MCP Servers for Finance and Trading: A Complete List (2026)

The Model Context Protocol (MCP) lets AI agents call external tools through a standardized interface. For traders and fintech developers, this means your Claude, Cursor, or custom agent can fetch market data, execute trades, and manage risk — without writing custom API wrappers for every broker.

This list covers every MCP server built for finance, trading, and DeFi as of March 2026. We categorize them by use case so you can find the right one for your stack.

What is an MCP server for trading?

An MCP server exposes tools that an AI agent can call. For trading, these tools typically fall into four categories: market data (prices, OHLCV, fundamentals), trade execution (place/cancel orders), memory and analytics (store trades, discover patterns), and risk management (position sizing, circuit breakers).

You install them via pip install or npx, add them to your MCP client config, and your agent gains access to the tools.

Complete list

TradeMemory ProtocolPythonTrading Memory15 tools

Persistent memory layer for AI trading agents. Three-layer architecture (raw trades → pattern discovery → strategy adjustment) with 15 MCP tools, Outcome-Weighted Memory, and an Evolution Engine that discovers strategies from raw price data.

Platforms: MT5, Binance, Alpaca, any platform · GitHub

CCXT MCP ServerTypeScriptExchange Data

Real-time crypto market data and trading via 20+ exchanges using the CCXT library. Supports spot, futures, OHLCV, balances, and order management.

Platforms: Binance, Coinbase, Kraken, 20+ exchanges · GitHub

Yahoo Finance integration for stock market data including options recommendations.

Platforms: US stocks · GitHub

VibeTrader MCPPythonTrading Bot

AI-powered trading bot platform. Create automated trading strategies with natural language via Alpaca brokerage.

Platforms: Alpaca · GitHub

Tasty AgentPythonBrokerage

Tastyworks API integration to handle trading activities on Tastytrade.

Platforms: Tastytrade · GitHub

FinancialData.Net MCPPythonMarket Data

End-of-day and intraday stock market data, financial statements, insider trading data, sustainability data, and earnings releases.

Platforms: US, global equities · GitHub

Stooq MCPRustMarket Data

Fetch real-time stock prices from Stooq without API keys. Supports US, Japan, UK, and Germany markets.

Platforms: US, JP, UK, DE equities · GitHub

Perp CLITypeScriptDeFi Trading18 tools

Multi-DEX perpetual futures trading. 18 tools for market data, trade execution with dry-run safety, funding rate arbitrage scanning, and portfolio analytics.

Platforms: Pacifica (Solana), Hyperliquid, Lighter (ETH) · GitHub

CryptoGuard ClientPythonRisk Management

Per-transaction deterministic crypto validator. Validate trades (PROCEED/CAUTION/BLOCK), scan tokens, detect rug pulls.

Platforms: Crypto · GitHub

BaoStock MCPPythonMarket Data

Access and analysis capabilities for Chinese stock market data via baostock.

Platforms: China A-shares · GitHub

Finmap MCPPythonMarket Data

Historical data from US, UK, Russian and Turkish stock exchanges. Sectors, tickers, company profiles, market cap, volume, treemap and histogram visualizations.

Platforms: US, UK, RU, TR equities · GitHub

How to choose

  • Need market data only? Use CCXT (crypto), FinancialData.Net (equities), or Stooq (global, no API key).
  • Need to execute trades? VibeTrader (Alpaca), Tasty Agent (Tastytrade), or Perp CLI (DeFi perps).
  • Need your agent to remember and learn from trades? TradeMemory Protocol — the only MCP server with persistent trade memory, pattern discovery, and strategy evolution.
  • Need risk checks before execution? CryptoGuard for crypto token validation.

Missing from this list?

New MCP servers ship every week. If you know one we missed, open an issue and we will add it.


Built by Mnemox AI. We build open-source MCP servers for trading memory and idea validation.