Test and activate

Australia Enable AI

Release

australia

ft:locale

de-DE

ft:publication_title

Australia Enable AI

ft:clusterId

platai

bundleId

platai

workflow

Platform

Test and activate a desktop action in Agentic Desktop

Freigeben Version: Australia

Aktualisiert 12. März 2026

2 Minuten Lesedauer

Test the desktop action and then activate it so that it’s available as a tool in the AI Agent Studio. You can add this tool to the AI agents that execute desktop actions in your desktop environment.

Vorbereitungen

Capture your automation steps. For more information, see Automate repetitive tasks by auto-capturing steps in Agentic Desktop or Automate repetitive tasks by manually capturing steps in Agentic Desktop.
Configure the properties for screens, anchors, and steps. For more information, see Screen, anchor, and step properties in Agentic Desktop.
Add desktop action details, such as name, description, and associated applications, and review inputs and outputs. For more information, see Add details to desktop actions in Agentic Desktop.

Role required: sn_aia.admin

Warum und wann dieser Vorgang ausgeführt wird

You can test the desktop action before you activate it. You can also test individual screens to quickly identify and troubleshoot issues.

Execution workspace is launched to run and test the selected screen or the desktop action.

Agentic Desktop showing the Shipping Management app with fields on a Windows 10 background.

After you run the test by entering the required inputs manually, test results are shown either for a screen or entire desktop action in the Execution workspace. While running tests, if you do not see the Execution workspace coming forth, switch to the Execution workspace manually.

While working on a desktop action, the test values you enter are retained across test runs until you close the desktop action or reset the test values.

Prozedur

From the desktop action header, select Test.
Option 1: Run the screen-level test.
1. Enter required inputs for the screen that you want to test.
2. Select Test screen.
3. After the test run is complete, check the status.
  - If test is successful, proceed with testing the next screen.
  - If test isn’t successful, fix the issue highlighted in the error message and run the screen-level test again.
4. Wahlweise: Reset test data by selecting Reset.
5. Wahlweise: Cancel running test by selecting Cancel.
Option 2: Run the desktop action-level test.
1. Enter required inputs for all screens in a desktop action.
2. Select Run test.
3. After the test run is complete, check the status.
  - If test is successful, proceed with activating the desktop action.
  - If test isn’t successful, select Back to Test to check the configuration and run the test again.
Wahlweise: Cancel running test by selecting Cancel.
After a successful completion of the desktop action test and when you are satisfied with the result, select Activate to publish the desktop action in AI Agent Studio.
The Activate button sets the desktop action as active or inactive. Only the activated desktop actions can be selected from the drop-down of UI block actions in AI Agent Studio.
Hinweis:
After activation, save the desktop action to make it available as a tool in AI Agent Studio.

If you get en error that the file size is beyond 10 MB, you can increase the file size in a payload by creating system properties in the ServiceNow instance. For more information, see Increase payload limit through system properties.

Nächste Maßnahme

Create an AI agent. For more information, see Creating AI agents for Agentic Desktop.

Enable AI agents to perform desktop actions by adding desktop actions as tools. For more information, see Add a desktop action to an AI agent.