HITL Overview

Let your agent and your React Native users work through tasks together.

What is this?

Human-in-the-loop (HITL) lets an agent pause mid-run to collect input, confirmation, or a choice from the user, then resume with that answer folded back into its reasoning. It turns an autonomous workflow into a collaborative one: the agent keeps its context, and the user keeps the steering wheel.

When should I use this?

Use HITL when you need:

  • Quality control: a human gate at high-stakes decision points.
  • Missing input: the agent needs a value only the user knows, such as a meeting time.
  • Risky actions: the agent must get a yes before it sends, pays, or deletes.
  • Expert input: the user supplies knowledge the model lacks.

Two patterns for HITL in CopilotKit

CopilotKit has two ways to pause an agent turn and ask the user something. They look alike from the outside (the chat pauses, your component appears, the user answers, the run resumes), but they're wired differently on the backend.

PatternWho decides to pause?Backend surface
useHumanInTheLoopThe LLM, by calling a tool your app registeredA frontend-only tool (schema + render)
useInterruptYour agent code, at a step that must stop for a personAn interrupt your agent emits, for example LangGraph interrupt() or Mastra suspend()

Pick useHumanInTheLoop when the model decides to ask the user, and you want the question to appear in the normal tool-call flow.

Pick useInterrupt when the pause is a checkpoint in your agent code: that code path always needs a human answer, and the server-side contract is an interrupt. On the Built-in Agent, interrupts are temporarily unavailable, so use useHumanInTheLoop there.

Install the OpenAI Agents AG-UI adapter

pip install ag-ui-openai-agents openai-agents

Ask for the human with a frontend tool

The simplest route is an ordinary frontend tool. Register it in the browser with useHumanInTheLoop, or with useFrontendTool and an async handler that resolves once the user has answered. The adapter hands the tool to the model on every run. When the model calls it, the run stops and the browser asks the user. The answer goes back with the next request as the call's result, and the model continues from there. Nothing waits on the server in between.

Name the tool in the agent's instructions

That leaves the agent with one job: knowing the tool exists and when to call it. A model that is never told about book_call will answer in prose and the picker will never mount.

hitl.py
from agents import Agent

from demo_agents._common import MODEL, chat_settings

HITL_IN_CHAT_PROMPT = (
    "You help users book an onboarding call with the sales team. "
    "When they ask to book a call, call the frontend-provided "
    "`book_call` tool with a short topic and the user's name. "
    "Keep any chat reply to one short sentence."
)


hitl_in_chat_agent = Agent(
    name="booking_assistant",
    instructions=HITL_IN_CHAT_PROMPT,
    model=MODEL,
    model_settings=chat_settings(),
)

Or gate a server tool with needs_approval

When the backend decides it needs the user, mark the tool needs_approval=True. The SDK pauses before the tool runs, and the adapter ends the AG-UI run with an interrupt for that call, which useInterrupt renders. The interrupt's reason is "tool_approval", its metadata carries tool_name and the parsed tool_input, and it asks for {"approved": boolean}.

interrupt_agent.py
@function_tool(needs_approval=True)
def schedule_meeting(ctx: ToolContext[Any], topic: str, attendee: str = "") -> str:
    """Ask the user to pick a meeting time, then confirm what was scheduled.

    Args:
        topic: Short description of the meeting purpose.
        attendee: Who the meeting is with, if known.
    """
    # Runs once the user approved, with the time they picked beside it.
    answer = _answers.get({}).get(ctx.tool_call_id, {})
    label = answer.get("chosen_label") or answer.get("chosen_time")
    if not label:
        return f"User did not pick a time. Meeting NOT scheduled: {topic}"
    return f"Meeting scheduled for {label}: {topic}"

The next request resumes the interrupt. {"approved": true} runs the tool before the model is called again. Any other answer gives the model the SDK's rejection text, and a cancelled interrupt answers the call as cancelled. If the user sends a new message instead of deciding, the approval stays open: the run ends with the same interrupt again, and the new message reaches the model once the approval is decided.

