Trading MCP Server

What You Can Do

• Account & portfolio: Check balances, buying power, positions, portfolio history, and account activity
• Trade: Place and manage orders for stocks, options, and crypto (market, limit, stop, trailing stop, brackets, multi-leg spreads)
• Market data: Query historical and live data: bars, quotes, trades, snapshots, option chains with Greeks
• Discover: Screen for most active stocks, market movers, news, corporate actions
• Organize: Create and manage watchlists

Prerequisites

• Python 3.10+ (official installation guide)
• uv with uvx (install guide)
• Alpaca API keys: a free paper trading account is enough to start
• MCP client: Claude Desktop, Cursor, VS Code, Claude Code, Gemini CLI, PyCharm, or any MCP-compatible tool

Setup

Add the server to your MCP client configuration with your API keys. No init commands or additional setup required.

Cursor

Add to ~/.cursor/mcp.json:

{
  "mcpServers": {
    "alpaca": {
      "command": "uvx",
      "args": ["alpaca-mcp-server"],
      "env": {
        "ALPACA_API_KEY": "your_alpaca_api_key",
        "ALPACA_SECRET_KEY": "your_alpaca_secret_key"
      }
    }
  }
}

Claude Desktop

Add to your Claude Desktop config (claude_desktop_config.json):

{
  "mcpServers": {
    "alpaca": {
      "command": "uvx",
      "args": ["alpaca-mcp-server"],
      "env": {
        "ALPACA_API_KEY": "your_alpaca_api_key",
        "ALPACA_SECRET_KEY": "your_alpaca_secret_key"
      }
    }
  }
}

VS Code

Add to your VS Code settings (.vscode/mcp.json):

{
  "servers": {
    "alpaca": {
      "command": "uvx",
      "args": ["alpaca-mcp-server"],
      "env": {
        "ALPACA_API_KEY": "your_alpaca_api_key",
        "ALPACA_SECRET_KEY": "your_alpaca_secret_key"
      }
    }
  }
}

Claude Code (CLI)

claude mcp add alpaca --scope user --transport stdio uvx alpaca-mcp-server \
  --env ALPACA_API_KEY=your_alpaca_api_key \
  --env ALPACA_SECRET_KEY=your_alpaca_secret_key

Docker

git clone https://github.com/alpacahq/alpaca-mcp-server.git
cd alpaca-mcp-server
docker build -t mcp/alpaca:latest .
docker run -e ALPACA_API_KEY=your_key -e ALPACA_SECRET_KEY=your_secret mcp/alpaca:latest

For other clients (Gemini CLI, PyCharm, ChatGPT), see the README setup section.

Verify

Restart your MCP client, then ask:

"What is my Alpaca account balance and buying power?"

If you get account details back, the server is connected.

Configuration

All configuration is via environment variables in your MCP client config.

Toolset Filtering

Control which tools are available to your AI assistant using ALPACA_TOOLSETS. This is useful for limiting scope (e.g. read-only market data) or staying within MCP client tool limits.

Example — read-only market data:

{
  "env": {
    "ALPACA_TOOLSETS": "account,stock-data,crypto-data,options-data,news"
  }
}

Supported MCP Clients

Available Tools

Alpaca's MCP Server exposes 65 tools across the Trading and Market Data APIs.

Account & Portfolio

Orders

Positions

Watchlists

Assets & Market Info

Stock Data

Crypto Data

Options Data

Fixed Income Data

Index Data

Corporate Actions & News

Example Prompts

Here are some things you can ask once connected:

Account & Portfolio

• "What's my current account balance and buying power?"
• "Show me all my open positions."
• "What's my portfolio performance over the last month?"

Trading

• "Buy 10 shares of AAPL at market price."
• "Place a limit order to sell 50 shares of TSLA at $350."
• "Set a trailing stop on my NVDA position at 5%."
• "Cancel all my open orders."

Market Data

• "Show me AAPL's daily price bars for the last week."
• "What's the latest quote for BTC/USD?"
• "Get a snapshot of MSFT including latest trade and quote."
• "What are the most active stocks right now?"

Options

• "Show me available AAPL options expiring next month."
• "What are the Greeks for this TSLA 300 put?"
• "Place a bull call spread on SPY."

Watchlists

• "Create a watchlist called Tech Earnings with AAPL, MSFT, NVDA, and GOOGL."

Tip: Use paper trading API keys to test strategies safely. Switch to live trading only when you set ALPACA_PAPER_TRADE=false with live API keys.

Important Considerations

• API key security: Never paste API keys into chat. Configure them only in your MCP client's env block or OS secret store.
• Order execution: Orders execute directly against Alpaca's Trading API. Review AI-suggested orders before confirming.
• Paper vs. live: The server defaults to paper trading. Set ALPACA_PAPER_TRADE=false with live keys to trade with real capital.
• Rate limits: Alpaca's API has rate limits per account. High-frequency querying may trigger rate limiting.
• Market data subscriptions: Some real-time data may require an Algo Trader Plus subscription.

Upgrading from V1

V2 is not a drop-in replacement for V1. Tool names and parameters changed. If you need V1, pin an older PyPI version:

{
  "args": ["alpaca-mcp-server==1.x.x", "serve"]
}

For details, see the Upgrading from V1 section in the README.

Resources

• GitHub Repository
• V2 Launch Blog Post
• Cursor Directory
• Contributing & Development (AGENTS.md)
• Alpaca Support

Disclosure

Insights generated by our MCP server and connected AI agents are for educational and informational purposes only and should not be taken as investment advice. Past performance from models does not guarantee future results. Please conduct your own due diligence before making any decisions.