Get Your CertAI-103PlayCompete
‹ All AI-103 questions

AI-103 Practice Question 44: Cross Project Agent Handoff

Microsoft Azure AI Apps and Agents Developer Associate (AI-103) · Test 5 · question 4 of 10

Implement generative AI and agentic solutionshard

A retailer's shopping-assistant prompt agent already carries a large toolbox for inventory lookup, order status, and recommendations. A separate fraud team owns and independently maintains its own specialized fraud-scoring agent in a different Foundry project. The shopping assistant needs to hand off suspicious-order checks to that agent for a small subset of orders, while staying in control of the conversation itself. The team wants to reuse the fraud team's existing agent as-is, not rebuild its logic, and avoid standing up a full multistep orchestration just for this one hand-off.

What should the team do?

Select one answer
Show answer and explanation

Answer:

  • B. Enable incoming A2A on the fraud-scoring agent, create an A2A connection from the shopping assistant's project, and add the A2A tool so the shopping assistant can call it directly

Microsoft names this exact case for the agent-to-agent (A2A) pattern: "for lightweight scenarios where one agent just needs to call another (without a formal, multistep workflow) expose an agent as an A2A endpoint and call it directly. This approach keeps the calling agent in control while it delegates a task to a specialized agent." Enabling incoming A2A on the fraud-scoring agent, wiring an A2A connection from the shopping assistant's project, and attaching the A2A tool lets the shopping assistant delegate without rebuilding anything or crossing project boundaries.

- A is wrong because function calling requires the shopping assistant's own application to implement and own the scoring logic. That reimplements exactly the capability the fraud team already built and maintains, instead of reusing it. - C is wrong because a Foundry workflow is built for repeatable, branching, human-in-the-loop orchestration across several steps. Standing up and maintaining a workflow graph for a single conditional hand-off is precisely the overhead the team wants to avoid. - D is wrong because it duplicates the fraud team's specialist inside another team's project instead of calling their independently owned and maintained agent; the two copies would drift, and connected agents (classic) are the older mechanism for agents built together in one project, not for delegating to another team's separately published agent.

Reference: Enable incoming A2A on a Foundry agent

Practice Test 5 ›

Study guides for AI-103

AI engineer roadmap 2026: skills, certs and the orderAI engineer roadmap for 2026: skills to learn in order, AI engineer vs ML engineer, which certifications fit each stage, and what to skip.
AI-200 exam guide: Azure AI Cloud Developer AssociateAI-200 replaced AZ-204: what the Azure AI Cloud Developer exam tests, how it differs from AI-103, price in USD and INR, format and how to prepare.
How AI Is Changing Software Jobs: Where to PrepareHow AI is changing software jobs: what history shows, the evidence on junior developers, what Klarna learned, new AI roles and five domains to prepare for.
AI-901 exam guide: Azure AI Fundamentals after AI-900AI-901 replaced AI-900: learn what changed, how much Python and Microsoft Foundry you need, exam costs in India, and how to prepare.
Best AI certifications 2026: AWS, Google, Microsoft, ClaudeThe best AI certifications in 2026 from AWS, Google, Microsoft, Claude and Databricks, by level, with price, status and which one to take first.
Microsoft AI certification path 2026: which exam firstMicrosoft AI certification path for 2026 - AI-901, AB-730, AI-103, AI-200, AI-300, AB-100, AI-500 and GH-300 by role, and which exam to take first.
AI-103 Certification Guide: What Changed From AI-102The AI-103 certification replaced AI-102 in 2026. Exam format, cost, the five skill domains, what changed and how to prepare, with free practice questions.
What Is RAG? How Retrieval-Augmented Generation WorksWhat RAG is and how it works: chunking, embeddings, vector search, hybrid search, reranking, GraphRAG and agentic RAG on AWS, Google and Azure.
Prompt engineering vs RAG vs fine-tuningWhen to use prompt engineering, retrieval-augmented generation, fine-tuning or continued pre-training, compared on cost, freshness, hallucination and privacy.