Online Access Free AB-620 Practice Test

A company uses an agent in Copilot Studio to generate structured responses for internal users.
The agent must meet the following requirements:
- Consistently follow a reusable instruction template for response tone and structure.
- Use an enterprise-approved foundation model for responses generated
through the template.
You need to configure custom prompts so the agent uses the Microsoft Foundry model catalog.
Which two actions should you perform? Each correct answer presents part of the solution.
NOTE: Each correct selection is worth one point.

• A.Configure Azure AI Search as the grounding source, so the model choice is enforced.
• B.Allow topic authors to define tone and structure directly in topic steps.
• C.Publish the agent to apply the Foundry model selection automatically.
• D.Enable citations, so the model is restricted to approved outputs.
• E.Select a model from the Foundry model catalog for the custom prompt configuration.
• F.Create a custom prompt template and apply it where the agent generates the structured response.

Drag and Drop Question
A team is preparing to assess an agent in Copilot Studio before expanding access to additional users.
The team requires an evaluation that provides controlled, repeatable, and consistently measured test runs.
The evaluation must:
- Be triggered directly by the team to control when testing occurs.
- Determine success based on clearly established acceptance criteria.
- Rely on a predefined source that ensures consistency across repeated
runs.
You need to configure the evaluation to meet the requirements.
What should you configure? To answer, move the appropriate configurations to the correct evaluation requirements. You may use each configuration once, more than once, or not at all.
You may need to move the split bar between panes or scroll to view content.
NOTE: Each correct selection is worth one point.

Correct Answer:

Explanation:
Box 1: Execute the evaluation on demand
To trigger structured, repeatable agent evaluations manually inside Microsoft Copilot Studio so your team can precisely control when testing occurs, you should click Evaluate directly from the Evaluate tab within the Copilot Studio user interface.
Selecting this option initiates an on-demand, interactive test run against your defined test sets.
Box 2: Comparing generated responses against defined expectation
Comparing generated responses against defined expectations is exactly how you establish pass/fail signals. Microsoft Copilot Studio automates this process so you can achieve controlled, repeatable, and consistently measured tests before expanding access.
Box 3: Use a curated prompt list as the interaction source.
Using a curated prompt list-known as an evaluation test set-is exactly how you ensure consistency. It removes subjective, one-off testing and establishes a structured framework for measuring response quality.
Reference:
https://learn.microsoft.com/en-us/microsoft-copilot-studio/analytics-agent-evaluation-overview

You need an agent to only be invoked by other trusted agents in the tenant, not directly by end users. What should you configure?

• A.Publish the agent to the public web channel
• B.Enable anonymous authentication
• C.Configure the agent as a connected agent without exposing a standalone chat channel
• D.Increase the token limit

A company uses an agent flow that calls a connector action. The flow sends an administrative notification if the action fails or times out.
Administrators require the following configuration to help troubleshoot and manage failures:
- Handle failure and timeout outcomes explicitly.
- Continue execution after a failure is handled.
- Avoid retrying the failed action automatically.
You need to configure the agent flow so that failures are handled in a controlled and predictable way.
Which two actions should you perform? Each correct answer presents part of the solution.
NOTE: Each correct selection is worth one point.

• A.Increase the timeout on the connector action.
• B.Configure the run after conditions for the notification action.
• C.Configure concurrency control and apply to each action.
• D.Add a scope action to group related actions.

Hotspot Question
A team is preparing to evaluate an agent in Copilot Studio before expanding access to additional users.
The team must select an evaluation method that meets the following requirements:
- Use a fixed set of prepared interactions.
- Determine responses against a predefined baseline.
- Support consistent comparison across repeated test runs.
You need to determine the evaluation method that satisfies the evaluation requirements.
Which evaluation method should you select for each requirement? To answer, select the appropriate option in the answer area.
NOTE: Each correct selection is worth one point.

Correct Answer:

Explanation:
Box 1: Use a curated prompt list.
The best evaluation method for this requirement is to use a curated prompt list.
Using a curated prompt list allows you to test the agent against a fixed, predefined set of prepared interactions. This method ensures consistency, as you can run the exact same test inputs before and after making updates to measure improvements or regressions accurately.
Copilot Studio supports automated testing and evaluation precisely through these types of structured test datasets.
Box 2: Compare against predefined expected outcomes.
To determine an agent's responses against a predefined baseline before expanding access, the best evaluation method is to compare against predefined expected outcomes.
Structured & Repeatable Validation: By configuring an evaluation set of conversations with exact or partial expected outcomes, you eliminate guesswork.
Targeted Graders: You can utilize different graders in the Copilot Studio Evaluation Tool (e.g., General Quality, Similarity, or Exact Match) to check if your agent's responses meet your business standards.
Box 3: Use a consistent scoring evaluation matrix
The best evaluation method to satisfy these requirements is to use a consistent scoring evaluation matrix via the built-in Agent Evaluation tool.
This method satisfies the need for repeated, reliable, and standardized comparisons across multiple testing cycles:
Automated & Structured: Agent Evaluation automatically runs up to 100 scenarios using consistent scoring mechanisms like exact text matching, semantic meaning comparison, or AI- driven quality grading.
Consistent Baselines: Every evaluation run generates a versioned, auditable scorecard, allowing you to see exactly whether changes improved or degraded the agent's performance over time.
Reference:
https://learn.microsoft.com/en-us/microsoft-copilot-studio/agents-experience/analytics-agent-evaluation-intro