Hand the tool more than a yes or no

An Agents SDK approval is only a yes or a no, so an approved tool runs with its own arguments and nothing else. The showcase's picker sends the chosen time beside approved. A small OpenAIAgentsAgent subclass keeps each request's answers, by tool call id, where schedule_meeting can read them:

interrupt_agent.py
class SchedulerAgent(OpenAIAgentsAgent):
    """Lets ``schedule_meeting`` read the time the user picked.

    The adapter runs an approved call with its arguments alone: an Agents SDK
    approval is a yes or a no, with nothing to hand the tool. The picker
    sends the time beside the approval, so this keeps each answer where the
    tool can read it.
    """

    async def run_streamed(self, input: RunAgentInput) -> AsyncIterator[BaseEvent]:
        _answers.set(
            {
                entry.interrupt_id: entry.payload
                for entry in input.resume or []
                if isinstance(entry.payload, dict)
            }
        )
        async with aclosing(super().run_streamed(input)) as events:
            async for event in events:
                yield event

Nothing is kept on the server

The paused call lives in the conversation the page holds, and the server trusts the history it is sent. Replaying the same resume request runs an approved tool again, so key side effects on ToolContext.tool_call_id where a retry must not repeat them, and authorize the decision yourself if that matters.

Pattern 1: useHumanInTheLoop (tool-based)

Register a HITL tool with useHumanInTheLoop. When the model calls it, CopilotKit renders your component for the call. While the call is running, the component gets a respond callback. The agent receives the answer as the tool result and continues from there.

src/MeetingApproval.tsx
import { Pressable, Text, View } from "react-native";
import { useHumanInTheLoop } from "@copilotkit/react-native/headless";
import { z } from "zod";

const SLOTS = [
  { iso: "2026-11-17T09:00:00Z", label: "Tue 9 AM" },
  { iso: "2026-11-18T14:00:00Z", label: "Wed 2 PM" },
];

export function useMeetingApproval() {
  useHumanInTheLoop({
    name: "schedule_meeting",
    description: "Ask the user to pick a meeting time.",
    parameters: z.object({
      topic: z.string(),
      attendee: z.string(),
    }),
    render: ({ args, respond, result }) => (
      <View style={{ padding: 12, borderRadius: 12, backgroundColor: "#f4f4f8", gap: 8 }}>
        <Text style={{ fontWeight: "600" }}>
          Pick a time for {args.topic ?? "the meeting"}
        </Text>
        {result ? (
          <Text>Done: {result}</Text>
        ) : (
          <>
            {SLOTS.map((slot) => (
              <Pressable
                key={slot.iso}
                disabled={!respond}
                onPress={() => respond?.({ chosen_time: slot.iso, chosen_label: slot.label })}
              >
                <Text>{slot.label}</Text>
              </Pressable>
            ))}
            <Pressable disabled={!respond} onPress={() => respond?.({ cancelled: true })}>
              <Text>None of these work</Text>
            </Pressable>
          </>
        )}
      </View>
    ),
  });
}

respond is set only while the call is running. Before that, the arguments are still streaming in, and args can be partial. After the user answers, result holds the tool result as a string.

Call useMeetingApproval() in a component under CopilotKitProvider. The slots are data your app owns, so you can swap in real availability or a calendar API.

Where the card renders

The rendered CopilotChat and CopilotModal from @copilotkit/react-native/components draw the card in the conversation. The root CopilotChat, CopilotPopup, and CopilotSidebar don't draw messages, so in your own chat, render tool calls with useRenderToolCall. See Headless UI.

Ask the agent to book a meeting with Alex about the launch. The card appears in the chat. Tap a slot, and the agent confirms the time you picked.

Pattern 2: useInterrupt (agent-paused)

With an interrupt, your agent code enforces the pause, for example with LangGraph's interrupt():

  1. A step calls interrupt({...}), and the run stops.
  2. The app receives the payload and renders your UI.
  3. The user answers, and the run resumes with the answer.

CopilotKit's useInterrupt hook is the render contract. The rendered CopilotChat from /components draws its element after the last message.

See the useInterrupt deep dive for the full walkthrough, including the backend and the render callback.

Going headless

Both patterns can drive UI outside the chat. For interrupts, set renderInChat: false and place the element that useInterrupt returns. For a tool-based pause, a frontend tool's Promise can drive your own app state. See the headless interrupts guide.

How is this guide?