Pausing the Agent for Input

Pause an agent run mid-tool, hand control to a custom React component, and resume with the user's answer.

In-Chat HITL (useInterrupt — low-level primitive)

Waking up the demo

What is this?

Some agent steps need a person's answer before the run can continue, such as approving a payment or picking a meeting time. useInterrupt pauses the run, shows your React component, and resumes the run with the answer. How the pause works depends on your agent framework.

This framework ships a first-class interrupt primitive that lets running work suspend itself and hand control to the client: interrupt() in LangGraph, interrupts in AWS Strands, a tool that returns suspend() in Mastra, or a @human_feedback step in a CrewAI Flow. The run is frozen server-side until the client resolves the interrupt with a payload. Then the run resumes, and the agent receives that payload as the answer.

CopilotKit's useInterrupt is the frontend half of that contract: it subscribes to the paused run, renders whatever component you give it, and calls the agent back with the user's answer.

When should I use this?

Reach for useInterrupt when the pause is a checkpoint in your agent code where the code path must stop and wait for a human, not an LLM-initiated tool call. Typical cases:

  • A sensitive action (payments, irreversible writes) must be approved
  • A required piece of state isn't known and can only be collected from the user
  • The agent explicitly reaches an approval node in a longer workflow
  • You want the server-side contract to be an interrupt that resumes with a payload

For LLM-initiated pauses where the model decides on the fly to ask the user, prefer useHumanInTheLoop.

The backend: pause the run in your agent

Pause a tool with Strands' native interrupt

AWS Strands ships a first-class interrupt primitive. A tool's callback receives a context whose interrupt({ name, reason }) call halts the agent loop and hands reason to the client as the interrupt payload. The AG-UI adapter finishes the run with RUN_FINISHED carrying outcome.type === "interrupt".

src/agent/interrupt-agent.ts
export const scheduleMeeting = tool({
  name: "schedule_meeting",
  description:
    "Ask the user to pick a meeting time, then confirm what was scheduled.",
  inputSchema: z.object({
    topic: z.string().describe("Short description of the meeting purpose."),
    attendee: z.string().optional().describe("Who the meeting is with."),
  }),
  callback: ({ topic, attendee }, context) => {
    // Typed optional by the SDK, so this is checked rather than asserted: with
    // no context there is nothing to pause on, and pretending otherwise would
    // schedule a meeting the user never saw.
    if (!context) {
      throw new Error("schedule_meeting needs a tool context to pause on");
    }

    // `attendee` is optional and the reason has to be JSON, which has no
    // `undefined`, so it is omitted rather than sent as undefined.
    const answer = context.interrupt<ResumeEnvelope>({
      name: "schedule_meeting",
      reason: attendee === undefined ? { topic } : { topic, attendee },
    });

    // Three cancel shapes reach here: each bridge's own sentinel for a
    // cancelled resume entry, and the picker's Cancel button, which resolves
    // with a `cancelled` flag inside the payload.
    const { choice, cancelled } = readResume(answer);
    if (cancelled) {
      return `User cancelled. Meeting NOT scheduled: ${topic}`;
    }

    const label = choice.chosen_label || choice.chosen_time;
    return label
      ? `Meeting scheduled for ${label}: ${topic}`
      : `User did not pick a time. Meeting NOT scheduled: ${topic}`;
  },
});

How the answer reaches the tool depends on the adapter version. The pinned @ag-ui/aws-strands 0.2.3 hands the client's payload through untouched and signals a cancel as { status: "cancelled" }; the Python adapter wraps an answer as { response: ... } and cancels with { cancelled: true }. Read both shapes, or a picked slot comes back to the model as though the user never picked one.

Keep the pausing tool off a client-executed name

useHumanInTheLoop registers its tool on the FRONTEND, so a name used there cannot also be a pausing backend tool. This showcase mounts a dedicated interrupt agent and points the interrupt demos' agent names at it, leaving schedule_meeting on the shared agent free for the frontend-tool flow.

Resume in the same process, or across a restart

Pause and resume on the same running process need no extra wiring. For a resume that survives a restart, give the agent a Strands SessionManager through StrandsAgentConfig.sessionManagerProvider; the adapter persists its interrupt checkpoint into that session.

