CopilotChatInput
Primary text input and control surface for chat interactions
Overview
CopilotChatInput is the primary text input and control surface for chat interactions. It provides a multi-line textarea that auto-grows up to a configurable maximum number of rows (default 5), action buttons for sending messages, voice transcription, and file attachment, and an optional tools menu for declarative command surfaces.
The component operates in three modes: "input" (default text entry), "transcribe" (replaces the textarea with an audio recorder), and "processing" (indicates background processing is underway). When text spans multiple rows, the layout stacks the textarea above the control row.
Import
import { CopilotChatInput } from "@copilotkit/react-core/v2";
import "@copilotkit/react-core/v2/styles.css";Props
Prop
Type
Prop
Type
Prop
Type
Prop
Type
Prop
Type
Prop
Type
Prop
Type
Prop
Type
Prop
Type
Prop
Type
Prop
Type
Prop
Type
Prop
Type
Prop
Type
Slots
All slot props follow the CopilotKit slot system: each accepts a replacement React component, a className string that is merged into the default component's classes, or a partial props object that extends the default component.
Prop
Type
Prop
Type
Prop
Type
Prop
Type
Prop
Type
Prop
Type
Prop
Type
ToolsMenuItem
The ToolsMenuItem type defines items in the tools menu. Items can trigger actions directly or contain nested submenus.
type ToolsMenuItem = { label: string } & (
| { action: () => void }
| { items: (ToolsMenuItem | "-")[] }
);Prop
Type
Prop
Type
Prop
Type
Separators
Use the string "-" as an array entry to render a visual separator between menu items:
toolsMenu={[
{ label: "Search", action: () => search() },
"-",
{ label: "Settings", action: () => openSettings() },
]}Nested Submenus
Menu items with an items array render as expandable submenus:
toolsMenu={[
{
label: "Export",
items: [
{ label: "As PDF", action: () => exportPDF() },
{ label: "As CSV", action: () => exportCSV() },
"-",
{ label: "Print", action: () => printDoc() },
],
},
]}CopilotChatAudioRecorder
A visual audio waveform component used during transcription mode. It displays a real-time waveform visualization of the microphone input and exposes an imperative API via ref.
Import
Imperative API
The component exposes the following imperative methods via React ref:
Prop
Type
Prop
Type
Prop
Type
Prop
Type
Usage
function CustomRecorder() {
const recorderRef = useRef(null);
return (
<div>
<CopilotChatAudioRecorder ref={recorderRef} />
<button onClick={() => recorderRef.current?.start()}>Record</button>
<button onClick={() => recorderRef.current?.stop()}>Stop</button>
</div>
);
}Usage
Basic Chat Input
function ChatInput() {
const [value, setValue] = useState("");
return (
<CopilotChatInput
value={value}
onChange={setValue}
onSubmitMessage={(text) => {
sendMessage(text);
setValue("");
}}
autoFocus
/>
);
}With Voice Transcription
function VoiceInput() {
const [mode, setMode] = useState<"input" | "transcribe">("input");
const [value, setValue] = useState("");
return (
<CopilotChatInput
mode={mode}
value={value}
onChange={setValue}
onSubmitMessage={(text) => sendMessage(text)}
onStartTranscribe={() => setMode("transcribe")}
onCancelTranscribe={() => setMode("input")}
onFinishTranscribeWithAudio={async (blob) => {
const text = await transcribeAudio(blob);
setValue(text);
setMode("input");
}}
/>
);
}With Tools Menu
function InputWithTools() {
return (
<CopilotChatInput
onSubmitMessage={(text) => sendMessage(text)}
toolsMenu={[
{ label: "Search the web", action: () => triggerSearch() },
{ label: "Analyze data", action: () => triggerAnalysis() },
"-",
{
label: "Export",
items: [
{ label: "As PDF", action: () => exportPDF() },
{ label: "As Markdown", action: () => exportMarkdown() },
],
},
]}
/>
);
}With Stop Button During Execution
function ChatInputWithAgent() {
const { agent } = useAgent();
return (
<CopilotChatInput
isRunning={agent.isRunning}
onSubmitMessage={(text) => {
agent.addMessage({
role: "user",
content: text,
id: crypto.randomUUID(),
});
agent.runAgent();
}}
onStop={() => agent.abortRun()}
autoFocus
/>
);
}Custom Layout with Children Render Prop
function AdaptiveInput() {
return (
<CopilotChatInput onSubmitMessage={(text) => sendMessage(text)}>
{({ isMultiline }) => (
<div
className={
isMultiline ? "border-2 border-blue-300 rounded-xl p-2" : ""
}
>
{isMultiline && (
<div className="text-xs text-gray-400 mb-1">Multi-line mode</div>
)}
</div>
)}
</CopilotChatInput>
);
}Styled Slots
function StyledInput() {
return (
<CopilotChatInput
onSubmitMessage={(text) => sendMessage(text)}
textArea="font-mono text-sm bg-gray-50 rounded-lg"
sendButton="bg-green-500 hover:bg-green-600 text-white rounded-full"
/>
);
}Behavior
- Auto-Growing Textarea: The default textarea automatically expands as the user types, up to a maximum of 5 rows. After reaching the maximum, the textarea becomes scrollable.
- Layout Stacking: When the input text spans multiple rows, the layout switches from inline (textarea and buttons side by side) to stacked (textarea above the control row containing the action buttons).
- Mode Switching: Setting
modeto"transcribe"replaces the textarea with theaudioRecorderslot component. The transcription control buttons (cancelTranscribeButton,finishTranscribeButton) replace the standard send button. - Controlled and Uncontrolled: The component supports both controlled mode (via
valueandonChange) and uncontrolled mode (internal state management). - Tools Menu: The
toolsMenuprop enables a declarative menu with support for flat items, separators ("-"), and nested submenus. The menu button is only rendered whentoolsMenuis provided. - Slot System: Each slot prop accepts three forms -- a replacement component, a className string merged into the default, or a partial props object that extends the default component's props.
Related
CopilotChatView-- Higher-level chat view that composes this componentCopilotChatAudioRecorder-- Audio recorder used during transcription modeuseCopilotChatConfiguration-- Provider for localized input labels