AB-410 Business Application Logic And Automation Practice Test 2 — 15 Questions

Question 1

Which THREE of the following are AI Hub features? (Choose THREE)

Accepted Answer

Document processing. AI Hub features include Document processing (A), which uses prebuilt and custom models to extract text, key-value pairs, and tables from documents such as invoices and receipts. Sentiment analysis (B) is also an AI Hub capability, evaluating text to determine whether the expressed opinion is positive, negative, or neutral. Text classification (E) is likewise an AI Hub feature, assigning custom or prebuilt category labels to text for scenarios like content tagging and intent detection. SQL Server query editor (C) belongs to database tooling such as SQL Server Management Studio, and Power BI dashboard builder (D) is part of Power BI for visualization and reporting, so neither is an AI Hub feature.

Answer

SQL Server query editor

Answer

Power BI dashboard builder

Question 2

Which THREE of the following are true about Dataverse 'Business Rules'? (Choose THREE)

Accepted Answer

They run on the client side. Option C is correct because Dataverse business rules execute on the client side (in model-driven apps and forms) to provide immediate, no-code logic without requiring server-side plug-ins. Option D is correct because business rules support field-level validation, such as checking that a value is within a range or that a required field is populated, and can display error messages to the user. Option E is correct because business rules can set field values (for example, defaulting or clearing a field based on another field's value) as part of their actions. Option A is incorrect because business rules are defined and run within Dataverse/model-driven apps, not exclusively in Power Automate. Option B is incorrect because business rules cannot call external APIs; that requires plug-ins, Power Automate, or custom connectors.

Answer

They can only be used in Power Automate

Answer

They can trigger external APIs

Question 3

You are building a Power Automate flow that integrates with an external REST API. The API requires a custom authentication header. Where do you configure this?

Accepted Answer

In the HTTP action Headers property. Custom authentication headers for a REST API call are configured on the HTTP action's Headers property, where you add key-value pairs such as Authorization or a proprietary API key header. This is the only place in a Power Automate flow where you can inject arbitrary HTTP headers into an outbound request. The Headers field accepts dynamic content, so you can reference a secure input or environment variable for the secret value.

Answer

In the trigger settings

Answer

In the connection settings

Answer

In the Environment variables

Question 4

You are analyzing AI Hub logs. Which status indicates that the model was successfully invoked but failed to return a prediction?

Accepted Answer

Failed. In AI Hub logging, a 'Failed' status specifically indicates that the model was invoked (i.e., the request reached the model endpoint) but the model did not return a valid prediction. This could be due to an error during inference, such as a runtime exception, timeout, or invalid input that the model could not process. The key distinction is that the invocation occurred, but the prediction step failed.

Answer

Succeeded

Answer

Skipped

Answer

Waiting

Question 5

You are building a Power Automate flow that triggers when a row is modified in Dataverse. You need to ensure the flow only processes records where the 'Status' column changes from 'Draft' to 'Active'. How should you configure this?

Accepted Answer

Configure Trigger Conditions in the Settings of the trigger action using the @equals expression.. Trigger Conditions are evaluated by the Dataverse connector before the flow run is even created, so only rows where the Status column transitions from Draft to Active will fire the flow. Using the @equals expression against the trigger output (e.g., @equals(triggerOutputs()?['body/status'],'Active')) combined with a check on the previous value filters at the source. This is the most efficient pattern because it prevents unnecessary runs entirely.

Answer

Use a Filter Array action immediately after the trigger.

Answer

Add a Condition action after the trigger to check the Status value.

Answer

Enable the 'Only trigger on specific columns' feature in the trigger settings.

Question 6

You have a Copilot in your canvas app that uses a custom prompt. You want the Copilot to reference real-time data from a SharePoint list. Which component is required to enable this integration?

Accepted Answer

A connector configured within the Copilot's knowledge base or plugin.. To let a Copilot in a canvas app reference real-time SharePoint data, you add a connector to the Copilot's knowledge base or plugin configuration. The Copilot uses this connector at runtime to query SharePoint directly, so the data is fetched live rather than cached or synced. This is the native extensibility mechanism for grounding Copilot responses in external data sources.

Answer

A custom connector to SharePoint.

Answer

A Power Automate flow with the 'Run a flow' action.

Answer

A Dataverse table that syncs with SharePoint.

Question 7

You need to share a cloud flow with a colleague so they can manage its runs and edit the workflow. What is the minimum permission level they require?

Accepted Answer

Owner. To manage a cloud flow's runs and edit its workflow, a colleague needs Owner permission. In Power Automate, Owner is the only sharing level that grants both editing rights and access to run history and flow management. Contributor allows editing but not full management, and Viewer/Run-only are read or execution-only.

Answer

Contributor

Answer

Viewer

Answer

Run-only user

Question 8

Which feature in AI Hub allows you to test AI models without creating a full flow?

Accepted Answer

Model Test tab. The Model Test tab in AI Hub (part of the AI Builder / Power Platform AI Hub experience) allows makers to test AI models directly without building a full Power Automate flow or app. It provides a sandbox to input sample data and view model predictions, making it ideal for quick validation before integration.

Answer

AI Playground

Answer

Power Apps Studio

Answer

Dataverse preview

Question 9

Which action should be used in Power Automate to handle errors when a specific step might fail due to transient network issues?

