The mcp-binance-futures server provides comprehensive access to Binance USDT-M Futures trading, enabling market analysis, account management, order execution, and position control.
Market Data (no authentication required)
ping — Test API connectivity
get_ticker — Latest price, 24h stats, mark/index price, and funding rate
get_order_book — Top N bids and asks (5–1000 depth levels)
get_recent_trades — Most recent public trades for a symbol
get_klines — OHLCV candlestick data (intervals: 1m to 1w)
get_symbol_info — Trading rules (tick size, lot size, min notional, max leverage, order types)
Account Management (requires API keys)
get_balance — Wallet balances for all non-zero assets
get_positions — Open positions with PnL, leverage, margin type, and liquidation price
get_account_summary — Total balance, unrealized PnL, margin usage, and open position count
Order Management (requires API keys)
place_order — Place LIMIT, MARKET, STOP, STOP_MARKET, TAKE_PROFIT, TAKE_PROFIT_MARKET, or TRAILING_STOP_MARKET orders
modify_order — Change price or quantity of an existing open LIMIT order
cancel_order — Cancel a single order by ID (supports regular and algo/conditional orders)
cancel_all_orders — Cancel all open orders for a symbol
get_open_orders — List all open orders (regular, algo, or all)
get_order — Retrieve a specific order by order ID or client order ID
get_order_history — Recent order history for a symbol
get_trade_history — Personal fill/execution history for a symbol
Position & Margin Control (requires API keys)
set_leverage — Set leverage multiplier (1–125×) for a symbol
set_margin_type — Switch between ISOLATED and CROSSED margin mode
adjust_isolated_margin — Add or remove margin from an isolated position
set_position_mode — Switch between One-way and Hedge mode
get_position_mode — Check the current position mode
get_leverage_brackets — View leverage tiers and maintenance margin rates
Provides tools for monitoring and trading Binance USDT-M Futures, including access to market data, account balances, position management, and order execution.
MCP server for Binance USDT-M Futures trading. Exposes tools for market data, account state, order management, and position/margin control — designed to give an LLM everything it needs to monitor, place, and manage futures trades.
Tool | Description |
| Test API connectivity |
| Price, 24 h stats, mark price, funding rate for a symbol |
| Top N bids/asks for a symbol |
| Latest public trades |
| OHLCV candlestick data (1m → 1w) |
| Trading rules: tick size, lot size, min notional, order types |
Tool | Description |
| Wallet balances (non-zero assets only) |
| Open positions with PnL, leverage, margin type — optionally scoped to one symbol |
| Total balance, unrealized PnL, margin usage, open position count |
Tool | Description |
| Place LIMIT, MARKET, STOP, STOP_MARKET, TAKE_PROFIT, TAKE_PROFIT_MARKET, TRAILING_STOP_MARKET |
| Change price or quantity of an open LIMIT order |
| Cancel a single order by ID |
| Cancel all open orders for a symbol |
| List all open orders for a symbol |
| Get a specific order by ID |
| Recent order history (all statuses) |
| Personal fill history for a symbol |
Tool | Description |
| Set leverage multiplier (1–125×) for a symbol |
| Switch between |
| Add or remove margin from an isolated position |
| Switch between One-way and Hedge Mode |
| Get current position mode |
| Leverage tiers with maintenance margin rates |
Related MCP server: binance-data MCP Server
Python 3.11+
uv (recommended) or pip
# with uv (recommended)
uv sync
# or with pip
pip install -e .Create a Binance API key with Futures trading enabled. Set environment variables:
export BINANCE_API_KEY="your_api_key"
export BINANCE_API_SECRET="your_api_secret"Security: Use IP whitelisting on your Binance API key. Never commit keys to version control.
