Stock Research

An MCP (Model Context Protocol) server that provides comprehensive stock research tools for Claude. Aggregates data from Alpha Vantage and Finnhub APIs.

Features

• Real-time quotes and historical price data
• Company profiles and fundamental analysis
• Technical indicators (SMA, EMA, RSI, MACD, Bollinger Bands)
• Support/resistance level calculations
• Analyst ratings and price targets
• News sentiment analysis
• Insider trading activity
• Macro economic indicators
• Deep research reports combining all data sources
• Trading execution via Alpaca (paper and live trading)

Installation

1. Clone the repository:

cd stock-research-mcp

2. Create a virtual environment and install the package:

uv venv .venv
source .venv/bin/activate
uv pip install -e ".[dev]"

3. Configure API keys in .env:

ALPHA_VANTAGE_API_KEY=your_key_here
FINNHUB_API_KEY=your_key_here

# Alpaca Trading (optional)
ALPACA_API_KEY=your_key_here
ALPACA_SECRET_KEY=your_secret_here
ALPACA_PAPER=true  # Set to false for live trading

# Risk Controls (optional)
TRADING_MAX_ORDER_VALUE=5000
TRADING_MAX_POSITION_SIZE=10000
TRADING_ALLOWED_SYMBOLS=AAPL,MSFT,GOOGL  # Leave empty for all symbols

Get free API keys at:

• Alpha Vantage: https://www.alphavantage.co/support/#api-key
• Finnhub: https://finnhub.io/register
• Alpaca: https://alpaca.markets/ (free paper trading account)

Usage with Claude Code

Add to your Claude Code MCP configuration:

claude mcp add stock-research -- .venv/bin/python -m stock_research.server

Or use the installed script:

claude mcp add stock-research -- .venv/bin/stock-research-mcp

Or manually add to ~/.claude/mcp.json:

{
  "mcpServers": {
    "stock-research": {
      "command": "/path/to/stock-research-mcp/.venv/bin/stock-research-mcp"
    }
  }
}

Available Tools

Market Data

• get_quote(ticker) - Real-time quote
• get_historical_prices(ticker, timeframe, interval) - Historical OHLCV data

Company Analysis

• get_company_profile(ticker) - Company overview
• get_financials(ticker) - Key financial metrics and ratios
• get_earnings(ticker) - Earnings history and surprises

Technical Analysis

• get_technical_indicators(ticker, indicators) - SMA, EMA, RSI, MACD, BBands
• get_support_resistance(ticker, lookback_days) - Support/resistance levels

Sentiment & Activity

• get_news_sentiment(ticker, limit) - News articles with sentiment
• get_insider_trades(ticker) - Insider buying/selling
• get_analyst_ratings(ticker) - Analyst consensus and price targets

Macro

• get_macro_context() - Economic indicators (Fed rate, yields, GDP, CPI)

Research

• generate_deep_research(ticker, research_type) - Comprehensive research report

Trading (Alpaca)

• get_trading_account() - Account info, buying power, equity
• get_positions() - All open positions
• get_position(symbol) - Position for specific symbol
• place_market_order(symbol, qty, side) - Market order
• place_limit_order(symbol, qty, side, limit_price) - Limit order
• place_stop_order(symbol, qty, side, stop_price) - Stop order
• place_stop_limit_order(symbol, qty, side, stop_price, limit_price) - Stop-limit order
• get_orders(status) - List orders (open/closed/all)
• cancel_order(order_id) - Cancel specific order
• cancel_all_orders() - Cancel all open orders
• close_position(symbol) - Close position for symbol
• close_all_positions() - Liquidate all positions
• get_trading_config() - View current risk limits

Development

Run tests:

pytest

Example Usage

Once configured, ask Claude:

"Give me a deep research report on NVDA"
"What's the current price and technical setup for AAPL?"
"Show me analyst ratings for MSFT"
"What's the macro environment looking like?"

Trading Examples

"Show me my current positions"
"Buy 10 shares of AAPL at market"
"Place a limit order for 5 shares of MSFT at $400"
"Set a stop loss on my NVDA position at $120"
"What's my account buying power?"

⚠️ Warning: Trading involves real money. Start with paper trading (ALPACA_PAPER=true) to test your strategies. The server includes risk controls but you are responsible for your trades.

Rate Limits

• Alpha Vantage Free Tier: 25 API calls/day
• Finnhub Free Tier: 60 API calls/minute

The server uses aggressive caching to stay within free tier limits.

License

MIT