A travel booking company wants to deploy an assistant through Azure AI Foundry that can answer general travel questions and also check a customer's real-time flight status by calling the airline's existing REST API. Which Foundry capability should the team use to let the model invoke that external API during a conversation?

Configuring an agent with a tool definition that lets the model call the flight-status function during the conversation is correct because it lets a Foundry-deployed model recognize when a request needs live data, hand off relevant parameters, and receive the airline system's actual response before continuing. The model's knowledge comes from training data frozen at some point in the past, which can explain baggage policy but cannot capture specific flight status right now because it changes by the minute. Increasing the model's maximum output token limit so it can describe how to check flight status in more detail is incorrect because longer output limits only let the model write more text; they do not give the model any way to reach external systems, so it would still be guessing at flight status. Switching to a larger foundation model in the catalog that has more general aviation knowledge is incorrect because a larger model has broader general knowledge from training but no access to real-time data feeds, so it cannot know specific flight status regardless of size. Adding more example conversations to the system message describing typical flight-delay scenarios is incorrect because examples can shape the model's tone or explain typical scenarios, but examples alone do not create a connection to a live airline system; the model would still be fabricating plausible-sounding answers instead of reporting fact. Only the tool-calling configuration bridges the model's language ability with the airline's authoritative, real-time source of truth, which is what this scenario requires.