Human Approval as a System Boundary
Design human approval around an exact operation, with reviewable payloads, durable decisions, expiry, and execution checks that survive retries.
Tool-Approval
Use approval for side effects, not for every tool call
Reserve human approval for actions where a wrong call would actually matter.
Human-in-the-Loop Agents: When the AI Must Ask Before Acting
How to implement human-in-the-loop approval flows in Microsoft Agent Framework for risky tool calls, side effects, workflow state, and resumable review.