# stdio transport (default — for MCP clients like Claude Desktop)
python server.py
# or via the installed script
mcp-binance-futuresAdd to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"binance-futures": {
"command": "python",
"args": ["/path/to/mcp-binance-futures/server.py"],
"env": {
"BINANCE_API_KEY": "your_api_key",
"BINANCE_API_SECRET": "your_api_secret"
}
}
}
}{
"mcpServers": {
"binance-futures": {
"command": "uv",
"args": ["run", "--project", "/path/to/mcp-binance-futures", "mcp-binance-futures"],
"env": {
"BINANCE_API_KEY": "your_api_key",
"BINANCE_API_SECRET": "your_api_secret"
}
}
}
}# install dev dependencies
uv sync --extra dev
# run all tests
pytest
# run with output
pytest -vTests use respx to mock all HTTP calls — no real API keys or network required.
1. get_ticker(symbol="BTCUSDT") → check current price
2. get_balance() → check available margin
3. get_positions(symbol="BTCUSDT") → confirm no existing position
4. set_leverage(symbol="BTCUSDT", leverage=10)
5. set_margin_type(symbol="BTCUSDT", margin_type="ISOLATED")
6. place_order(symbol="BTCUSDT", side="BUY", order_type="MARKET", quantity=0.01)
7. place_order(symbol="BTCUSDT", side="SELL", order_type="STOP_MARKET",
stop_price=45000, close_position=True)
8. place_order(symbol="BTCUSDT", side="SELL", order_type="TAKE_PROFIT_MARKET",
stop_price=55000, close_position=True)1. get_open_orders(symbol="BTCUSDT") → find the order ID
2. modify_order(symbol="BTCUSDT", order_id=123456, side="BUY",
quantity=0.01, price=48500)1. cancel_all_orders(symbol="BTCUSDT")
2. place_order(symbol="BTCUSDT", side="SELL", order_type="MARKET",
quantity=<position_size>, reduce_only=True)server.py — FastMCP server, all tool definitions
client.py — Async HTTP client: signing, transport, error handling
tests/
test_client.py — Unit tests for BinanceClient (signing, HTTP, errors)
test_server.py — Integration tests for all MCP toolsThe client and server are intentionally kept in separate files: client.py handles all Binance API mechanics (HMAC signing, error parsing, HTTP verbs) while server.py contains only tool logic and MCP wiring. This makes both independently testable.
Add or remove margin from an isolated position.
Only valid when the symbol is in ISOLATED margin mode with an open position.
Inputs: symbol (required), amount (required), direction (required), position_side. Returns: object
Cancel all open orders for a symbol.
With source='all' (default), cancels both regular and algo (conditional) orders in parallel.
Inputs: symbol (required), source. Returns: object
Cancel a single open order by orderId or clientOrderId.
For algo (conditional) orders returned by get_open_orders with '_isAlgo': True, set is_algo=True — this routes to the Algo API using algoId.
Inputs: symbol (required), order_id, client_order_id, is_algo. Returns: object
Get account-level summary: total balance, unrealized PnL, margin ratio, positions count.
Inputs: no named parameters. Returns: object
Get futures wallet balances for all assets with non-zero balance.
Returns list of: asset, balance, availableBalance, crossWalletBalance, unrealizedProfit.
Inputs: no named parameters. Returns: result
Get OHLCV candlestick data for a symbol.
Returns list of dicts with: openTime, open, high, low, close, volume, closeTime, quoteVolume, trades, takerBuyVolume, takerBuyQuoteVolume.
Inputs: symbol (required), interval, limit. Returns: result
Get leverage brackets for a symbol: max leverage per notional tier with maintenance margin rates.
Inputs: symbol (required). Returns: result
Get open orders for a symbol.
Returns list of: orderId, clientOrderId, symbol, status, type, side, positionSide, price, origQty, executedQty, avgPrice, stopPrice, timeInForce, reduceOnly, closePosition, updateTime. Algo (conditional) orders also include '_isAlgo': True.
Inputs: symbol (required), source. Returns: result
Get details of a specific order by orderId or clientOrderId.
Inputs: symbol (required), order_id, client_order_id. Returns: object
Get order book bids and asks for a symbol.
Returns top limit bids and asks as [[price, qty], ...] lists.
Inputs: symbol (required), limit. Returns: object
Get recent order history for a symbol (all statuses).
Inputs: symbol (required), limit. Returns: result
Get current position mode: Hedge Mode or One-way Mode.
