Integrate Agents with Azure AB-620 Practice Questions (Page 1)

Questions 1–5

1. 1foundation · easy

   What is the primary purpose of the Foundry model catalog in the context of Copilot Studio?

   • ATo store conversation logs from agent interactions
   • BTo create and manage Azure AI Search indexes
   • CTo manage the lifecycle of agent deployments
   • DTo provide a selection of AI models that can be used to build custom prompts
   Reveal Solution Discussion Select an answer first
2. 2foundation · easy

   When you connect Copilot Studio to Application Insights, what type of data can you analyze to understand agent performance?

   • AAzure resource utilization metrics
   • BUser interaction logs and custom telemetry events
   • CBilling and cost data
   • DNetwork traffic patterns
   Reveal Solution Discussion Select an answer first
4. 3expert · hard

   A banking company has a Copilot Studio agent that uses Azure AI Search and a custom prompt from Foundry. The security team requires that Application Insights logs do not contain any personal data, such as user names or account numbers. However, the support team needs to trace specific user issues. What should they do?

   • AUse Azure AI Search query logs instead of Application Insights.
   • BLog the user's name and account number in Application Insights but restrict access to the logs.
   • CUse a correlation ID in Application Insights logs and store personal data separately in a secure database.
   • DDisable Application Insights and rely on manual logging.
   Reveal Solution Discussion Select an answer first
5. 4application · medium

   A real estate company is building a Copilot Studio agent that answers questions about property listings. They have a dataset of listings in a CSV file stored in Azure Blob Storage. They want to use Azure AI Search to provide generative answers. The team wants to ensure that the agent can handle queries like 'Show me all 3-bedroom houses under $500k'. What should they do?

   • AUpload the CSV file directly to the index and rely on the model to parse the data.
   • BEnable Application Insights to analyze the query and generate a filter.
   • CUse Azure AI Search's built-in natural language processing to understand the query and filter automatically.
   • DCreate an index with fields for bedrooms and price, and use the Foundry prompt to instruct the model to filter based on these fields.
   Reveal Solution Discussion Select an answer first
6. 5application · medium

   A travel agency is creating a Copilot Studio agent that recommends vacation packages. They have a database of packages and a set of customer reviews. They want to use the Foundry model catalog to configure a custom prompt that combines the package data with the reviews to generate personalized recommendations. They also want to use Azure AI Search to retrieve relevant reviews. What should they do?

   • AEnable Application Insights to analyze customer reviews and generate recommendations.
   • BUse the Foundry model catalog to create a prompt that calls the travel agency's database directly.
   • CUse the Foundry prompt to define the recommendation logic, and use Azure AI Search to retrieve reviews that are then passed to the prompt.
   • DUse Azure AI Search to store the packages and reviews, and let the search results alone generate the recommendations.
   Reveal Solution Discussion Select an answer first