Skip to main content
AgentOS provides a set of capabilities for controlling machines. What’s available depends on the control mode and runtime mode you use.

Desktop Control

CapabilityHost ModeCompanion ModeDescription
ScreenshotsAvailableAvailable (via HDMI capture)Capture the current screen state.
GPU-accelerated screenshotsAvailableTake screenshots on GPU for better performance.
Keyboard inputAvailableAvailable (via USB / Bluetooth HID)Simulate key presses and combinations.
Real Unicode typingAvailableNative Unicode input (not clipboard-based).
Mouse inputAvailableAvailable (via USB / Bluetooth HID)Click, move, drag, and scroll.
Multi-display supportAvailableInteract with multiple monitors.
Window managementAvailableSelect, resize, move, and focus windows.
Process managementAvailableStart, stop, and monitor OS processes.
Automation recoveryAvailableRecover automatically from display changes.
CLIAvailableEnhanced command-line interface.

Windows Service (OS Service Mode Only)

These capabilities require AgentOS to be installed as a Windows service.
CapabilityStatusDescription
SYSTEM-level serviceAvailableRun as a Windows service with SYSTEM privileges.
RDP disconnect recoveryAvailableKeep automation running when RDP sessions disconnect.
Login screen automationAvailableInteract with the Windows logon screen.
Secure Attention SequenceAvailableSend CTRL+ALT+DEL programmatically.
Background session automationAvailableAutomate sessions that are not in the foreground.
Multi-user session controlPlannedControl multiple user sessions on the same machine.
Virtual displayPlannedCreate virtual displays for headless environments.

Mobile Devices

CapabilityStatusDescription
ADB (Android)AvailableControl Android devices via USB.
IDB (iOS)PlannedControl iOS devices via USB.

Hardware (Companion Mode Only)

CapabilityStatusDescription
HID over USBAvailableSimulate keyboard and mouse via USB connection.
HID over BluetoothAvailableSimulate keyboard and mouse via Bluetooth.
Screenshot via media streamAvailableCapture screen output via HDMI-to-USB dongle or webcam.
Raspberry Pi supportPlannedUse a Raspberry Pi as the companion device.

Platform Support

PlatformHost ModeCompanion Mode
Windows 10 22H2+Available
Windows 11Available
Windows Server 2022+Available
Windows 10 < 22H2Available
Windows Server 2019Available
macOSAvailable
LinuxAvailable
AndroidAvailable (ADB)
iOSPlanned (IDB)

Future

CapabilityStatusDescription
C# APIPlannedNative C# interface for .NET integrations.
Future automation APIsPlannedIntegration with emerging OS automation frameworks.