What is Agent Execution?
Agent execution is how Granite runs your automations. An AI agent interprets your task, plans the steps, and executes them on a Windows machine.Key Features
Human-in-the-Loop
Approve actions before they execute
Live Video
Watch execution in real-time
Screenshots
Automatic captures at each step
Recordings
Full video recordings of sessions
Execution Flow
Actions Execute
Each action:
- Shown to you (with HITL)
- Screenshot captured
- Executed on desktop
- Result verified
Agent Capabilities
The agent can:| Action | Examples |
|---|---|
| Click | Buttons, menus, links, icons |
| Type | Text fields, documents, search boxes |
| Navigate | Open apps, switch windows, scroll |
| Read | Extract text, verify content |
| Wait | For elements, for loading |
| Decide | Handle variations, recover from errors |
Execution Modes
- With HITL
- Without HITL
Human-in-the-Loop enabled (default)
- Agent pauses before sensitive actions
- You approve, modify, or cancel
- Maximum control and oversight
Real-Time Monitoring
During execution, you see:- Live video - The Windows desktop in real-time
- Chat log - Each action described
- Progress - Current stage indicator
- Controls - Pause, cancel buttons
Error Handling
When something goes wrong:- Agent describes the error
- Screenshot shows the state
- You can:
- Modify the approach
- Skip the action
- Cancel execution
After Execution
Once complete, you have:- Execution summary - Success/failure, duration
- Screenshot gallery - All captured images
- Video recording - Full session replay
- Detailed logs - Timestamped trace
Best Practices
Start with HITL enabled
Start with HITL enabled
Until you’re confident the automation works, keep HITL on.
Watch the video
Watch the video
Even successful runs may reveal optimization opportunities.
Use clear task descriptions
Use clear task descriptions
The clearer your description, the better the agent performs.
Test before going unattended
Test before going unattended
Run with HITL several times before disabling oversight.