VNC (Virtual Network Computing) access provides a graphical desktop environment for your Daytona Sandbox directly in the browser. This allows you to interact with GUI applications, desktop tools, and visual interfaces running inside your sandbox.
VNC and Computer Use work together to enable both manual and automated desktop interactions. VNC provides the visual interface for users to manually interact with the desktop, while Computer Use provides the programmatic API for AI agents to automate mouse, keyboard, and screenshot operations. Through VNC, you can observe AI agents performing automated tasks via Computer Use in real-time.
- GUI application development: build and test desktop applications with visual interfaces
- Browser testing: run and debug web applications in a full browser environment
- Visual debugging: inspect graphical output and UI behavior in real-time
- Desktop tool access: use graphical IDEs, design tools, or other desktop software
- Agent observation: watch AI agents perform automated tasks through Computer Use
Access VNC from Dashboard
Access the VNC desktop environment directly from the Daytona Dashboard ↗.
- Navigate to Daytona Sandboxes ↗
- Locate the sandbox you want to access via VNC
- Click the options menu (⋮) next to the sandbox
- Select VNC from the dropdown menu
This opens a VNC viewer in your browser with a Connect button.
- Click Connect to establish the VNC session
Once connected, a full desktop environment loads in your browser, providing mouse and keyboard control over the sandbox’s graphical interface.
Programmatic VNC management
Daytona provides methods to start, stop, and monitor VNC sessions and processes programmatically using the Computer Use references as part of automated workflows.
Start VNC
Start all VNC processes (Xvfb, xfce4, x11vnc, novnc) in the sandbox to enable desktop access.
result = sandbox.computer_use.start()print("VNC processes started:", result.message)const result = await sandbox.computerUse.start();console.log('VNC processes started:', result.message);result = sandbox.computer_use.startputs "VNC processes started: #{result.message}"err := sandbox.ComputerUse.Start(ctx)if err != nil { log.Fatal(err)}defer sandbox.ComputerUse.Stop(ctx)
fmt.Println("VNC processes started")curl 'https://proxy.app.daytona.io/toolbox/{sandboxId}/computeruse/start' \ --request POSTFor more information, see the Computer Use reference.
Stop VNC
Stop all VNC processes in the sandbox.
result = sandbox.computer_use.stop()print("VNC processes stopped:", result.message)const result = await sandbox.computerUse.stop();console.log('VNC processes stopped:', result.message);result = sandbox.computer_use.stopputs "VNC processes stopped: #{result.message}"err := sandbox.ComputerUse.Stop(ctx)if err != nil { log.Fatal(err)}
fmt.Println("VNC processes stopped")curl 'https://proxy.app.daytona.io/toolbox/{sandboxId}/computeruse/stop' \ --request POSTFor more information, see the Computer Use reference.
Get VNC status
Check the status of VNC processes to verify they are running.
response = sandbox.computer_use.get_status()print("VNC status:", response.status)const status = await sandbox.computerUse.getStatus();console.log('VNC status:', status.status);response = sandbox.computer_use.statusputs "VNC status: #{response.status}"status, err := sandbox.ComputerUse.GetStatus(ctx)if err != nil { log.Fatal(err)}
fmt.Printf("VNC status: %v\n", status["status"])curl 'https://proxy.app.daytona.io/toolbox/{sandboxId}/computeruse/status'For more information, see the Computer Use reference.
For additional process management operations including restarting individual processes and viewing logs, see the Computer Use reference.
Automating desktop interactions
Once VNC is running, you can automate desktop interactions using Computer Use. This enables AI agents to programmatically control the mouse, keyboard, and capture screenshots within the VNC session.
Available operations:
- Mouse: click, move, drag, scroll, and get cursor position
- Keyboard: type text, press keys, and execute hotkey combinations
- Screenshot: capture full screen, regions, or compressed images
- Display: get display information and list open windows
For complete documentation on automating desktop interactions, see Computer Use.
Example: Automated browser interaction
# Start VNC processessandbox.computer_use.start()
# Click to open browsersandbox.computer_use.mouse.click(50, 50)
# Type a URLsandbox.computer_use.keyboard.type("https://www.daytona.io/docs/")sandbox.computer_use.keyboard.press("Return")
# Take a screenshotscreenshot = sandbox.computer_use.screenshot.take_full_screen()// Start VNC processesawait sandbox.computerUse.start();
// Click to open browserawait sandbox.computerUse.mouse.click(50, 50);
// Type a URLawait sandbox.computerUse.keyboard.type('https://www.daytona.io/docs/');await sandbox.computerUse.keyboard.press('Return');
// Take a screenshotconst screenshot = await sandbox.computerUse.screenshot.takeFullScreen();Required packages
The default sandbox image includes all packages required for VNC and Computer Use. If you are using a custom image, you need to install the following packages.
VNC and desktop environment
| Package | Description |
|---|---|
xvfb | X Virtual Framebuffer for headless display |
xfce4 | Desktop environment |
xfce4-terminal | Terminal emulator |
x11vnc | VNC server |
novnc | Web-based VNC client |
dbus-x11 | D-Bus session support |
X11 libraries
| Library | Description |
|---|---|
libx11-6 | X11 client library |
libxrandr2 | X11 RandR extension (display configuration) |
libxext6 | X11 extensions library |
libxrender1 | X11 rendering extension |
libxfixes3 | X11 fixes extension |
libxss1 | X11 screen saver extension |
libxtst6 | X11 testing extension (input simulation) |
libxi6 | X11 input extension |