Examples

Ready-to-run snippets you can paste into Node, the browser, or any other Socket.IO client.

Live market chart

Subscribe to BTC/USD 1-minute candles and push updates into a chart library.

import { io } from 'socket.io-client';

const socket = io('ws://localhost:8091', {
  auth: { token: 'Bearer ' + process.env.LEDGERCORE_TOKEN },
  transports: ['websocket'],
});

socket.on('connected', () => {
  socket.emit('request_market_data', {
    symbol: 'BTC-USD',
    resolution: '1m',
    window: '24h',
  });
});

socket.on('market_chart_initial_data', ({ series }) => {
  chart.setData(series);
});

socket.on('market_chart_live_update', (tick) => {
  chart.updateLastCandle(tick);
});

socket.on('market_data_error', ({ code, message }) => {
  console.error(code, message);
});

Streaming query results

Fire off a SQL query against the LedgerCore query engine and render rows as they arrive.

const queryId = crypto.randomUUID();

socket.emit('query', {
  sql: `SELECT date, SUM(volume_usd) AS v
        FROM trades
        WHERE venue = 'binance' AND date > NOW() - INTERVAL 7 DAY
        GROUP BY 1 ORDER BY 1`,
  applicationId: 'app_2xMn1psnHkYRTH00nxdvp02IWfr',
});

socket.on('query_started', ({ queryId: id, estimatedMs }) => {
  console.log('ETA', estimatedMs, 'ms');
});

socket.on('query_progress', ({ pct }) => {
  progressBar.set(pct);
});

socket.on('query_response', ({ rows }) => {
  table.append(rows);
});

socket.on('query_completed', ({ totalRows, durationMs }) => {
  console.log('done —', totalRows, 'rows in', durationMs, 'ms');
});

Live credit balance

Your personal room delivers every credit movement without a subscribe call. Perfect for showing a balance widget that never goes stale.

socket.on('credit_balance_updated', ({ newBalance, applicationId }) => {
  balanceWidget.set(applicationId, newBalance);
});

socket.on('credits_deducted', ({ amount, action, transactionId }) => {
  toast(`Used ${amount} credits for ${action}`);
});

socket.on('insufficient_credits', ({ required, available, action }) => {
  modal.open('topup', { required, available, action });
});

LedgerBrain AI — streaming chat

The assistant streams deltas just like an LLM chat. Concatenate them until isFinal === true.

const conversationId = crypto.randomUUID();
let buffer = '';

socket.emit('chat', {
  conversationId,
  message: 'Show me the top 5 CEX inflows over the last 24 hours.',
});

socket.on('chat_typing', () => ui.showTyping());
socket.on('chat_stop_typing', () => ui.hideTyping());

socket.on('chat_message', ({ delta, isFinal }) => {
  buffer += delta;
  ui.renderMarkdown(buffer);
  if (isFinal) {
    buffer = '';
  }
});

socket.on('chat_response', ({ sql }) => {
  if (sql) codePane.show(sql);
});

Long-running address analysis

Fire-and-forget compliance job with progress updates.

socket.emit('address_analysis', {
  chain: 'bitcoin',
  address: '1Drt3c8pSdrkyjuBiwVcSSixZwQtMZ3Tew',
  hops: 3,
});

socket.on('address_analysis_started', ({ jobId }) => {
  console.log('job started', jobId);
});

socket.on('address_analysis_progress', ({ stage, pct }) => {
  statusBar.set(`${stage} — ${pct}%`);
});

socket.on('address_analysis_response', ({ riskScore, riskLevel, signals }) => {
  panel.render({ riskScore, riskLevel, signals });
});

socket.on('address_analysis_completed', ({ durationMs }) => {
  console.log('done in', durationMs, 'ms');
});

Python client

Any Socket.IO 4.x client works. Here's the equivalent of the market chart example in Python:

import os
import socketio

sio = socketio.Client()

@sio.on('connected')
def on_ready(data):
    sio.emit('request_market_data', {
        'symbol': 'BTC-USD',
        'resolution': '1m',
        'window': '24h',
    })

@sio.on('market_chart_initial_data')
def on_initial(data):
    print('got', len(data['series']), 'candles')

@sio.on('market_chart_live_update')
def on_tick(tick):
    print(tick['timestamp'], tick['close'])

sio.connect(
    'ws://localhost:8091',
    auth={'token': f"Bearer {os.environ['LEDGERCORE_TOKEN']}"},
    transports=['websocket'],
)
sio.wait()