Create and Manage Copilots with Microsoft Copilot Studio · 15% of the exam

Extend Copilots with Plugins and Connectors: free practice questions

5 sample questions from our 12-question bank for this domain — answers and explanations included. These are the same scenario-based style as the real Microsoft exam.

1. Tailspin Toys has a Copilot Studio copilot that uses generative answers over a Dataverse knowledge source. During testing, the team observes that the AI sometimes generates answers that mix data from unrelated Dataverse tables, causing confidentiality concerns. What is the BEST approach to address this?

  • A. Disable generative answers entirely and author explicit topics for every supported question.
  • B. Apply row-level security (RLS) on the Dataverse tables and ensure the copilot's service account only has access to permitted tables, combined with content moderation settings.✓ Correct
  • C. Switch the knowledge source from Dataverse to a SharePoint document library to avoid cross-table data mixing.
  • D. Set the generative answers node to require a confidence score above 90% before returning any response.
Explanation

The correct approach is to enforce row-level security (RLS) on the Dataverse tables so the copilot's service identity can only access permitted data, and to configure content moderation to further filter responses. This addresses both the data-access root cause and adds a safety layer. Option A is overly restrictive and eliminates valuable AI functionality. Option C does not inherently solve the problem—confidentiality issues can also occur with SharePoint if permissions are not managed. Option D is incorrect because the 'confidence score' threshold controls whether a generative answer is surfaced at all; it does not restrict WHICH Dataverse tables or rows are accessed during generation.

2. A developer wants to configure a Copilot Studio copilot so that when no topic matches a user's message, the copilot attempts to answer using AI-generated responses from configured knowledge sources rather than displaying a default 'I didn't understand that' message. Which TWO steps are required to achieve this? (Select TWO.)

  • A. Enable the 'Generative answers' (Boosted conversations) setting at the copilot level.✓ Correct
  • B. Configure at least one knowledge source (such as a website URL, SharePoint site, or uploaded file) for the copilot.✓ Correct
  • C. Create a fallback topic and manually add a generative answers node with a hardcoded prompt.
  • D. Publish the copilot to a Microsoft Teams channel so Azure OpenAI can access it.
  • E. Enable the 'Allow AI to use its own general knowledge' option in the content moderation settings.
Explanation

To enable boosted conversations (generative answers when no topic matches), you must: (1) turn on the Generative answers / Boosted conversations feature at the copilot settings level, and (2) have at least one knowledge source configured so the AI has content to search. These two steps together activate the automatic fallback behavior. Option C is an alternative manual approach using the System Fallback topic, but it is not the required steps for the built-in boosted conversations feature described in the question. Option D is incorrect—publishing to Teams is a channel deployment step unrelated to enabling generative answers; Azure OpenAI access is handled by the platform, not the channel. Option E misrepresents content moderation controls—'Allow AI to use general knowledge' is a separate toggle and is not required for knowledge-source-based generative answers.

3. A developer is configuring generative answers in Copilot Studio and notices that some AI-generated responses contain content that is off-topic or potentially harmful. Which feature should the developer enable to address this?

  • A. Fallback topic with a redirect node to a human agent.
  • B. Content moderation settings to filter harmful or off-topic generative answers.✓ Correct
  • C. Disable generative answers entirely and replace them with authored topics.
  • D. Set the knowledge source confidence threshold to 100%.
Explanation

Content moderation is the dedicated feature in Copilot Studio that filters harmful, off-topic, or otherwise undesirable generative answers before they reach the user. Option A (fallback topic with handoff) addresses unanswered questions but does not filter harmful content from responses that are generated. Option C is an overreaction that eliminates useful functionality rather than controlling output quality. Option D is incorrect because there is no single '100% confidence threshold' control for content safety—confidence thresholds relate to topic triggering, not content moderation.

4. Contoso is building a copilot that must call an existing Microsoft 365 connector (such as the SharePoint connector available in Power Platform) directly from a conversational turn, without using a separate Power Automate flow. What type of object should the developer add to the copilot?

  • A. A knowledge source pointing to the connector endpoint.
  • B. A connector action that references the Microsoft 365 connector.✓ Correct
  • C. A generative plugin manifest that maps to the connector's operations.
  • D. A Dataverse virtual table linked to the connector.
Explanation

Connector actions allow a Copilot Studio copilot to call Microsoft 365 or third-party connectors directly within a conversation turn, without requiring a separate Power Automate flow. Option A is incorrect because knowledge sources are for AI-driven information retrieval, not triggering connector operations. Option C (generative plugin manifest) relates to Microsoft 365 Copilot extension scenarios, not inline connector calls in Copilot Studio. Option D (Dataverse virtual table) is a data integration pattern unrelated to directly invoking connector actions in a copilot.

5. A developer is writing a plugin manifest for a Microsoft 365 Copilot extension. Which THREE elements are essential in the plugin manifest to enable AI orchestration to correctly identify and invoke the plugin? (Select THREE.)

  • A. A human-readable description of what the plugin does, written in natural language for AI reasoning.✓ Correct
  • B. A reference to the API or connector operations the plugin exposes (functions/actions).✓ Correct
  • C. The conversation history of previous user interactions with the plugin.
  • D. Parameter descriptions for each operation, including data types and intent.✓ Correct
  • E. A list of Microsoft Teams channels approved to use the plugin.
  • F. The billing plan and license tier required to use the plugin.
Explanation

A plugin manifest for Microsoft 365 Copilot AI orchestration must include: (1) a natural-language description so the AI knows WHEN to use the plugin; (2) references to the API operations/functions the plugin exposes so the AI knows WHAT it can do; and (3) detailed parameter descriptions with data types so the AI can correctly populate inputs. Option C (conversation history) is not part of a manifest—it is runtime state. Option E (Teams channel approvals) is an administrative deployment concern, not a manifest element. Option F (billing/license tier) is irrelevant to the manifest's AI orchestration purpose.

7 more questions in this domain

Practice the full bank with instant grading, flashcards, and a timed mock exam.

Start practicing free