AI-103 Certification Practice Question #30

Question

You have a Microsoft Foundry project that contains a deployed ticket-triage agent.
You discover that sometimes the agent responds without calling any tools, even when a tool is required.
You need to ensure that the agent calls a tool during execution.
How should you complete the Python code? To answer, drag the appropriate values to the correct targets. Each value may be used once, more than once, or not at all. You may need to drag the split bar between panes or scroll to view content.
NOTE: Each correct selection is worth one point.

Accepted Answer

The agent sometimes answers without calling a required tool because tool selection defaults to `auto`, letting the model decide. MS Learn's tool-calling guidance is explicit: for deterministic tool invocation, set `tool_choice` to `required`, which forces the model to call one or more tools. In the `run_payload` dictionary the single empty key:value pair must therefore be `"tool_choice": "required"` — the key `"tool_choice"` goes in the left slot and the value `"required"` in the right. The other tiles (`"tools"`, `"response_format"`, `"type"`, `"auto"`) do not force a tool call (`"auto"` is in fact the cause of the failure). Confidence is 85% — there is no answer image, Suggested Answer, or community vote, but the MS Learn mapping of "must call a tool → `tool_choice: required`" is unambiguous and the image layout (one key:value slot inside the run payload) leaves only this assignment.

More AI-103 practice questions