AI-103: Developing AI Apps and Agents on Azure

By Microsoft · Question Mode

QuestionQ12

VerifiedImplement generative AI and agentic solutions

You have a Python application named App1 that integrates with a Microsoft Foundry project named Project1. You need App1 to meet the following requirements:

  • Authenticate by using a Microsoft Entra managed identity.
  • Send prompts to a deployed model using the Azure OpenAI Responses API.

How should you complete the Python code?

Select
from azure.identity import DefaultAzureCredential
from azure.ai.projects import AIProjectClient

credential = ()

project_client = AIProjectClient(
    endpoint="https://contosoai.services.ai.azure.com/api/projects/project1",
    credential=credential,
)

with project_client.get_openai_client() as openai_client:
    response = openai_client.responses.(
        model="trail-guide-chat",
        input="Create a 3-day hiking itinerary near Seattle.",
    )

    print(response.output_text)
Explanation

A managed identity is consumed in the Azure SDK through azure.identity.DefaultAzureCredential, which automatically discovers and uses the assigned managed identity (or other available credential sources) without requiring a stored key or client secret - unlike AzureKeyCredential (API-key auth) or ClientSecretCredential (requires a stored application secret). Once an AIProjectClient exposes an OpenAI-compatible client through get_openai_client(), sending a new prompt through the Responses API is done by calling responses.create(model=..., input=...); create() issues a new model response, while retrieve() only fetches an existing response and compact() is unrelated to submitting a prompt.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!