HITL Overview

Allow your agent and users to collaborate on complex tasks.

In-Chat HITL (useHumanInTheLoop)

Waking up the demo

See this in Inspector

Open Inspector on localhost. Go to Inspect, then Frontend Tools. Your tool and its schema are listed.

More detail: Inspector.

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's what turns an autonomous workflow into a collaborative one: the agent keeps its context, 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 domain knowledge the model lacks.

Two patterns for HITL in CopilotKit

CopilotKit ships two complementary ways to pause an agent turn and ask the human something. They look similar from the outside (the chat pauses, a custom component appears, the user answers, the run resumes) but they're wired differently on the backend, and each has its own niche.

PatternWho decides to pause?Backend surface
useHumanInTheLoopThe LLM, by calling a registered client-side toolA frontend-only tool description (Zod 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 pause is an agent-initiated decision (the model chose to ask the user) and you want the picker UI inlined into the normal tool-call flow.

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

useHumanInTheLoop: no frontend-tool code on the agent

AGUIStream turns every tool the client declares into a client-side tool, so a tool you register with useHumanInTheLoop reaches the model through the AG-UI request. Do not declare it on the AG2 agent. Name it in the agent's prompt so the model calls it.

To keep the user's answer, have a backend tool write it to context.variables. When a tool changes the variables, AGUIStream sends a STATE_SNAPSHOT at the end of the run:

agent.py
from typing import Annotated

from fastapi import FastAPI, Header
from fastapi.responses import StreamingResponse
from pydantic import Field

from ag2 import Agent, Context, tool
from ag2.ag_ui import AGUIStream, RunAgentInput
from ag2.config import OpenAIResponsesConfig

@tool
def store_user_choice(
    choice: Annotated[str, Field(description="The option selected by the user")],
    context: Context,
) -> str:
    """Store the latest user choice in shared state."""
    context.variables["hitl"] = {"latest_choice": choice}
    return f"Stored choice: {choice}"

agent = Agent(
    name="assistant",
    prompt=(
        "When you need the user to choose, call the frontend tool `offerOptions`. "
        "After it returns, call `store_user_choice` with the selected value."
    ),
    config=OpenAIResponsesConfig(model="gpt-6-luna"),
    tools=[store_user_choice],
)

stream = AGUIStream(agent)
app = FastAPI()

@app.post("/chat")
async def run_agent(
    message: RunAgentInput,
    accept: str | None = Header(None),
) -> StreamingResponse:
    return StreamingResponse(
        stream.dispatch(message, accept=accept),
        media_type=accept or "text/event-stream",
    )

Approve a tool call with interrupts

AG2 can also pause a run on the backend. A tool wrapped in ApprovalRequired is held until the user approves or rejects the call. AG2 sends the held call to the client as an AG-UI interrupt, and CopilotKit's useInterrupt renders it.

tools.py
from typing import Annotated

from pydantic import Field

from ag2 import Context, tool
from ag2.middleware.builtin.tools.approval import ApprovalRequired

@tool(
    description="Save a city to the user's favorites. The user confirms before anything is saved.",
    middleware=[
        ApprovalRequired("Save this city to your favorites?\n{tool_arguments}", allow_always=False)
    ],
)
async def save_favorite_city(
    context: Context,
    city: Annotated[str, Field(description="The city name to save, e.g. 'Lisbon'")],
) -> list[str]:
    favorites = list(context.variables.get("favorites", []))
    if city not in favorites:
        favorites.append(city)
    context.variables["favorites"] = favorites
    return favorites

Serve the agent with AGUIStream as usual. Do not set a hitl_hook on the agent. With one, AG2 asks in-process and no interrupt reaches the browser.

The run ends with an interrupt of reason tool_call. The backend reads the answer as a boolean, so resolve(true) approves the call and resolve(false) refuses it. A refusal does not fail the run: the model is told the call was refused and carries on.

page.tsx
import { useInterrupt } from "@copilotkit/react-core/v2";

export function Page() {
  useInterrupt({
    render: ({ interrupt, resolve }) => (
      <div>
        <p>{interrupt?.message ?? "The assistant needs your confirmation."}</p>
        <button onClick={() => resolve(true)}>Approve</button>
        <button onClick={() => resolve(false)}>Reject</button>
      </div>
    ),
  });

  // ...
}

Ask Save Lisbon to my favorites. The approval card appears in the chat. Approve it, and the city shows up in the shared state.

useInterrupt().resolve() and .cancel() send no metadata in the resume entry. That is enough for a gated tool call like the one above, but not for an AG2 server built with require_resume_proof=True, which needs the interrupt's metadata copied into the resume entry. See Asking the human a question in the AG2 docs.