Inputs: no named parameters. Returns: object
Get current open positions (non-zero size).
Per position: symbol, side, size, entryPrice, markPrice, unrealizedPnl, percentage, leverage, marginType, isolatedMargin, liquidationPrice.
Inputs: symbol. Returns: result
Get the most recent public trades for a symbol.
Inputs: symbol (required), limit. Returns: result
Get trading rules for a symbol: tick size, lot size, min notional, max leverage, etc.
Inputs: symbol (required). Returns: object
Get latest price, 24 h stats, and mark/index prices for a symbol.
Returns a merged dict with:
price, priceChange, priceChangePct, high, low, volume, quoteVolume
markPrice, indexPrice, fundingRate, nextFundingTime
Inputs: symbol (required). Returns: object
Get your personal trade execution history for a symbol (fills).
Inputs: symbol (required), limit. Returns: result
Modify price or quantity of an existing open LIMIT order (PUT /fapi/v1/order).
Inputs: symbol (required), side (required), quantity (required), order_id, +2 more. Returns: object
Test connectivity to the Binance Futures API. Returns {} on success.
Inputs: no named parameters. Returns: object
Place a new futures order.
ORDER TYPE GROUPS — choose the right one:
Immediate orders (fill now or queue at price): MARKET: side=BUY, quantity=0.01 LIMIT: side=SELL, quantity=0.01, price=50000, time_in_force=GTC
Conditional orders (wait for stop_price trigger, then execute): Stop-loss full close: side=SELL, type=STOP_MARKET, stop_price=45000, close_position=True Take-profit full close: side=SELL, type=TAKE_PROFIT_MARKET, stop_price=60000, close_position=True Stop-loss partial: side=SELL, type=STOP_MARKET, stop_price=45000, quantity=0.01, reduce_only=True Trailing stop: side=SELL, type=TRAILING_STOP_MARKET, stop_price=45000, quantity=0.01, callback_rate=1.0
IMPORTANT — close_position=True vs quantity+reduce_only: close_position=True → closes the ENTIRE position, no quantity needed, max 1 SL + 1 TP active at a time. reduce_only=True → closes a PARTIAL quantity, multiple allowed simultaneously. Never mix both on the same order.
Inputs: symbol (required), side (required), order_type (required), quantity, +8 more. Returns: object
Set leverage for a symbol. Returns the new leverage and max notional value.
Inputs: symbol (required), leverage (required). Returns: object
Switch margin type for a symbol between ISOLATED and CROSSED.
Note: Cannot change margin type while a position or open order exists.
Inputs: symbol (required), margin_type (required). Returns: object
Switch between One-way Mode and Hedge Mode for the account.
Note: Cannot change while any positions or open orders exist.
Inputs: hedge_mode (required). Returns: object
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 23 tools
Most tools have distinct purposes targeting specific futures trading operations like orders, positions, account info, and market data. However, some overlap exists between get_order and get_order_history (specific vs. recent orders) and between get_ticker and get_symbol_info (price stats vs. trading rules), though descriptions help clarify their differences.
Tool names follow a highly consistent verb_noun pattern throughout, such as get_account_summary, cancel_order, set_leverage, and adjust_isolated_margin. All tools use snake_case with clear, descriptive verbs, making the set predictable and easy to navigate.
With 23 tools, the count is borderline high for a single server, though it covers a comprehensive range of futures trading operations. It feels slightly heavy but is justifiable given the complexity of the domain, including orders, positions, account management, and market data.
The tool set provides complete coverage for Binance Futures trading, including full CRUD/lifecycle operations for orders (place, modify, cancel, get), positions (get, adjust margin), account settings (leverage, margin type, position mode), and market data (klines, ticker, order book). No obvious gaps are present for core trading workflows.
Trading MCP server for AI agents, with live market data, account reads and controlled execution.
MCP server for OpenMM — exposes market data, account, trading, and strategy tools to AI agents
Hosted MCP server for MT4 and MT5: read accounts, prices and candles, and place or close orders.
MCP server for Gainium — manage trading bots, deals, and balances via AI assistants
