Fumadocs on TanStack Start

Overview

Enable streaming to receive agent responses token-by-token using Server-Sent Events (SSE). This provides a better user experience for longer responses.

Request

Set stream: true in your chat request:

curl https://api.your-domain.com/v1/chat \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "message": "Explain quantum computing",
    "stream": true
  }'

Response Format

The response is a stream of SSE events:

data: {"type": "start", "conversation_id": "conv_abc123"}

data: {"type": "token", "content": "Quantum"}
data: {"type": "token", "content": " computing"}
data: {"type": "token", "content": " is"}
data: {"type": "token", "content": " a"}

data: {"type": "sources", "sources": ["https://example.com/quantum"]}

data: {"type": "end", "message_id": "msg_xyz789"}

Client Implementation

JavaScript

const response = await fetch('https://api.your-domain.com/v1/chat', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer YOUR_API_KEY',
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    message: 'Explain quantum computing',
    stream: true,
  }),
});

const reader = response.body.getReader();
const decoder = new TextDecoder();

while (true) {
  const { done, value } = await reader.read();
  if (done) break;

  const chunk = decoder.decode(value);
  const lines = chunk.split('\n').filter(l => l.startsWith('data: '));

  for (const line of lines) {
    const data = JSON.parse(line.slice(6));
    if (data.type === 'token') {
      process.stdout.write(data.content);
    }
  }
}

Event Types

Type	Description
`start`	Stream has started, includes `conversation_id`
`token`	A single token of the response
`sources`	Source documents used for the response
`end`	Stream complete, includes `message_id`
`error`	An error occurred during generation

Streaming