (apps, install, reinstall, install-from-source), open apps/deep links/URLs (open), close apps or sessions (close), manage sessions (session), and check foreground app state (appstate).

UI Inspection

• Capture accessibility snapshots (snapshot), diff snapshots (diff), find elements by text/label/role/id (find), get element text/attributes (get), and assert UI state (is).

UI Interaction

• Tap/click (click, press), long press (longpress), fill text fields (fill), type into focused fields (type), focus inputs (focus), scroll (scroll), swipe (swipe), run structured gestures like pan/fling/pinch/rotate (gesture), navigate back (back), go home (home), open app switcher (app-switcher), rotate orientation (rotate), handle alerts (alert), manage keyboard (keyboard), read/write clipboard (clipboard), wait for conditions (wait), change OS settings/permissions (settings), push notifications (push), and trigger app-defined events (trigger-app-event).

Evidence & Diagnostics

• Capture screenshots (screenshot), record video (record), collect logs (logs), monitor network traffic (network), gather performance metrics (perf), and start/stop traces (trace).

Workflow Automation

• Replay recorded .ad scripts (replay), run multiple scripts with retry and JUnit reporting (test), and batch multiple commands in a single request (batch).

React Native / Metro

• Prepare Metro runtime, reload React Native apps (metro), and dismiss React Native overlays (react-native).

Supports native, Expo, Flutter, and React Native apps on simulators, emulators, and physical devices.

Automates Android emulators and devices, enabling app opening, UI accessibility tree inspection, element interaction (fill, tap), and evidence collection (screenshots, videos, logs, CPU/memory/perf).

Enables testing of Expo apps on iOS and Android, including opening Expo Go or dev clients, UI snapshotting, element interaction, and gathering evidence such as screenshots, videos, logs, and performance traces.

Automates iOS simulators and devices, enabling app opening, UI accessibility tree inspection, element interaction, and evidence collection (screenshots, videos, logs, network, performance).

Automates Linux desktop applications, enabling UI inspection via accessibility snapshots, element interaction, and evidence collection (screenshots, logs, performance).

Automates macOS desktop applications, enabling UI inspection via accessibility snapshots, element interaction, and evidence collection (screenshots, logs, performance).