- Home
- MCP servers
- Alpaca
Alpaca
- python
538
GitHub Stars
python
Language
4 months ago
First Indexed
3 weeks ago
Catalog Refreshed
Documentation & install
Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.
You can run and connect Alpaca's MCP Server to your preferred MCP client to trade programmatically using natural language prompts. It provides real-time market data, account, positions, orders, and diverse trading capabilities for stocks, crypto, and options within a single, extensible server.
How to use
Connect the Alpaca MCP Server to an MCP client such as Claude Desktop, Cursor, VS Code, or Claude Code. Use the client to issue natural language commands that get translated into trading actions, place orders, fetch market data, manage watchlists, and inspect your portfolio. The server supports real-time quotes, trades, and price bars; historical data; account and position management; and a wide range of order types and options strategies. Start by configuring your MCP server credentials, then add the server to your MCP client of choice. When you issue prompts, the MCP client routes them to the server and returns structured results you can review before execution.
How to install
Prerequisites you need before installing the MCP server.
Then follow concrete installation options to get started quickly.
Security and credentials
Keep your Alpaca API keys secure. Treat all credentials as sensitive. When you use remote hosting or cloud deployments, ensure access is restricted and secured with proper authentication and TLS. The MCP client configuration can override server env settings, so verify that your keys and trading mode (paper vs live) are correct in the appropriate config.
Troubleshooting
If you encounter issues, verify that your environment variables are set, the MCP server is reachable from your MCP client, and the client is restarted after any config change. Check for port conflicts when using HTTP transport and consider using transport with a dedicated port or a proxy with authentication for remote access.
Notes and examples
Several installation and configuration options are provided to suit your environment. You can install locally with uvx, use install.py for Cursor or Claude Desktop, or run via Docker. For remote hosting, you can deploy the MCP server in the cloud and connect it to Claude Web, Claude Mobile, or other MCP clients.
Appendix: Quick configuration references
You can configure the MCP server to run locally using uvx. The following example starts the server and serves MCP requests with environment variables for Alpaca credentials.
Appendix: Available prompts and outputs
Prompts and outputs demonstrate typical trading tasks, such as checking balances, placing orders, retrieving quotes, and inspecting positions. The MCP server returns structured responses that describe the action taken, its status, and any data retrieved or generated.
Available tools
get_account_info
View balance, buying power, and account status
get_all_positions
List all held assets
get_open_position
Detailed info on a specific position
get_asset
Search asset metadata
get_all_assets
List all tradable instruments with filtering options
get_corporate_actions
Historical and future corporate actions (earnings, dividends, splits)
get_portfolio_history
Retrieve account portfolio history with equity and P/L over time
create_watchlist
Create a new watchlist with symbols
get_watchlists
Retrieve all saved watchlists
update_watchlist_by_id
Modify an existing watchlist by ID
get_watchlist_by_id
Get a specific watchlist by ID
add_asset_to_watchlist_by_id
Add an asset to a watchlist
remove_asset_from_watchlist_by_id
Remove an asset from a watchlist
delete_watchlist_by_id
Delete a specific watchlist
get_calendar
Holidays and trading days in a date range
get_clock
Market open/close times and current status
get_stock_bars
OHLCV historical bars with flexible timeframes
get_stock_quotes
Historical quote data (level 1 bid/ask) for a stock
get_stock_trades
Trade-level history for a stock
get_stock_latest_bar
Most recent OHLCV bar for a stock
get_stock_latest_quote
Real-time bid/ask quote for one or more symbols
get_stock_latest_trade
Latest market trade price for a symbol
get_stock_snapshot
Comprehensive snapshot for a stock
get_crypto_bars
Historical price bars for cryptocurrency with configurable timeframe
get_crypto_quotes
Historical quote data for crypto
get_crypto_trades
Historical trade prints for cryptocurrency
get_crypto_latest_quote
Latest quote for one or more crypto symbols
get_crypto_latest_bar
Latest minute bar for crypto
get_crypto_latest_trade
Latest trade for crypto
get_crypto_snapshot
Comprehensive crypto snapshot
get_option_contracts
Get option contracts with flexible filtering
get_option_latest_quote
Latest bid/ask on contract
get_option_snapshot
Get Greeks and underlying
get_orders
Retrieve all or filtered orders
place_stock_order
Place a stock order of any type (market, limit, stop, stop_limit, trailing_stop)
place_crypto_order
Place a crypto order with market, limit, and stop_limit types
place_option_order
Place an options order (single or multi-leg)
cancel_all_orders
Cancel all open orders
cancel_order_by_id
Cancel a specific order
close_position
Close part or all of a position
close_all_positions
Liquidate entire portfolio
exercise_options_position
Exercise a held option contract