In the tool or step that needs the user's answer, call your framework's interrupt primitive with the data your UI needs: interrupt() in LangGraph, suspend() in Mastra, or a @human_feedback step in a CrewAI Flow. The example agent does this in its schedule_meeting tool:

interrupt-agent.ts
export const scheduleMeeting = tool({
  name: "schedule_meeting",
  description:
    "Ask the user to pick a meeting time, then confirm what was scheduled.",
  inputSchema: z.object({
    topic: z.string().describe("Short description of the meeting purpose."),
    attendee: z.string().optional().describe("Who the meeting is with."),
  }),
  callback: ({ topic, attendee }, context) => {
    // Typed optional by the SDK, so this is checked rather than asserted: with
    // no context there is nothing to pause on, and pretending otherwise would
    // schedule a meeting the user never saw.
    if (!context) {
      throw new Error("schedule_meeting needs a tool context to pause on");
    }

    // `attendee` is optional and the reason has to be JSON, which has no
    // `undefined`, so it is omitted rather than sent as undefined.
    const answer = context.interrupt<ResumeEnvelope>({
      name: "schedule_meeting",
      reason: attendee === undefined ? { topic } : { topic, attendee },
    });

    // Three cancel shapes reach here: each bridge's own sentinel for a
    // cancelled resume entry, and the picker's Cancel button, which resolves
    // with a `cancelled` flag inside the payload.
    const { choice, cancelled } = readResume(answer);
    if (cancelled) {
      return `User cancelled. Meeting NOT scheduled: ${topic}`;
    }

    const label = choice.chosen_label || choice.chosen_time;
    return label
      ? `Meeting scheduled for ${label}: ${topic}`
      : `User did not pick a time. Meeting NOT scheduled: ${topic}`;
  },
});

The run stops here until the client resolves the interrupt. The resolve payload comes back to the agent as the answer: the return value of interrupt() in LangGraph, resumeData in Mastra, or the HumanFeedbackResult that the next step of a CrewAI Flow receives. The example turns that value into a string for the model.

Two things to note:

  • The payload ({"topic": topic, "attendee": attendee}) is what the frontend reads off the interrupt. Keep it a plain, serializable object. It's the "pause-time context" the UI needs to render.
  • The return-side contract ({chosen_label, chosen_time} or {cancelled: true}) is entirely yours. The client can send anything as the resolve payload; the tool is the one that gives it meaning.

The frontend: useInterrupt render prop

On the client you register a useInterrupt hook per agent. When the paused run arrives, render receives both the raw event and the interrupt itself, and resolve(...) is how you resume the run. Your agent framework decides where the payload sits:

  • Custom-event interrupt (on_interrupt): event.value.
  • Standard AG-UI interrupt: interrupt.metadata, or a JSON string in interrupt.message.
page.tsx
import {
  CopilotKit,
  CopilotChat,
  useInterrupt,
} from "@copilotkit/react-core/v2";
import type { TimeSlot } from "./_components/time-picker-card";
import { TimePickerCard } from "./_components/time-picker-card";
import { generateFallbackSlots } from "../_shared/interrupt-fallback-slots";
import { useGenUiInterruptSuggestions } from "./suggestions";

// Shape the backend `schedule_meeting` tool pauses with: its `interrupt()`
// `reason` payload. `slots` is absent on the Strands path (the tool passes only
// topic and attendee), so the picker falls back to generated slots.
type SchedulingPayload = {
  topic?: string;
  attendee?: string;
  slots?: TimeSlot[];
};

// Read the tool's `interrupt()` reason off an AG-UI interrupt.
//
// The two bridges expose it on different channels: `ag_ui_strands` (Python)
// carries the reason object under `metadata.reason`, while the published
// `@ag-ui/aws-strands` 0.2.3 JSON-encodes it into `message` instead. Both are
// read so one page serves both, and the legacy event value is read last for
// adapters that pass the payload through unwrapped.
/**
 * JSON.parse that never throws and never returns a primitive. Both readers run
 * inside a React render callback, where a throw takes the whole pane down.
 */
function parseObject(raw: string | undefined): Record<string, unknown> | null {
  if (!raw) return null;
  try {
    const parsed: unknown = JSON.parse(raw);
    return parsed && typeof parsed === "object"
      ? (parsed as Record<string, unknown>)
      : null;
  } catch {
    return null;
  }
}

function readSchedulingPayload(
  interrupt: { metadata?: unknown; message?: string } | null | undefined,
  eventValue: unknown,
): SchedulingPayload {
  const metadata = interrupt?.metadata as
    | { reason?: SchedulingPayload }
    | undefined;
  if (metadata?.reason && typeof metadata.reason === "object") {
    return metadata.reason;
  }

  // The published TypeScript bridge JSON-encodes the reason into `message`
  // instead of carrying it on metadata.
  const decoded = parseObject(interrupt?.message);
  if (decoded) {
    const nested = (decoded as { reason?: SchedulingPayload }).reason;
    return nested && typeof nested === "object"
      ? nested
      : (decoded as SchedulingPayload);
  }

