A team is building an app that must take a user question, retrieve relevant passages from a company knowledge base, then pass both the question and retrieved passages to an LLM to produce a grounded answer. They want to visually design, test, and iterate on these connected steps before deploying. Which Azure AI Foundry tool fits this need?
Select an answer to reveal the explanation.
Short Explanation
What this team needs is a way to sketch out a sequence of steps, first go fetch the relevant background material, then hand that material along with the original question to the model, and be able to test and tweak that whole chain before it goes live. A tool for configuring what content gets filtered out is solving a different problem entirely, since it is about moderation, not about wiring steps together. Browsing and filtering a list of available models helps you pick which brain to use, but it says nothing about how the pieces of your pipeline connect to each other. And a dashboard that shows you how much money you are spending is useful for budgeting, but it has nothing to do with designing the logic of your app. The tool built specifically for visually connecting steps like retrieval and generation, and letting you run and iterate on that flow before shipping it, is the right fit here.
Full Explanation
The correct answer is B. Prompt flow is Foundry's tool for visually building, connecting, and testing a sequence of steps such as retrieval followed by generation, letting the team iterate on how the question and retrieved passages flow into the LLM before deploying the app. Option A is incorrect because Content Safety Studio is focused on configuring and testing content moderation filters, not on wiring together retrieval and generation steps into an orchestrated pipeline. Option C is incorrect because the model catalog's search and filter tools only help the team locate and compare candidate models to use; they play no role in defining or testing the multi-step logic that connects retrieval to generation. Option D is incorrect because the cost management dashboard tracks spending across deployments and resources, which is unrelated to designing or debugging the sequence of steps an application executes at runtime. Prompt flow is purpose-built for exactly this kind of multi-step, retrieval-plus-generation orchestration and testing workflow.