Market Data

You can run the Market Data MCP Server to provide real-time market data to any MCP-compatible AI assistant. It exposes a suite of financial data tools and transports, letting your agent fetch quotes, movers, screeners, indicators, and catalysts without requiring API keys. This setup is designed for straightforward local use or network access via SSE or Streamable HTTP.

How to use

To connect an MCP client to the Market Data MCP Server, run it locally in the background and point your client to the appropriate transport. You can use the standard local transport for Claude Desktop, or enable network access via SSE or Streamable HTTP.

Default local transport (stdin/stdio) connects directly for Claude Desktop. You can also expose the server over the network using SSE on port 8100 or Streamable HTTP on port 8100. Use these transports to query any of the available market data tools from your MCP client.

How to install

Prerequisites: ensure you have Python 3.10 or newer installed on your system.

1. Clone the repository and enter the project directory.

git clone https://github.com/smashtoes/market-data-mcp.git
cd market-data-mcp

2. Install dependencies in editable mode so you can run the server directly from the source.

pip install -e .

3. Start the server with the default local transport. This runs the server as a standard MCP provider for Claude Desktop.

python server.py

If you want network access, run the server with the desired transport and port.

# For SSE network access on port 8100
python server.py --transport sse --port 8100

# For Streamable HTTP network access on port 8100
python server.py --transport streamable-http --port 8100

Additional sections

What you get from this server: eleven financial tools cover equities, crypto, screening, macro data, and biotech catalysts. All tools return structured JSON even on failure, and there are built-in fallbacks for resilience. The server is production-quality with type hints, docstrings, logging, and a clean architecture.

Configuration and troubleshooting tips are included to help you maintain a stable setup. No API keys are required, and all data sources are free and public, with a local FDA calendar data file used for biotech catalysts.

If you plan to use Claude Desktop, add the market data MCP server to your Claude Desktop configuration to enable the tools within the chat interface. If you prefer a CLI-driven workflow, you can connect via the Claude Code MCP interface using the provided commands.

Available tools

get_stock_quote

Real-time stock price, change, volume, market cap, PE ratio, and 52-week range for a given ticker.

get_crypto_price

Cryptocurrency price with 24h/7d/30d changes, market cap, and ATH distance.

get_market_movers

Today’s top gainers and losers from a universe of liquid stocks.

get_sector_performance

Performance data for all 11 commonly tracked sector ETFs over 1 day, 5 days, and 1 month.

get_earnings_calendar

Upcoming earnings reports with EPS and revenue estimates.

get_technical_indicators

Technical indicators such as RSI, MACD, SMA/EMA, Bollinger Bands, and volume analysis with interpreted signals.

get_fear_greed_index

CNN Fear & Greed Index score with historical comparison and fallback behavior when the primary source is unavailable.

get_fda_calendar

Upcoming FDA PDUFA drug approval dates and major biotech catalysts.

screen_stocks

TradingView-style stock screener with filters for market, sector, price, volume, and technicals.

screen_crypto

TradingView-style crypto screener with market cap filtering to reduce exchange noise.

top_movers

Market-wide top movers computed via screener engine for gainers, losers, and activity.