Question 61
CCA-F is a newly released exam. These practice questions reflect our best current understanding — the exact wording and question formats may change in the live exam until the certification matures. Use them to learn the concepts, not to memorize.
Scenario: Agent Architecture and Orchestration - An engineer builds an agent loop that calls Claude with tools. To decide when the task is finished, the loop scans the assistant's text for the phrase "task complete" and also stops after a fixed `max_iterations=5`. In production the agent sometimes halts mid-task and other times keeps looping after it is actually done. What is the correct completion signal for the agent loop?
- AKeep the text scan but raise `max_iterations` to 15 so the agent has more room to finish.
- BContinue the loop while `stop_reason == "tool_use"` and stop when `stop_reason == "end_turn"`.
- CStop as soon as the assistant returns any text content block alongside tool calls.
- DAsk the model at each step to self-report a boolean `is_done` field and stop when it is true.