robinhood-mcp

Robinhood Portfolio Tracker (CLI)

This project provides a command-line portfolio tracker for Robinhood with login, portfolio viewing, and basic trading. It also includes a paper-trading mode for safe testing.

Important: Real-money trading is risky. Use paper mode first. You are responsible for your account and regulatory compliance.

Features

• Login using username/password or device token with session caching
• View portfolio positions, cash, performance snapshot
• Place market buy/sell orders (limit support planned)
• Paper-trading mode with local state file
• Simple momentum strategy example and rebalance command

Quick Start

1) Create and activate a virtual environment ``bash python3 -m venv .venv && source .venv/bin/activate ``

2) Install dependencies ``bash pip install -r requirements.txt ``

3) Set environment variables ```bash export RH_USERNAME="your_email@example.com" export RH_PASSWORD="your_password" export RH_DEVICE_TOKEN="your_device_token_optional" export RH_PAPER="true" # set to "false" to enable live trading

Optional: if you already have an access token from the web app

export RH_ACCESS_TOKEN="paste_access_token_here" export RH_TOKEN_TYPE="Bearer" ```

4) Run the CLI ``bash python -m robinhood_tracker --help ``

Commands

• login: Authenticate and cache session
• portfolio: Display positions, equity, and cash
• buy --symbol AAPL --quantity 1: Market buy
• sell --symbol AAPL --quantity 1: Market sell
• rebalance --symbols AAPL,MSFT,GOOGL --allocations 40,30,30: Example strategy

Trading Bot Commands

• bot add --symbol AAPL --quantity 10 --stop-loss -5.0 --take-profit 10.0: Add position to monitor
• bot remove --symbol AAPL: Remove position from monitoring
• bot start --interval 5: Start bot (checks every 5 minutes)
• bot stop: Stop the bot
• bot status: Show monitored positions and current P&L
• bot check: Manually check all positions once

MCP (Model Context Protocol) Integration

The project includes a comprehensive MCP adapter for AI model integration.

Available Tools

The MCP adapter provides 15 trading and portfolio management tools:

Portfolio Management

• get_portfolio - Get portfolio summary and positions
• get_health - Check API server status
• login - Authenticate with Robinhood

Trading Operations

• buy_stock - Buy shares with symbol and quantity
• sell_stock - Sell shares with symbol and quantity
• rebalance_portfolio - Auto-rebalance to target allocations

Trading Bot Management

• get_bot_status - Get bot status and monitored positions
• add_bot_position - Add position with stop-loss/take-profit
• remove_bot_position - Remove position from monitoring
• start_bot - Start automated trading bot
• stop_bot - Stop automated trading bot
• check_bot_positions - Manually check all positions

Options Trading

• get_options_positions - Get current options holdings
• get_options_orders - Get recent options orders
• get_options_instruments - Get available options for symbol

Usage Examples

# List all available tools
python3 mcp_robinhood_adapter.py --list-tools

# Get portfolio
python3 mcp_robinhood_adapter.py --tool get_portfolio

# Buy stock
python3 mcp_robinhood_adapter.py --tool buy_stock --params '{"symbol": "AAPL", "quantity": 1}'

# Add position to bot
python3 mcp_robinhood_adapter.py --tool add_bot_position --params '{"symbol": "AAPL", "quantity": 1, "stop_loss": 5, "take_profit": 10}'

AI Model Integration

Configure your AI model to use the MCP server:

{
  "mcpServers": {
    "robinhood": {
      "command": "python3",
      "args": ["/path/to/robinhood_tracker/mcp_robinhood_adapter.py"]
    }
  }
}

Security Notes

• Never commit credentials. Use environment variables or a .env file not tracked by git.
• Paper mode writes local state to .paper_state.json.
• MCP server communicates with Python backend via local HTTP API (127.0.0.1:5000).

Disclaimer

This is not financial advice. Use at your own risk.