An enterprise is designing a Copilot Studio multi-agent solution for an HR department. A front-door agent must handle all incoming employee requests. Requests related to benefits should go to a Benefits agent, requests about payroll should go to a Payroll agent, and requests about leave management should go to a Leave agent. Requests that don't clearly match any category should be handled by the front-door agent itself using generative AI answers from an HR knowledge base. The three specialized agents already exist and are published. Which front-door agent configuration achieves this design?

Copilot Studio's multi-agent architecture supports a hub-and-spoke model where a front-door agent uses generative AI orchestration to dynamically route requests to registered specialized agents. This design leverages the language model's intent understanding rather than static routing rules.

Option A is correct because it uses all three of the required platform mechanisms together. First, enabling generative AI orchestration on the front-door agent allows it to classify incoming messages by semantic intent. Second, registering the Benefits, Payroll, and Leave agents as agent connections in the front-door agent's configuration makes them visible to the orchestration layer as routing targets. The description field for each agent connection is critical — the language model uses these descriptions to determine which agent is the best match for a given request. Third, attaching the HR knowledge base to the front-door agent provides the generative AI fallback capability: when no specialized agent's domain matches the intent, the front-door agent answers using the knowledge base.

Option B is incorrect because trigger phrase-based routing cannot handle semantic variations. Employees might ask about 'my PTO balance' (Leave intent), 'vacation days remaining' (Leave intent), or 'time off allowance' (Leave intent) — only a small fraction of actual phrasings would match a fixed trigger phrase list. The Power Automate fallback for knowledge base queries adds unnecessary complexity compared to the native knowledge source attachment in option A.

Option C is incorrect because using Azure OpenAI externally and routing to agent REST endpoints bypasses Copilot Studio's native A2A agent connection framework. Specialized agents in Copilot Studio do not expose REST endpoints for direct invocation in the way this option implies — they are invoked through the platform's agent connection mechanism. This approach would require significant custom development and would lose native context propagation.

Option D is incorrect because collapsing all domain knowledge into a single agent eliminates the architectural benefits of specialized agents — separate security contexts, independent lifecycle management, and specialization-optimized knowledge sources. Condition-node routing based on entity extraction is also a rule-based approach that suffers from the same limitations as trigger phrases when messages don't fit expected entity patterns.

Exam tip: The multi-agent hub-and-spoke pattern requires three setup steps on the front-door agent: (1) enable generative orchestration, (2) register specialized agents as agent connections with descriptive domain descriptions, (3) attach fallback knowledge sources. The language model uses agent descriptions for routing decisions — write these descriptions carefully. 'Benefits agent helps with health insurance, dental, vision, 401k, and FSA questions' is better than 'handles benefits.'