Alpaca’s MCP server lets LLM clients (Claude Desktop, Cursor, VS Code, etc.) trade stocks & options, manage portfolios, and stream market data through Alpaca’s API via natural-language.
https://github.com/alpacahq/alpaca-mcp-serverStop switching between trading platforms, data feeds, and analysis tools. Alpaca's MCP server brings professional-grade trading directly into Claude Desktop, Cursor, VS Code, and any MCP-compatible environment. Execute trades, analyze market data, and manage portfolios using plain English—no GUI clicking required.
Traditional trading means juggling multiple interfaces: one for market data, another for order entry, a third for portfolio analysis. You're constantly context-switching between tools, manually copying symbols and prices, and translating your analysis into concrete actions.
This MCP server eliminates that friction. Ask Claude "What's my buying power and current AAPL position?" or tell it "Place a bull call spread on SPY using July options, 5% OTM" and watch it execute directly through Alpaca's infrastructure. The same conversation that analyzes your strategy can place the trades.
Complete Trading Operations
Advanced Options Support
Intelligent Market Data
Portfolio Intelligence
Instead of opening multiple browser tabs and trading apps, your entire workflow lives in one conversation:
"Show me my current positions and account balance"
→ Displays complete portfolio overview
"Get me a quote on NVDA and show the next earnings date"
→ Real-time quote plus corporate announcement data
"Place a bull call spread on AAPL using June 6th options: buy the 190 strike, sell the 200 strike"
→ Executes multi-leg options order with full confirmation details
"Create a watchlist called 'AI Stocks' with NVDA, AMD, and GOOGL"
→ Saves watchlist for future monitoring
Each command executes directly through Alpaca's trading infrastructure with institutional-grade execution quality. No manual order entry, no copying symbols between applications.
The server supports both paper trading (for testing strategies risk-free) and live trading with real capital. Configuration takes minutes:
git clone https://github.com/alpacahq/alpaca-mcp-server.git
cd alpaca-mcp-server
python -m venv venv && source venv/bin/activate
pip install -r requirements.txt
Add your Alpaca API credentials to your environment or Claude Desktop config, and you're trading through natural language. The server includes Docker support for containerized deployments and handles all the API authentication automatically.
For Claude Desktop integration, you just point to your local Python environment in the MCP configuration. The server manages the connection to Alpaca's infrastructure while your AI assistant handles the natural language interpretation.
This isn't a third-party integration—it's official tooling from Alpaca themselves. They understand that modern developers and traders want to incorporate AI into their decision-making process without abandoning professional trading infrastructure.
The MCP server bridges that gap, giving you institutional-quality execution through conversational interfaces. Whether you're backtesting strategies, managing existing positions, or executing complex options trades, the natural language interface removes the mechanical overhead of traditional trading platforms.
Security Note: The server can place real trades with real money. Start with paper trading to test your workflows, and treat your API credentials with the same security you'd use for direct access to your brokerage account.
Ready to trade through conversation rather than clicking? Clone the repository and connect your first AI-powered trading session in under 10 minutes.