Configuring Computer Use and Testing Pane in Copilot Studio

AB-620 By islever Editorial Last updated

Configure and monitor computer use for an agent

Short answer

CE Enable the Computer use capability and access the remote session within the testing pane to monitor real-time execution.

Question

A company needs an agent in Copilot Studio to complete tasks by interacting with a legacy desktop application that does not use APIs. The builder needs to validate that UI automation can run in the supported execution context and needs a way to inspect how the automation behaves during testing conversations.
To meet the business needs, the solution must meet the following requirements:
Enable the agent to perform UI automation against the legacy application.
Inspect execution details for the automation during conversations.
You need to configure the agent based on the requirements.
Which two actions should you perform? Each correct answer presents part of the solution.
NOTE: Each correct selection is worth one point.
  1. Configure a custom connector for the desktop app so the agent can automate the UI through authenticated calls.
  2. Add a Power Platform Desktop flow to control the desktop app.
  3. Access the remote session within the testing pane to monitor the real-time execution of the agent's desktop interactions. Correct Answer
  4. Publish the agent so the automation runs in the supported environment automatically.
  5. Enable the Computer use capability. Correct Answer

Key Takeaways

Tests knowledge of Copilot Studio's Computer use feature for UI automation without APIs and the testing pane's remote session for debugging.

Learn how to enable Computer use capability in Copilot Studio for legacy app automation and use the testing pane's remote session to inspect agent behavior.

Common mistake: Choosing custom connectors (A) is a common error, but they require APIs, which are absent here. Candidates must recognize that Computer use is the specific tool for API-less desktop interaction.

Expert Analysis

The scenario describes a legacy desktop application that lacks APIs, requiring the agent to interact with the user interface directly. In Copilot Studio, this is achieved through the Computer use capability (Option E). This feature allows the agent to perform UI automation tasks such as clicking buttons and typing text on a supported execution context (typically Windows VMs or specific remote environments). Without enabling Computer use, the agent cannot perform these actions.

The second requirement is to inspect how the automation behaves during testing conversations. Copilot Studio provides a testing pane where builders can run test conversations. When Computer use is enabled, this pane includes access to a remote session, allowing the builder to view the real-time execution of the agent's desktop interactions. This visibility is crucial for debugging UI automation issues. Therefore, accessing the remote session within the testing pane (Option C) satisfies the inspection requirement.

Option A is incorrect because custom connectors rely on REST APIs, which the legacy app does not have. Option B mentions Power Platform Desktop flows; while RPA can automate desktop apps, the question specifically asks about configuring the agent in Copilot Studio to perform UI automation via its built-in capabilities, not integrating an external flow for this specific validation step. Option D is incorrect because publishing the agent does not provide the testing/inspection mechanism required by the prompt.

Community comments correctly identify Options C and E as the solution, noting that Computer use is the key feature for API-less desktop interaction and the testing pane offers the necessary visibility.

Exam Strategy

When a question specifies 'no API' and 'desktop application', immediately look for 'Computer use' or 'UI automation' options. For testing requirements involving visual feedback or real-time monitoring of these actions, select options related to the 'testing pane' or 'remote session'. Avoid connector-based solutions when APIs are explicitly unavailable.

Frequently Asked Questions

What is the Computer use capability in Copilot Studio?

It is a feature that enables agents to perform UI automation on desktop applications by simulating user interactions like clicks and keystrokes, useful for apps without APIs.

How can I debug a computer use agent in Copilot Studio?

You can access the remote session within the testing pane to view the real-time execution of the agent's desktop interactions during test conversations.

More AB-620 FAQ →

Community Votes

C
50%
E
50%

50% of anonymous learners picked answer C. Votes are pick records left by other test-takers — they are not the verified answer.

Discussion 4 contributions

Compiled from public community threads and reader submissions, lightly edited for clarity. Handles and vote tallies are shown as recorded, and comments reflect the views of individual contributors.

From public community threads 4

  1. kevinirl 👍 2
    E. Enable the Computer use capability — This is the feature that lets a Copilot Studio agent perform UI automation against applications that have no API. Computer use drives the interface directly (clicking, typing, reading the screen) the way a person would, which is exactly what a legacy desktop app without APIs requires. This satisfies the first requirement. C. Access the remote session within the testing pane — Computer use runs in a remote/cloud session, and the testing pane lets you watch that session live so you can inspect how the automation behaves during test conversations. This satisfies the second requirement (inspect execution details during conversations).
  2. Joselito79 👍 2
    ✅ C. Access the remote session within the testing pane to monitor the real-time execution of the agent's desktop interactions. ✅ E. Enable the Computer use capability.
  3. Snoopy70 👍 1
    C. Access the remote session within the testing pane to monitor the real-time execution of the agent's desktop interactions. E. Enable the Computer use capability.
  4. massnonn 👍 1
    Access the remote session

Reader comments & corrections

No reader corrections yet — spotted an error or have a note? Share it below.

Log in to comment, report an error, or add a note about this question.

Copilot Studio Generative Answer Disclaimer Strategy

S-grade

Evaluate Copilot Agent Configuration Compliance Status

S-grade

Export Copilot Studio Agent for Production ALM

S-grade

Configuring Copilot Agent Content Control and Formatting

S-grade

← Back to AB-620 Study Guide