  // Legacy channel: some adapters pass the payload through as the event value,
  // JSON-encoded or not.
  const legacy =
    typeof eventValue === "string" ? parseObject(eventValue) : eventValue;
  if (!legacy || typeof legacy !== "object") return {};
  const wrapped = (legacy as { metadata?: { reason?: SchedulingPayload } })
    .metadata?.reason;
  if (wrapped && typeof wrapped === "object") return wrapped;
  return legacy as SchedulingPayload;
}

export default function GenUiInterruptDemo() {
  return (
    <CopilotKit runtimeUrl="/api/copilotkit" agent="gen-ui-interrupt">
      <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() {
  // The failure banner lives OUTSIDE the interrupt element on purpose. The hook
  // caches the element it rendered, so a picker already on screen never sees an
  // updated prop: a flag threaded into the card would keep showing a stale
  // failure on the NEXT booking, before that booking's resume has even started.
  const [resumeFailed, setResumeFailed] = useState(false);

  return (
    <div className="flex h-full flex-col">
      {resumeFailed && (
        <div
          role="alert"
          data-testid="time-picker-resume-error"
          className="m-4 rounded-lg border border-red-200 bg-red-50 p-3 text-sm text-red-800 dark:border-red-900 dark:bg-red-950/40 dark:text-red-300"
        >
          Your previous selection could not be sent. Please try booking again.
        </div>
      )}
      <div className="min-h-0 flex-1">
        <InterruptChat setResumeFailed={setResumeFailed} />
      </div>
    </div>
  );
}

function InterruptChat({
  setResumeFailed,
}: {
  setResumeFailed: (failed: boolean) => void;
}) {
  useGenUiInterruptSuggestions();

  // Native interrupt path. The backend `schedule_meeting` tool calls Strands'
  // `context.interrupt(...)`; the @ag-ui/aws-strands bridge finishes the run
  // with `outcome.type === "interrupt"`. Where the tool's `reason` travels
  // depends on the bridge version, so the reader below accepts both channels.
  // `resolve(...)` resumes the same Strands run, handing the selection back to
  // that `interrupt()` call.
  useInterrupt({
    agentId: "gen-ui-interrupt",
    renderInChat: true,
    render: ({ event, interrupt, resolve }) => {
      const payload = readSchedulingPayload(interrupt, event.value);
      const slots =
        payload.slots && payload.slots.length > 0
          ? payload.slots
          : generateFallbackSlots();
      return (
        <TimePickerCard
          topic={payload.topic ?? "a call"}
          attendee={payload.attendee}
          slots={slots}
          onSubmit={(result) => {
            setResumeFailed(false);
            // Defer resolve so React commits the picked/cancelled badge before
            // useInterrupt clears the interrupt element (a single rAF is not
            // reliable: it can fire before React's commit). The rejection is
            // re-surfaced rather than dropped, because the card has already
            // shown a green "Booked" badge by then and a silently failed
            // resume would leave the user reading a success that never
            // happened.
            window.setTimeout(() => {
              void resolve(result).catch((error: unknown) => {
                setResumeFailed(true);
                queueMicrotask(() => {
                  throw error;
                });
              });
            }, 500);
          }}
        />
      );
    },
  });

Whatever you pass to resolve is sent back to the agent as the answer to the matching interrupt, for example the return value of interrupt(...) in LangGraph.

Ask the agent to book a call. The time picker appears in the chat. Pick a slot, and the agent confirms the booked time.

Key props

  • agentId: must match a runtime-registered agent. If omitted, the hook uses the agent of the surrounding chat configuration, then the agentId set on CopilotKitProvider, then "default". A mismatch means the interrupt never fires.
  • render: receives { event, interrupt, resolve }.
  • renderInChat: when true (as above), the picker appears inline in the chat transcript, between the paused assistant turn and the still-pending continuation. To show it outside the chat, see Headless interrupts.

Multiple interrupts? Add a type and gate with enabled

If your agent issues more than one kind of interrupt (e.g. "ask" vs "approval"), tag each with a type field on the payload and install one useInterrupt per shape, each gated by an enabled predicate:

// The LangGraph adapter sends `event.value` as a JSON string.
const readValue = (value: unknown) =>
  typeof value === "string" ? JSON.parse(value) : value;

useInterrupt({
  agentId: "gen-ui-interrupt",
  enabled: (event) => readValue(event.value).type === "ask",
  render: ({ event, resolve }) => (
    <AskCard question={readValue(event.value).content} onAnswer={resolve} />
  ),
});

useInterrupt({
  agentId: "gen-ui-interrupt",
  enabled: (event) => readValue(event.value).type === "approval",
  render: ({ event, resolve }) => (
    <ApproveCard content={readValue(event.value).content} onAnswer={resolve} />
  ),
});

