Session 8· 12 questions

End-of-session quiz

What you'll learn

12 questions across single agents, memory, streaming, structured output, HITL, and multi-agent. Aim for 9+ before heading to the capstone projects.

Single agent (00–03)

Knowledge Check

What does create_agent give you that Session 6's manual code did not?

Code Check

How does @wrap_model_call middleware change available tools?

Knowledge Check

When a tool raises an exception, what should @wrap_tool_call return?

Code Check

What two things do you need for multi-turn memory?

Knowledge Check

What does SummarizationMiddleware do?

Code Check

Which stream_mode shows individual tokens?

Code Check

Where does the structured result live when using response_format?

Knowledge Check

What are the three human decisions during an HITL interrupt?

Code Check

How does a handoff tool switch agents?

Knowledge Check

Why do subagents start with a clean context on every invocation?

Knowledge Check

Which multi-agent pattern would you use for a customer support bot with triage → specialist → billing stages?

Scored 9+ out of 12?

You are ready for the capstone projects. Apply everything from Sessions 1–8 in the BookingAgent and InterviewerBOT.