Provides read-only tools for accessing Binance Spot market data (price, klines, order book, trading rules) and account information (balances, open orders, trade history) via the Binance Spot Testnet API.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Binance Trading Agent MCP Server what's the current price of BTCUSDT?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A safety-first Model Context Protocol (MCP) server that connects any LLM client (Claude, Cursor, custom agents) to the Binance Spot API — with a hard focus on guardrails, testnet-first development, and auditable order flow.
7 read-only MCP tools for reading market data and account info on Binance Spot Testnet.
Tool | Type | Input | Purpose |
| public | symbol | price, 24h change, bid/ask, high/low, volume |
| public | symbol, interval, limit | candlestick data |
| public | symbol, limit | order book depth |
| public | symbol | minQty, stepSize, tickSize, minNotional |
| private | hideZero | free/locked balances, canTrade |
| private | symbol? | open orders |
| private | symbol, limit | recent trades |
Not in Phase 1: order placement/cancellation, futures, margin, OCO, websockets, autonomous actions.
Related MCP server: Binance MCP Server
src/
binance/ Layer 1 — Binance HTTP client
client.ts signedRequest / publicRequest, retry, timeout
signing.ts HMAC SHA256
errors.ts typed error classes
rateLimiter.ts weight tracking + backoff
types.ts response types
services/ Layer 2 — domain services (normalization)
marketDataService.ts
accountService.ts
tools/ Layer 3 — MCP tools (1 file = 1 tool)
getSpotPrice.ts ... getTradeHistory.ts
config.ts .env validation (zod)
logger.ts structured JSON logging
server.ts MCP server registration
tests/
unit/ signing, errors, rateLimiter, services, client# 1. Install
npm install
# 2. Configure (Binance Spot TESTNET keys)
cp .env.example .env
# Edit .env with your testnet keys
# 3. Run the MCP server
npm run devAdd to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"binance": {
"command": "npx",
"args": ["tsx", "/path/to/binance-trading-agent-mcp/src/server.ts"]
}
}
}Add to .cursor/mcp.json:
{
"mcpServers": {
"binance": {
"command": "npx",
"args": ["tsx", "/path/to/binance-trading-agent-mcp/src/server.ts"]
}
}
}Get BTC price:
tool: get_spot_price
args: { "symbol": "BTCUSDT" }Get 1h klines (last 100 candles):
tool: get_klines
args: { "symbol": "BTCUSDT", "interval": "1h", "limit": 100 }Get order book (top 20 levels):
tool: get_order_book
args: { "symbol": "ETHUSDT", "limit": 20 }Get symbol trading rules:
tool: get_symbol_rules
args: { "symbol": "BTCUSDT" }Get account balances:
tool: get_account_balances
args: { "hideZero": true }Get open orders:
tool: get_open_orders
args: { "symbol": "BTCUSDT" }Get trade history:
tool: get_trade_history
args: { "symbol": "BTCUSDT", "limit": 50 }# Run all unit tests
npm test
# Watch mode
npm run test:watchLanguage: TypeScript (Node 20+)
Protocol: Model Context Protocol (MCP)
Exchange API: Binance Spot REST
Tooling: vitest, zod, pino
Phase 1 — Binance client + 7 read-only tools
Phase 2 — Safe testnet trading: risk/policy layer, preview/place/cancel, full logging
Phase 3 — Strategy design + backtesting on historical klines + paper trading
Phase 4 — (optional) minimal real-money trading with strict limits
Phase 5 — (optional) futures support
This is engineering infrastructure, not financial advice. Automated trading with real funds carries real risk of loss. Use the Binance Testnet and dry_run mode by default. You are solely responsible for any keys, funds, strategy, and regulatory compliance in your jurisdiction.
MIT
This server cannot be deployed
Unified financial infrastructure connecting AI agents directly to trade live/demo brokerage accounts, Web3 non-custodial wallets, real-time market data across equities, ETFs, crypto, forex, options, DeFi swaps, and prediction markets, institutional research feeds, and algorithmic strategy backtesters.
Connect AI agents to bank accounts, transactions, balances, and investments.
DeFi safety layer for AI agents: wallet safety, token risk, tx decode/simulate. 20 tools.
Connect AI agents to financial institution origination, analytics, and compliance workflows.