A held turn lives in the AG2 process that served the run, so the resuming run must reach that same process. See the AG2 deployment notes before you run several replicas.

Pattern 1: useHumanInTheLoop (tool-based)

Register a HITL tool in your React app with useHumanInTheLoop. When the LLM calls that tool, CopilotKit routes the call through your render function, which shows a custom component and calls respond with the user's answer. The agent sees the answer as the tool result and continues from there.

page.tsx
import React from "react";
import {
  CopilotKit,
  CopilotChat,
  useHumanInTheLoop,
  useConfigureSuggestions,
} from "@copilotkit/react-core/v2";
import { z } from "zod";
import { TimePickerCard, TimeSlot } from "./time-picker-card";

const DEFAULT_SLOTS: TimeSlot[] = [
  { label: "Tomorrow 10:00 AM", iso: "2026-04-19T10:00:00-07:00" },
  { label: "Tomorrow 2:00 PM", iso: "2026-04-19T14:00:00-07:00" },
  { label: "Monday 9:00 AM", iso: "2026-04-21T09:00:00-07:00" },
  { label: "Monday 3:30 PM", iso: "2026-04-21T15:30:00-07:00" },
];

export default function HitlInChatDemo() {
  return (
    <CopilotKit runtimeUrl="/api/copilotkit" agent="hitl-in-chat">
      <div className="flex justify-center items-center h-screen w-full">
        <div className="h-full w-full max-w-4xl">
          <Chat />
        </div>
      </div>
    </CopilotKit>
  );
}

function Chat() {
  useConfigureSuggestions({
    suggestions: [
      {
        title: "Book a call with sales",
        message:
          "Please book an intro call with the sales team to discuss pricing.",
      },
      {
        title: "Schedule a 1:1 with Alice",
        message: "Schedule a 1:1 with Alice next week to review Q2 goals.",
      },
    ],
    available: "always",
  });

  useHumanInTheLoop({
    agentId: "hitl-in-chat",
    name: "book_call",
    description:
      "Ask the user to pick a time slot for a call. The picker UI presents fixed candidate slots; the user's choice is returned to the agent.",
    parameters: z.object({
      topic: z
        .string()
        .describe("What the call is about (e.g. 'Intro with sales')"),
      attendee: z
        .string()
        .describe("Who the call is with (e.g. 'Alice from Sales')"),
    }),
    render: ({ args, status, respond }: any) => (
      <TimePickerCard
        topic={args?.topic ?? "a call"}
        attendee={args?.attendee}
        slots={DEFAULT_SLOTS}
        status={status}
        onSubmit={(result) => respond?.(result)}
      />
    ),
  });

The sample imports TimePickerCard from its own file. The card renders the slots and calls onSubmit with the user's choice. These are its props:

time-picker-card.tsx
export interface TimePickerCardProps {
  topic: string;
  attendee?: string;
  slots: TimeSlot[];
  status: TimePickerStatus;
  onSubmit: (
    result: { chosen_time: string; chosen_label: string } | { cancelled: true },
  ) => void;
}

Each slot button calls onSubmit({ chosen_time: s.iso, chosen_label: s.label }). The "None of these work" button calls onSubmit({ cancelled: true }). The full file is in the Code view of the demo at the top of this page.

The picker UI is fed a static list of candidate slots. This is just data the demo page owns, so you can swap in real availability, a calendar API, or anything else:

page.tsx
import React from "react";
import {
  CopilotKit,
  CopilotChat,
  useHumanInTheLoop,
  useConfigureSuggestions,
} from "@copilotkit/react-core/v2";
import { z } from "zod";
import { TimePickerCard, TimeSlot } from "./time-picker-card";

const DEFAULT_SLOTS: TimeSlot[] = [
  { label: "Tomorrow 10:00 AM", iso: "2026-04-19T10:00:00-07:00" },
  { label: "Tomorrow 2:00 PM", iso: "2026-04-19T14:00:00-07:00" },
  { label: "Monday 9:00 AM", iso: "2026-04-21T09:00:00-07:00" },
  { label: "Monday 3:30 PM", iso: "2026-04-21T15:30:00-07:00" },
];

Pattern 2: useInterrupt (agent-paused)

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

  1. A node calls interrupt({...}), and the run stops.
  2. The client 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.

See the useInterrupt deep dive for the full walkthrough, including the backend tool and render-prop wiring.

Going headless

Both patterns above ship with a render prop. CopilotKit handles the "when to show the picker" logic for you. If you want to drive interrupt resolution from a custom UI that lives anywhere in the tree (not necessarily inside a chat), see the headless interrupts guide. It shows how to compose useAgent, agent.subscribe, and copilotkit.runAgent to build your own useInterrupt equivalent.

How is this guide?