Designs agents with human approval checkpoints. Expected output is a workflow with escalation logic. Prompting style: structured, agentic.
Design an AI agent workflow that includes human-in-the-loop checkpoints for the task below. Specify when human review is required and why.
Task:
{{task}}