Quiz — Session 3 (Advanced): Claude Agent SDK

Program : Applied AI — Yann Isola 10 MCQs (Multiple Choice Questions) — only one correct answer per question, unless otherwise stated. Level: preparation Claude Certified Architect . Correction commented at the end of the document.

Question 1 — In the Claude Agent SDK (Software Development Kit), which component executes the agentic loop (sending messages, executing tool calls, iterating until the final response)?

HAS The class Agent B The decorator @tool C THE Runner D The hook on_agent_start

Question 2 — When defining a tool with `@tool` , where the description transmitted to the model?

HAS From the function name only B From the function docstring C From a parameter description= obligatory from the decorator D It is automatically generated by the model on first use

Question 3 — How does the SDK build the parameter diagram of a tool?

HAS From the type annotations (type hints) of the Python signature B By analyzing the function body at runtime C You have to write the JSON (JavaScript Object Notation) schema by hand D Parameters are always passed as a single string

Question 4 — What is the fundamental difference between a handoff and a tool call ?

HAS The handoff is faster because it avoids a call to the model B Tool call transfers the complete history, handoff only the arguments C After a handoff, the target agent maintains control of the conversation; after a tool call, the calling agent takes control D There is no difference: the handoff is a tool call alias

Question 5 — One input guardrail triggers his tripwire. What is happening?

HAS The message is automatically reformulated then sent back to the model B The run is interrupted and a dedicated exception (InputGuardrailTripwireTriggered ) is raised, to be intercepted by the application C The agent automatically switches to a more secure model D The guardrail is disabled for the rest of the run

Question 6 — A coordinator must be able to delegate work to subagents in the Claude runtime environment. What configuration is essential ?

HAS max_turns greater than 50 B An output guardrail on each sub-agent C The tool "Task" present in the allowedTools of the coordinator D The same model for the coordinator and subagents

Question 7 — What is the role of context variables (generic parameter `Agent[MonContexte]`) ?

HAS Inject additional text into the system prompt B Share a typed state between agents, tools, guardrails and hooks, without expose it to the model C Cache model responses between two runs D Store chat history persistently

Question 8 — You run five independent analyzes in parallel with `asyncio.gather(...)`. What option prevents a single failure from canceling all four successes?

HAS timeout=None B return_exceptions=True C max_turns=1 D Wrap each task in an output guardrail

Question 9 — A trainer shows this delegation code: `Runner.run(agent_redacteur, "Rédige la section 2.")`. Why is this an anti-pattern?

HAS Runner.run cannot be called on a subagent B The subagent does not share the coordinator's working memory: without a plan, tone, context and deliverable expected in the prompt, the output will be off-topic C You should always use run_async for subagents D Prompt string is too short to charge

Question 10 — A tool `initier_remboursement` can be retried after a timeout. Which architectural precaution has priority?

HAS Increase timeout to 10 minutes B Remove tool from list and replace with instructions C Make the tool idempotent (idempotence key) so that a retry does not trigger two refunds D Entrust the tool to a more powerful model

Quiz — Session 3 (Advanced): Claude Agent SDK

Question 1 — In the Claude Agent SDK (Software Development Kit), which component executes the agentic loop (sending messages, executing tool calls, iterating until the final response)?

Question 2 — When defining a tool with @tool , where the description transmitted to the model?

Question 3 — How does the SDK build the parameter diagram of a tool?

Question 4 — What is the fundamental difference between a handoff and a tool call ?

Question 5 — One input guardrail triggers his tripwire. What is happening?

Question 6 — A coordinator must be able to delegate work to subagents in the Claude runtime environment. What configuration is essential ?

Question 7 — What is the role of context variables (generic parameter Agent[MonContexte]) ?

Question 8 — You run five independent analyzes in parallel with asyncio.gather(...). What option prevents a single failure from canceling all four successes?

Question 9 — A trainer shows this delegation code: Runner.run(agent_redacteur, "Rédige la section 2."). Why is this an anti-pattern?

Question 10 — A tool initier_remboursement can be retried after a timeout. Which architectural precaution has priority?

Question 2 — When defining a tool with `@tool` , where the description transmitted to the model?

Question 7 — What is the role of context variables (generic parameter `Agent[MonContexte]`) ?

Question 8 — You run five independent analyzes in parallel with `asyncio.gather(...)`. What option prevents a single failure from canceling all four successes?

Question 9 — A trainer shows this delegation code: `Runner.run(agent_redacteur, "Rédige la section 2.")`. Why is this an anti-pattern?

Question 10 — A tool `initier_remboursement` can be retried after a timeout. Which architectural precaution has priority?