core practice · system design prompt
Design exercise: Agent Communication & Interfaces
Typed messages, events, artifacts, streaming updates, human checkpoints, agent-to-agent protocols, and UI status. Implement designOutline() returning non-empty values for: messageSchema, statusModel, provenance. Each value must name a concrete mechanism or decision.
The mechanism
Reliable agent interfaces exchange typed state and evidence, not vague prose. Messages need identity, intent, status, result, error, and provenance.
Expected evidence
A design outline with messageSchema, statusModel, provenance plus an explicit failure mode or trade-off.
Try this drill in the practice workspace →
Hints when you need them
- Name the failure mode before selecting the mechanism.