Testing automations
Amazon Quick Automate provides comprehensive capabilities for testing your automations. This section describes the tools and features available for validating your automation behavior.
Test modes
Choose from two modes to test your automation:
Run mode - Executes the automation from start to finish without stopping. Any breakpoints in the automation will be ignored. This mode is best for testing the complete flow of your automation.
Debug mode - Executes the automation and pauses at any breakpoints you've set or if an exception occurs. This allows you to inspect the automation state, monitor variables, and validate behavior at specific points in your process.
Test pane
The Test pane includes includes two main tabs that help you monitor and control your automation test:
Setup tab
The Setup tab provides tools for managing your test configuration:
Breakpoint manager - Lists all breakpoints currently set in your automation. Remove breakpoints directly from this view without returning to the canvas.
Test navigation - Quick access links to view:
View runs - See all automation runs from testing
View cases - View cases created during testing
View tasks - Review any human-in-the-loop tasks generated
Monitor tab
The Monitor tab displays real-time information about your running automation:
UI streaming - Watch browser automation actions execute in real-time
Variable watcher - Monitor and modify variable values
Logs - Track detailed automation progress
Test statuses
During testing, your automation will be in one of these statuses:
Running - The automation is actively executing actions
Paused (breakpoint) - Execution has stopped at a user-defined breakpoint
Paused (exception) - Execution has stopped due to an error condition
Completed - The automation has finished all steps successfully
Failed - The automation encountered an unrecoverable error and stopped
Running and debugging
Test automations with inputs and outputs
When you test an automation that has input and output fields defined, Amazon Quick Automate validates the input values and displays both inputs and outputs as expandable cards in the Monitor pane alongside the execution logs.
Running an automation with inputs from the authoring studio
When you choose Run or Debug to test an automation with inputs and outputs, the authoring studio (Studio) displays a dynamic input form titled Provide input values. Amazon Quick Automate automatically generates this form based on the schema you defined.
The form includes:
Field names from your schema.
Input controls appropriate for each data type (text fields for strings, number inputs for integers, and file upload buttons for files).
Required field indicators (marked with an asterisk).
Default values pre-populated where you defined them.
Client-side validation for data types, required fields, and file size limits (maximum 5 MB for files).
To run an automation with inputs:
Open your automation in Studio and choose Run or Debug.
In the Provide input values form, enter values for each field:
For text and number fields, type the values directly
For file fields, choose Browse to select a file from your local machine
Optional fields can be left blank unless you want to override default values
Choose Start to begin the automation with your provided inputs.
Amazon Quick Automate performs two layers of validation:
Client-side validation – Checks for blank required fields, correct data types, and file size limits before execution starts
Runtime validation – Validates the complete input payload against your JSON schema during execution
Viewing inputs and outputs
After the automation runs, input and output values appear as collapsible cards in the Monitor pane:
Input card – Displays at the beginning of the run logs, showing all values you provided when starting the automation
Output card – Displays at the end of the run logs after execution completes, showing all values your automation produced
Both cards are collapsed by default. When you expand them, you can:
View the complete schema definition and actual values
Download any file objects included in the inputs or outputs
Copy the structured data as JSON
Download the complete payload as a JSON file
Note
If the automation terminates with an error, the output card won't be available.
Setting breakpoints
Breakpoints allow you to pause your automation at specific points:
To add a breakpoint:
Hover over an action or step on the canvas
Click Add breakpoint
The breakpoint icon appears on the action or step
To remove a breakpoint:
Click Remove breakpoint on the action
Or use the breakpoint manager in the Setup tab
Note
Set up breakpoints before starting debug mode.
Debug controls
When your automation is paused, you have several control options:
Next - Run the next action in your automation and pause again
Continue - Resume the run until the next breakpoint is reached
Stop - End the current test run
Ignore - If paused on an exception, this will clear the exception and continue as if that action was successful. This is useful if the exception is not blocking the rest of your test and you want to continue.
Retry - If paused on an exception, this will attempt to run the current action again. This is useful if the issue is temporary or if you want to change the value of a variable and retry with the new value
Deactivating steps
To temporarily disable any actions or steps without removing them, you can deactivate them before starting your test. Deactivated actions or steps will be skipped during testing.
Select the step to deactivate
Click the Menu options
Choose Deactivate
The step appears grayed out
To reactivate:
Select the deactivated step
Click the Menu options
Choose Activate
Variable watcher
The Variable watcher in the Monitor tab provides visibility to inspect data being used by your automation during the test. To inspect a variable:
Open the Variable watcher in the Monitor tab while the test is paused
Click Add to search and select variables to inspect
View the current value for the variable
Modify values to test different scenarios:
Click the pencil icon next to any variable
Enter a new value to test
Continue the run to proceed with the new value
This feature is particularly useful for:
Testing different data conditions
Validating error handling
Simulating edge cases
Troubleshooting
Fix with Assistant
When your automation encounters an exception, the Automation Assistant can help identify and resolve issues:
Click Fix with Assistant when paused at an error
The Assistant will:
Analyze the error message
Review the run logs
Check its troubleshooting knowledge base
Provide a clear explanation of the issue
Suggest potential solutions
Chat with the Assistant to:
Ask questions about the error
Update the automation based on any of the recommended changes
Test metrics
After your test completes, Amazon Quick Automate provides comprehensive test metrics to help understand automation performance and validate test coverage. View key metrics about your test run:
Total running time - How long the test took from start to finish
Number of completed cases - Count of cases that finished processing (excluding those created or pending)
Average time per case - The processing average time for each completed case
Case success rate - Percentage of cases that completed without exceptions
Count of tasks created - Number of human-in-the-loop tasks generated during the test
Action coverage - Percentage of actions in your automation that executed during the test, helping identify untested paths
