xiaohongshu
xiaohongshu
ai-sdk-ui
Expert guidance for building chat UIs with AI SDK React hooks. Use when: (1) Building chatbots with useChat hook, (2) Implementing tool UIs with server/client execution, (3) Handling message persistence and stream resumption, (4) Creating generative UI with React components, (5) Integrating with ...
Full skill instructions
AI SDK UI provides framework-agnostic hooks for building interactive chat, completion, and assistant applications with real-time streaming and state management.
'use client';
import { useChat } from '@ai-sdk/react';
import { DefaultChatTransport } from 'ai';
import { useState } from 'react';
export default function Chat() {
const { messages, sendMessage, status } = useChat({
transport: new DefaultChatTransport({ api: '/api/chat' }),
});
const [input, setInput] = useState('');
return (
<>
{messages.map(message => (
<div key={message.id}>
{message.role === 'user' ? 'User: ' : 'AI: '}
{message.parts.map((part, index) =>
part.type === 'text' ? <span key={index}>{part.text}</span> : null
)}
</div>
))}
<form onSubmit={e => {
e.preventDefault();
if (input.trim()) {
sendMessage({ text: input });
setInput('');
}
}}>
<input
value={input}
onChange={e => setInput(e.target.value)}
disabled={status !== 'ready'}
/>
<button type="submit" disabled={status !== 'ready'}>
Submit
</button>
</form>
</>
);
}
import { convertToModelMessages, streamText, UIMessage } from 'ai';
import { openai } from '@ai-sdk/openai';
export const maxDuration = 30;
export async function POST(req: Request) {
const { messages }: { messages: UIMessage[] } = await req.json();
const result = streamText({
model: openai('gpt-4o'),
system: 'You are a helpful assistant.',
messages: await convertToModelMessages(messages), // v6: now async
});
return result.toUIMessageStreamResponse();
}
| Hook | Use Case | Stream Type | Best For |
|---|---|---|---|
useChat | Multi-turn conversations | Messages with parts (text, tools, files) | Chatbots, assistants, tool-calling UIs |
useObject | Structured data streaming | Typed objects with Zod schema | Forms, dashboards, real-time data |
useCompletion | Single-turn text generation | Plain text | Autocomplete, simple generation |
Decision tree:
useChatuseObjectuseCompletionconst { status, stop } = useChat();
// status values: 'submitted' | 'streaming' | 'ready' | 'error'
{(status === 'submitted' || status === 'streaming') && (
<div>
{status === 'submitted' && <Spinner />}
<button onClick={() => stop()}>Stop</button>
</div>
)}
const { error, reload } = useChat();
{error && (
<>
<div>An error occurred.</div>
<button onClick={() => reload()}>Retry</button>
</>
)}
Server-side error messages:
return result.toUIMessageStreamResponse({
onError: error => {
if (error instanceof Error) return error.message;
return 'Unknown error';
},
});
const { messages, setMessages } = useChat();
const handleDelete = (id: string) => {
setMessages(messages.filter(m => m.id !== id));
};
const handleEdit = (id: string, newText: string) => {
setMessages(messages.map(m =>
m.id === id
? { ...m, parts: [{ type: 'text', text: newText }] }
: m
));
};
const [files, setFiles] = useState<FileList | undefined>();
<form onSubmit={e => {
e.preventDefault();
sendMessage({ text: input, files });
setFiles(undefined);
}}>
<input
type="file"
onChange={e => setFiles(e.target.files ?? undefined)}
multiple
/>
</form>
// Per-request customization (recommended)
sendMessage(
{ text: input },
{
headers: { Authorization: 'Bearer token' },
body: { temperature: 0.7, user_id: '123' },
metadata: { sessionId: 'abc' },
}
);
// Hook-level configuration
const { sendMessage } = useChat({
transport: new DefaultChatTransport({
api: '/api/chat',
headers: () => ({ Authorization: `Bearer ${getToken()}` }),
body: { systemContext: 'expert' },
}),
});
// Server: Attach metadata
return result.toUIMessageStreamResponse({
messageMetadata: ({ part }) => {
if (part.type === 'start') {
return { createdAt: Date.now(), model: 'gpt-4o' };
}
if (part.type === 'finish') {
return { totalTokens: part.totalUsage.totalTokens };
}
},
});
// Client: Access metadata
{messages.map(m => (
<div key={m.id}>
{m.metadata?.createdAt && new Date(m.metadata.createdAt).toLocaleString()}
{m.parts.map(part => part.type === 'text' ? part.text : null)}
{m.metadata?.totalTokens && <span>{m.metadata.totalTokens} tokens</span>}
</div>
))}
const { regenerate, stop, status } = useChat();
<>
<button onClick={stop} disabled={status !== 'streaming'}>
Stop
</button>
<button onClick={regenerate} disabled={!(status === 'ready' || status === 'error')}>
Regenerate
</button>
</>
Messages use a parts array instead of content for flexible multi-modal rendering:
type MessagePart =
| { type: 'text'; text: string }
| { type: 'file'; filename: string; mediaType: string; url: string }
| { type: 'tool-invocation'; toolName: string; input: unknown; result?: unknown }
| { type: 'tool-result'; toolName: string; result: unknown }
| { type: 'reasoning'; text: string } // DeepSeek R1, Claude 3.7 Sonnet
| { type: 'source-url'; id: string; url: string; title?: string } // Perplexity, Google
| { type: 'source-document'; id: string; title?: string };
// Render pattern
{message.parts.map((part, index) => {
switch (part.type) {
case 'text':
return <span key={index}>{part.text}</span>;
case 'file':
return part.mediaType.startsWith('image/')
? <img key={index} src={part.url} alt={part.filename} />
: null;
case 'reasoning':
return <pre key={index}>{part.text}</pre>;
case 'source-url':
return <a key={index} href={part.url}>{part.title ?? 'Source'}</a>;
case 'tool-invocation':
return <ToolUI key={index} tool={part} />;
default:
return null;
}
})}
| Framework | Package | Hooks |
|---|---|---|
| React | @ai-sdk/react | useChat, useCompletion, useObject |
| Vue.js | @ai-sdk/vue | useChat, useCompletion, useObject |
| Svelte | @ai-sdk/svelte | Chat, Completion, StructuredObject |
| Angular | @ai-sdk/angular | Chat, Completion, StructuredObject |
| SolidJS | ai-sdk-solid (community) | useChat, useCompletion, useObject |
Pre-built UI components for chat interfaces: https://ai-sdk.dev/elements
Includes: Message bubbles, input fields, tool UIs, and more.
