[!IMPORTANT] :test_tube: This project is experimental and could be subject to breaking changes.

Massive.com MCP Server

A Model Context Protocol (MCP) server that provides access to the full Massive.com financial data API through an LLM-friendly interface.

Rather than exposing one tool per endpoint, this server gives the LLM three composable tools — search, call, and query — that cover the entire Massive.com API surface. Data can be stored in an in-memory SQLite database, and enriched with built-in financial functions.

Tools

Built-in Functions

Functions can be applied to API results or query output via the apply parameter on call_api and query_data. Use search_endpoints with scope="functions" to discover them.

Data Coverage

The server dynamically indexes all Massive.com API endpoints at startup from llms.txt, so it automatically stays in sync with the API. Coverage includes:

• Stock, options, forex, crypto, and futures aggregates
• Real-time and historical trades and quotes
• Market snapshots, gainers/losers
• Ticker details and reference data
• Dividends, splits, IPOs
• Financial fundamentals
• Analyst ratings and news (Benzinga)
• Treasury yields, inflation data
• Market status and holidays

Installation

Prerequisites

• Python 3.12+
• A Massive.com API key
  
• Astral UV (v0.4.0+)

Claude Code

First, install Claude Code

npm install -g @anthropic-ai/claude-code

Install the MCP server, then register it with Claude Code:

# Install the server (one-time — downloads dependencies ahead of time)
uv tool install "mcp_massive @ git+https://github.com/massive-com/[email protected]"

# Register with Claude Code
claude mcp add massive -e MASSIVE_API_KEY=your_api_key_here -- mcp_massive

To upgrade to a new version later:

uv tool upgrade mcp_massive

[!NOTE] Upgrading from uvx or uv run --with? Previous versions recommended uvx --from ... mcp_massive or uv run --with. These commands download dependencies on every cold start, which can cause the server to exceed Claude's 30-second connection timeout. Switch to uv tool install as shown above — it downloads dependencies once and starts instantly after that.

This command will install the MCP server in your current project. If you want to install it globally, you can run the command with -s <scope> flag. See claude mcp add --help for more options.

To start Claude Code, run claude in your terminal.

• If this is your first time using, follow the setup prompts to authenticate

You can also run claude mcp add-from-claude-desktop if the MCP server is installed already for Claude Desktop.

Claude Desktop

1. Follow the Claude Desktop MCP installation instructions to complete the initial installation and find your configuration file.
2. Install the server:

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

3. Find the installed binary path:

# Mac/Linux
which mcp_massive

# Windows
where mcp_massive

4. Add the server to your Claude Desktop config. Replace <path_to_mcp_massive> with the output from the previous step, and fill in the remaining fields.

{
    "mcpServers": {
        "massive": {
            "command": "<path_to_mcp_massive>",
            "env": {
                "MASSIVE_API_KEY": "<your_api_key_here>",
                "HOME": "<your_home_directory>"
            }
        }
    }
}

Configuration

Environment Variables

Transport

By default, STDIO transport is used. The transport can be set via the --transport CLI argument or the MCP_TRANSPORT environment variable (CLI argument takes precedence).

# CLI argument
MASSIVE_API_KEY=<your_api_key_here> uv run mcp_massive --transport streamable-http

# Environment variable
MCP_TRANSPORT=streamable-http MASSIVE_API_KEY=<your_api_key_here> uv run mcp_massive

Usage Examples

Once integrated, you can prompt Claude to access Massive.com data:

Get the latest price for AAPL stock
Show me yesterday's trading volume for MSFT
What were the biggest stock market gainers today?
Get me the latest crypto market data for BTC-USD
Calculate the 20-day SMA for AAPL closing prices over the last 3 months
Compute Black-Scholes delta for these option contracts

Development

Running Locally

Check to ensure you have the Prerequisites installed.

# Sync dependencies
uv sync

# Run the server
MASSIVE_API_KEY=your_api_key_here uv run mcp_massive

uv tool install --force /path/to/mcp_massive

{
  "mcpServers": {
    "massive": {
      "command": "mcp_massive",
      "env": {
        "MASSIVE_API_KEY": "your_api_key_here"
      }
    }
  }
}

Debugging

For debugging and testing, we recommend using the MCP Inspector:

npx @modelcontextprotocol/inspector uv --directory /path/to/mcp_massive run mcp_massive

This will launch a browser interface where you can interact with your MCP server directly and see input/output for each tool.

Select the STDIO transport type in the browser UI with the command uv and run mcp_massive as the arguments.

Code Linting

This project uses just for common development tasks. To lint your code before submitting a PR:

just lint

This will run ruff format and ruff check --fix to automatically format your code and fix linting issues.

Links

• Massive.com Documentation
• Model Context Protocol
• MCP Python SDK

Privacy Policy

This MCP server interacts with Massive.com's API to fetch market data. All data requests are subject to Massive.com's privacy policy and terms of service.

• Massive.com Privacy Policy: https://massive.com/legal/privacy
• Data Handling: This server does not store or cache any user data. All requests are proxied directly to Massive.com's API.
• API Key: Your Massive.com API key is used only for authenticating requests to their API.
• User-Agent: API requests include a User-Agent string containing the MCP server version (e.g., MCP-Massive/0.x.y). No personally identifiable information is included.

Contributing

If you found a bug or have an idea for a new feature, please first discuss it with us by submitting a new issue. We will respond to issues within at most 3 weeks. We're also open to volunteers if you want to submit a PR for any open issues but please discuss it with us beforehand. PRs that aren't linked to an existing issue or discussed with us ahead of time will generally be declined.