Preprocess with handler

For cases where the interrupt can sometimes be resolved without user input (e.g. the current user already has permission), pass a handler that runs before render. The handler can call resolve(...) itself to resume the agent early. The interrupt card unmounts when the resume run starts. Or return a value that render receives as result:

useInterrupt({
  agentId: "gen-ui-interrupt",
  handler: async ({ event, resolve }) => {
    const value =
      typeof event.value === "string" ? JSON.parse(event.value) : event.value;
    const dept = await lookupUserDepartment();
    if (value.accessDepartment === dept || dept === "admin") {
      resolve({ code: "AUTH_BY_DEPARTMENT" });
      return; // agent will resume; card unmounts when the run starts
    }
    return { dept };
  },
  render: ({ result, event, resolve }) => (
    <RequestAccessCard
      dept={result?.dept}
      onRequest={() => resolve({ code: "REQUEST_AUTH" })}
      onCancel={() => resolve({ code: "CANCEL" })}
    />
  ),
});

AG-UI standard interrupt flow vs. custom-event flow

This section is a reference for developers who write render code for their own agent. The useInterrupt reference lists every option. useInterrupt handles two kinds of interrupt, and your agent framework decides which kind it sends.

Standard flow (RUN_FINISHED with outcome.type === "interrupt")

When the agent backend conforms to the AG-UI protocol, it signals an interrupt by emitting a RUN_FINISHED event whose outcome carries the interrupts array:

outcome.type === "interrupt"
outcome.interrupts  // Interrupt[]

The hook detects this on onRunFinishedEvent and exposes the interrupts on the render props after onRunFinalized fires. Your render function receives:

  • interrupt: the primary Interrupt (interrupts[0]), with shape { id, reason, message?, toolCallId?, responseSchema?, expiresAt?, metadata? }.
  • interrupts: the full open set (usually one, but multi-interrupt is supported, as described below).
  • resolve(payload?, interruptId?): records { status: "resolved", payload } for the targeted interrupt (defaults to the primary). The agent run resumes once every open interrupt has a response.
  • cancel(interruptId?): records { status: "cancelled" } for the targeted interrupt. Same accumulate-then-submit logic applies.

Each resume entry carries the interrupt id, the status, and your payload. resolve and cancel send no metadata.

Custom-event flow (on_interrupt)

Agents that emit a custom on_interrupt event use this flow. The LangGraph adapter emits this event by default. The hook detects this on onCustomEvent and sets interrupt to null and interrupts to []. The payload is in event.value. Calling resolve(payload) resumes via forwardedProps.command (the custom-event resume mechanism). cancel() dismisses the interrupt without resuming. The agent never receives a response.

Priority

If both signals appear on the same run, the standard flow wins.

Approve / Cancel example

function ApprovalInterrupt() {
  useInterrupt({
    render: ({ interrupt, resolve, cancel }) => (
      <div className="p-3 border rounded">
        <p>{interrupt?.message ?? "Approve this action?"}</p>
        <div className="mt-2 flex gap-2">
          <button onClick={() => resolve({ approved: true })}>Approve</button>
          <button onClick={() => cancel()}>Cancel</button>
        </div>
      </div>
    ),
  });
  return null;
}

resolve({ approved: true }) records a resolved entry and submits the resume array to the agent. cancel() records a cancelled entry and does the same. Both return the RunAgentResult once the run restarts.

Multi-interrupt behavior

Some agents issue more than one interrupt in a single run (e.g. two independent approvals). Each interrupt has its own id. Address them individually:

useInterrupt({
  render: ({ interrupts, resolve, cancel }) => (
    <ul>
      {interrupts.map((i) => (
        <li key={i.id}>
          {i.message}
          <button onClick={() => resolve({ ok: true }, i.id)}>Approve</button>
          <button onClick={() => cancel(i.id)}>Cancel</button>
        </li>
      ))}
    </ul>
  ),
});

The agent run only resumes once every open interrupt has been addressed. Calling resolve or cancel with a specific interruptId marks that interrupt done; the hook auto-submits the accumulated responses when the last one is addressed. If you omit interruptId, the primary interrupt (interrupts[0]) is targeted.

responseSchema: surface only, no client-side validation

The Interrupt type exposes a responseSchema field (a JSON Schema object) that the agent can use to describe the expected payload shape. useInterrupt surfaces this field on interrupt.responseSchema for your UI to read (e.g. to drive a form), but it does not validate resolve payloads against it. Validation is the agent's responsibility on resume.

How is this guide?