Build fluency in the core building blocks of the OpenAI Agents SDK.
0 / 5 completed
1 / 5
At standup, a dev wants to define an autonomous unit with instructions and tools in the OpenAI Agents SDK. Which primitive fits?
An Agent in the OpenAI Agents SDK bundles a name, instructions, model, and a set of tools into one reusable unit. It is the core building block you configure and invoke. Other concepts like runs and handoffs operate on top of Agents.
2 / 5
During a design review, the team wants one agent to delegate a task to a specialist agent. Which mechanism enables this?
A handoff lets one agent transfer control of the conversation to another agent better suited for the sub-task. The receiving agent takes over with the existing context. This composes specialized agents into a larger workflow.
3 / 5
In a code review, a dev validates user input before the agent runs. Which SDK feature is designed for this?
A guardrail runs a check (often another lightweight agent or function) against input or output to catch problems early, like unsafe requests. Guardrails can halt execution before wasting a full agent run. They are the SDK's built-in safety mechanism.
4 / 5
An incident report shows a dev couldn't tell which tool call caused a bad output. Which SDK feature would help diagnose this?
The Agents SDK ships built-in tracing that records each step, tool call, and handoff during a run for inspection. Reviewing a trace pinpoints exactly which tool call produced the faulty output. This observability is essential once agents chain multiple steps.
5 / 5
During a PR review, a teammate wants the agent to remember prior turns across calls automatically. Which concept manages this?
A session persists conversation history so subsequent runs automatically include prior turns without manual bookkeeping. This keeps multi-turn agent interactions stateful. It removes the need to manually thread messages between calls.