Accepted Answer

Configure run after. The 'Configure run after' setting on an action lets you define behavior when a previous step fails, times out, or is skipped. For transient network issues, you can set the next action to run after the failed step and implement retry logic or error handling. This is the standard way to handle errors in Power Automate.

Answer

Scope action

Answer

Terminate action

Answer

Do until loop

Question 10

You need to retrieve the current user's email address within a Power Automate flow triggered by a Power App. What is the most reliable way to do this?

Accepted Answer

Use the 'Get user profile (V2)' action.. The 'Get user profile (V2)' action in Power Automate queries Microsoft Entra ID (formerly Azure AD) via the Office 365 Users connector and returns the full profile of the signed-in user, including the mail and userPrincipalName attributes. Because the flow is triggered from a Power App, the user context is passed through the connection, so this action reliably resolves the current user's email without needing any input parameters. It is the standard, supported method for retrieving the caller's identity details inside a flow.

Answer

Use an environment variable.

Answer

Use a formula in the flow input.

Answer

Use the 'Get my profile' trigger.

Question 11

When configuring AI Builder for Document Processing, which THREE steps are required for a successful model deployment?

Accepted Answer

Define the document fields to extract.. Option A is correct because AI Builder document processing models must be taught which fields to extract (for example invoice number, total, or date) by defining the document fields during model configuration, which establishes the schema the model learns to predict. Option D is correct because after training and testing, the model must be published so it becomes available to Power Apps, Power Automate, and other consumers; an unpublished model cannot be used in production flows or apps. Option E is correct because AI Builder requires a minimum of five sample documents to train a document processing model, providing the labeled examples from which the model learns field extraction. Option B is not required because AI Builder is a Microsoft Power Platform capability licensed through Power Apps/Power Automate and does not require a dedicated Azure subscription. Option C is not required because AI Builder trains on the original document formats (such as PDF, JPG, or PNG), not on documents converted to JSON.

Answer

Create a dedicated Azure subscription.

Answer

Convert all documents to JSON format.

Question 12

You are designing a Power Automate cloud flow that uses an AI Builder custom document processing model to extract fields from invoices. The flow must handle documents that fail extraction by logging the error and continuing without stopping the entire run. Which configuration should you apply to the AI Builder action?

Accepted Answer

Set the action's retry policy to none and enable 'Configure run after' on the subsequent logging action to run only if the AI Builder action fails.. Using run-after configuration on a subsequent action to trigger on failure or timeout, combined with disabling retries, allows the flow to log the error and continue. This is the correct pattern for graceful error handling in Power Automate without halting the entire run.

Answer

Add a parallel branch that runs the logging action simultaneously with the AI Builder action.

Answer

Wrap the AI Builder action in a Scope action and set the Scope's concurrency control to 1, then place the logging action after the Scope.

Answer

Enable 'Secure Inputs' and 'Secure Outputs' on the AI Builder action and add a Terminate action with status Succeeded after it.

Question 13

You are building an intelligent application in Power Apps that must automatically classify incoming support emails as either 'Urgent' or 'Normal' based on the email body text. You want to use a prebuilt AI model to avoid custom training. Which AI Builder model should you use?

Accepted Answer

Category classification model. The Category classification model in AI Builder is purpose-built to assign text to one of several predefined categories. It supports training with your own labeled data to improve accuracy, but also works as a prebuilt model for common scenarios. It directly addresses the need to classify emails into 'Urgent' or 'Normal', making it the correct choice.

Answer

Sentiment analysis model

Answer

Key phrase extraction model

Answer

Entity extraction model

Question 14

You are building a Power Automate cloud flow that must trigger when a new email arrives in a shared mailbox. Which trigger should you use?

Accepted Answer

When a new email arrives (V3) with the shared mailbox address specified. The correct trigger is 'When a new email arrives (V3)' with the shared mailbox address specified in the 'Original Mailbox Address' field. This configuration ensures the flow triggers only for emails in the shared mailbox. Other options either filter on the wrong field or reference a non-existent trigger, so they would not achieve the desired behavior.

Answer

When a new email arrives in a shared mailbox (V2)

Answer

When a new email arrives (V3) with the shared mailbox address in the 'To' field

Answer

When a new email arrives (V3) with the shared mailbox address in the 'From' field

Question 15

You are building a Power Automate cloud flow that must call a Copilot Studio agent to summarize customer emails. The agent requires a bearer token for authentication. You need to ensure the flow can authenticate to the agent without storing secrets in the flow definition. What should you configure in the flow?

Accepted Answer

Use the Azure AD OAuth 2.0 connection with a service principal and store the client secret in Azure Key Vault.. The requirement is to authenticate to a Copilot Studio agent without embedding secrets in the flow. Using a service principal with OAuth 2.0 and retrieving the client secret from Azure Key Vault at runtime achieves this securely. It aligns with Microsoft's recommended practices for secret management and supports automated, non-interactive authentication.

Answer

Add a 'HTTP with Azure AD' action and manually paste the bearer token into the action's headers.

Answer

Enable the 'Run only user' trigger and rely on the signed-in user's credentials to authenticate to the agent.

Answer

Configure the agent to allow anonymous access and call it without authentication.