| Reference | Topics |
|---|---|
| usechat-fundamentals.md | Hook API, transport config, status lifecycle, message state |
| tool-integration.md | Tool calling, client/server execution, tool approval, type inference |
| generative-ui.md | React components in streams, dynamic UIs, RSC integration |
| persistence.md | Message storage, resume streams, optimistic updates, sync patterns |
| hooks-reference.md | Complete API for useChat/useObject/useCompletion, options reference |
| backend.md | Next.js/Node/Fastify/Nest routes, convertToModelMessages, toUIMessageStreamResponse |
| production.md | Error boundaries, retry strategies, throttling, security best practices |
| migration.md | v6 migration guide, breaking changes, codemod usage |
const { messages } = useChat({
onFinish: ({ message, messages, isAbort, isDisconnect, isError }) => {
// Log completion, update analytics, trigger side effects
if (!isError) logMessage(message);
},
onError: error => {
// Custom error handling, fallback UI
Sentry.captureException(error);
},
onData: data => {
// Process data parts, validate responses
// Throw error to abort processing
},
});
const { sendMessage } = useChat({
transport: new DefaultChatTransport({
prepareSendMessagesRequest: ({ id, messages, trigger, messageId }) => {
if (trigger === 'submit-user-message') {
return {
body: {
id,
message: messages[messages.length - 1],
messageId,
},
};
}
// Handle regenerate, custom triggers
},
}),
});
import { InferUITools, InferAgentUIMessage, ToolSet, UIMessage } from 'ai';
const tools = {
weather: {
description: 'Get weather',
inputSchema: z.object({ location: z.string() }),
execute: async ({ location }) => `Sunny in ${location}`,
},
} satisfies ToolSet;
type MyUITools = InferUITools<typeof tools>;
type MyUIMessage = UIMessage<never, never, MyUITools>;
// Or for agent messages:
// type MyAgentUIMessage = InferAgentUIMessage<typeof myAgent>;
const { messages } = useChat<MyUIMessage>();
// Enable reasoning (DeepSeek R1, Claude 3.7 Sonnet)
return result.toUIMessageStreamResponse({ sendReasoning: true });
// Enable sources (Perplexity, Google)
return result.toUIMessageStreamResponse({ sendSources: true });
// Render reasoning and sources
{message.parts.map(part => {
if (part.type === 'reasoning') return <pre>{part.text}</pre>;
if (part.type === 'source-url') return <a href={part.url}>{part.title}</a>;
})}
const { messages } = useChat({
experimental_throttle: 50, // React only: throttle to 50ms
});
import { TextStreamChatTransport } from 'ai';
const { messages } = useChat({
transport: new TextStreamChatTransport({ api: '/api/chat' }),
});
Note: Tools, usage, and finish reasons unavailable with TextStreamChatTransport.
xiaohongshu
technical spec
product ux expert
database patterns
Conduct multi-agent task orchestration and workflow coordination.
Initialize project with Conductor artifacts (product definition,
Expert in web animations, transitions, and motion design using Framer Motion and CSS
Creates Mermaid and ASCII diagrams for flowcharts, architecture, ERDs, state machines, mindmaps, and more. Use when user mentions diagram, flowchart, mermaid, ASCII diagram, text diagram, terminal diagram, visualize, C4, mindmap, architecture diagram, sequence diagram, ERD, or needs visual docume...
PostgreSQL bindings for H3 hexagonal grid system. Use when working with H3 cells in Postgres, including spatial indexing, geometry/geography integration, and raster analysis.
Context-Driven Development skill for projects using Conductor. Use this skill when you detect a `conductor/` directory in the project, when working on tasks defined in a `plan.md` file, or when the user asks about tracks, specs, or plans. Automatically applies TDD workflow, tracks task completion...
Display project status, active tracks, and next actions
Official Stakpak application containerization standard operating procedure, a step-by-step guidline to properly dockerize applications. This is a rule book curated by the Stakpak Team.
Generate, edit, and beat-sync AI video with leading models in one workspace.
The world's fastest calendar for remote work
Transform Your Design with AI Designer by ImgCreator.ai
Revolutionizing Video Production with AI-Powered Creativity
Extend an image past the frame and let AI fill the new aspect ratio.
Discover your celebrity doppelgänger with StarByFace!
ChainClarity explains 700+ crypto whitepapers in plain English, with layered summaries, comparisons, research tools, alerts, and a $4.99 Pro plan.
Opus.ai: Revolutionize Your Web Experience