home / mcp / massive.com mcp server

Massive.com MCP Server

Provides access to Massive.com data via an MCP server with four composable tools and in-memory DataFrame storage for SQL queries and analytics.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "massive-com-mcp_massive": {
      "command": "uv",
      "args": [
        "run",
        "mcp_massive"
      ],
      "env": {
        "HOME": "<your_home_directory>",
        "MCP_TRANSPORT": "streamable-http",
        "MASSIVE_API_KEY": "YOUR_API_KEY",
        "POLYGON_API_KEY": "<POLYGON_API_KEY>",
        "MASSIVE_MAX_ROWS": "50000",
        "MASSIVE_MAX_TABLES": "50",
        "MASSIVE_API_BASE_URL": "https://api.massive.com",
        "MASSIVE_LLMS_TXT_URL": "https://massive.com/docs/rest/llms.txt"
      }
    }
  }
}

Massive.com MCP Server provides a single MCP endpoint surface that enables an LLM to access the full Massive.com financial data API through four composable tools. It stores results in-memory as DataFrames, supports SQL querying, and exposes built-in financial analytics to enrich responses.

How to use

You will run the MCP server locally and connect your MCP client to it. Once running, you can prompt your client to search for endpoints, retrieve endpoint documentation, call any Massive.com API endpoint, or run SQL queries against in-memory DataFrames. Use natural language prompts to explore the API surface, fetch parameter documentation, execute API calls, and analyze data with built-in financial functions.

How to install

Prerequisites and commands you need to set up the Massive.com MCP Server are shown here. Follow these steps exactly to ensure the server runs reliably with your client.

uv tool install "mcp_massive @ git+https://github.com/massive-com/[email protected]"

Register the MCP server with Claude Code so your client can connect. Replace YOUR_API_KEY with your actual Massive.com API key.

claude mcp add massive -e MASSIVE_API_KEY=YOUR_API_KEY -- mcp_massive

To upgrade to a newer MCP server version later, run the following.

uv tool upgrade mcp_massive

If you previously used older installation methods, switch to the current approach shown above. This downloads dependencies once and starts instantly after that.

Configuration and startup

Run the MCP server locally using the standard startup command. You must provide your Massive.com API key as an environment variable. You can also specify the transport method if you want to use a different communication channel.

uv run mcp_massive

Example with an API key and a different transport (optional):

MASSIVE_API_KEY=your_api_key_here uv run mcp_massive --transport streamable-http

Notes on transport and environment variables

The server defaults to stdio transport. You can switch to streamable-http by setting the transport option. The API key is required for authentication, and you may configure additional environment variables as needed.

Available tools

search_endpoints

Search for API endpoints and built-in functions by natural language query. Returns names, URL patterns, and descriptions. Supports scoping to endpoints, functions, or all.

get_endpoint_docs

Get parameter documentation for a specific endpoint by passing the docs URL from search results.

call_api

Call any Massive.com REST API endpoint. Can store results as an in-memory DataFrame (store_as) and apply post-processing functions (apply). Handles paginated responses with a next-page hint.

query_data

Run SQL against stored DataFrames using SQLite. Supports SHOW TABLES, DESCRIBE <table>, DROP TABLE <table>, CTEs, window functions, and more. Results can be post-processed with apply.

bs_price

Black-Scholes option price using the Greeks functions.

bs_delta

Black-Scholes delta using the Greeks functions.

bs_gamma

Black-Scholes gamma using the Greeks functions.

bs_theta

Black-Scholes theta using the Greeks functions.

bs_vega

Black-Scholes vega using the Greeks functions.

bs_rho

Black-Scholes rho using the Greeks functions.

simple_return

Compute simple returns from price data.

log_return

Compute logarithmic returns from price data.

cumulative_return

Compute cumulative returns over time.

sharpe_ratio

Calculate the Sharpe ratio for performance analysis.

sortino_ratio

Calculate the Sortino ratio for downside-risk-adjusted performance.

sma

Compute simple moving average.

ema

Compute exponential moving average.