trackmcp
Back to directory
polygon-io

mcp_massive

View on GitHub

An MCP server for Massive.com Financial Market Data Python-based implementation.

191 stars PythonAI & Machine Learning Updated Nov 4, 2025
cryptocurrencyfintechllmmarket-datamcpmcp-serveroptionsstocks

Documentation

> [!IMPORTANT]

> :test_tube: This project is experimental and could be subject to breaking changes.

Massive.com MCP Server

GitHub release

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

ToolDescription
`search_endpoints`Search for API endpoints and built-in functions by natural language query. Returns titles, path patterns, and descriptions. Set `detail` to `"more"` for query parameter docs, or `"verbose"` for full documentation. Use `max_results` to limit results.
`call_api`Call any Massive.com REST API endpoint. Supports storing results as an in-memory database table (`store_as`) and applying post-processing functions (`apply`). Paginated responses include a next-page hint.
`query_data`Run SQL against stored SQLite DB. Supports `SHOW TABLES`, `DESCRIBE `, `DROP TABLE `, CTEs, window functions, and more. Results can also be post-processed with `apply`.

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.

CategoryFunctions
Greeks`bs_price`, `bs_delta`, `bs_gamma`, `bs_theta`, `bs_vega`, `bs_rho` — Black-Scholes option pricing and greeks
Returns`simple_return`, `log_return`, `cumulative_return`, `sharpe_ratio`, `sortino_ratio`
Technical`sma` (simple moving average), `ema` (exponential moving average)

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 [![Button]][Link]
  • Astral UV (v0.4.0+)

Claude Code

First, install Claude Code

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

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

bash
# Install the server (one-time — downloads dependencies ahead of time)
uv tool install "mcp_massive @ git+https://github.com/massive-com/mcp_massive@v0.10.0"

# 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:

bash
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 ` 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.

1. Install the server:

bash
uv tool install "mcp_massive @ git+https://github.com/massive-com/mcp_massive@v0.10.0"

3. Find the installed binary path:

bash
# Mac/Linux
which mcp_massive

# Windows
where mcp_massive

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

claude_desktop_config.json

json
{
    "mcpServers": {
        "massive": {
            "command": "",
            "env": {
                "MASSIVE_API_KEY": "",
                "HOME": ""
            }
        }
    }
}

Configuration

Environment Variables

VariableRequiredDefaultDescription
`MASSIVE_API_KEY`YesYour Massive.com API key
`POLYGON_API_KEY`NoDeprecated alias for `MASSIVE_API_KEY`
`MCP_TRANSPORT`No`stdio`Transport protocol: `stdio`, `sse`, or `streamable-http`
`MASSIVE_API_BASE_URL`No`https://api.massive.com`Base URL for API requests
`MASSIVE_LLMS_TXT_URL`No`https://massive.com/docs/rest/llms.txt`URL for the endpoint index
`MASSIVE_MAX_TABLES`No`50`Maximum number of in-memory tables
`MASSIVE_MAX_ROWS`No`50000`Maximum rows per stored tables

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).

bash
# CLI argument
MASSIVE_API_KEY= uv run mcp_massive --transport streamable-http

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

Usage Examples

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

code
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.

bash
# Sync dependencies
uv sync

# Run the server
MASSIVE_API_KEY=your_api_key_here uv run mcp_massive

Local Dev Config for claude_desktop_config.json

Install from your local checkout, then reference the binary directly:

bash
uv tool install --force /path/to/mcp_massive
json
{
  "mcpServers": {
    "massive": {
      "command": "mcp_massive",
      "env": {
        "MASSIVE_API_KEY": "your_api_key_here"
      }
    }
  }
}

Debugging

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

bash
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:

bash
just lint

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

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.

[Link]: https://massive.com/?utm_campaign=mcp&utm_medium=referral&utm_source=github 'Massive.com Home Page'

[Button]: https://img.shields.io/badge/Get_One_For_Free-5F5CFF?style=for-the-badge&logoColor=white

Frequently asked questions

What is mcp_massive?

mcp_massive is An MCP server for Massive.com Financial Market Data Python-based implementation.

How do I install mcp_massive?

Open the GitHub repository and follow its README. Most MCP servers are added to your client's MCP config, then called by your agent.

Is mcp_massive open source?

Yes — it is hosted on GitHub at https://github.com/polygon-io/mcp_polygon and has 191 stars.

Related MCP tools

tadata-orgfastapi_mcp

Expose your FastAPI endpoints as Model Context Protocol (MCP) tools, with Auth! Python-based implementation. Trusted by 11000+ developers.

11,002 Python
aiauthenticationauthorization+10
qdrantmcp-server-qdrant

An official Qdrant Model Context Protocol (MCP) server implementation Python-based implementation. Trusted by 1000+ developers.

1,043 Python
claudecursorllm+4
TickDBtickdb-unified-realtime-marketdata-api

TickDB: AI-native real time stock API and market data API for US stocks, HK stocks, A-shares, forex, crypto, indices and commodities. Skill, CLI, MCP, REST API and WebSocket. AI 原生实时股票 API 与金融行情数据 API,支持美股、港股、A 股、外汇、加密货币、指数和大宗商品。

751 Python
commoditiescryptocurrencyfinancial-data+17
zhayujiechatgpt-on-wechat

基于大模型搭建的聊天机器人,同时支持 微信公众号、企业微信应用、飞书、钉钉 等接入,可选择ChatGPT/Claude/DeepSeek/文心一言/讯飞星火/通义千问/ Gemini/GLM-4/Kimi/LinkAI,能处理文本、语音和图片,访问操作系统和互联网,支持基于自有知识库进行定制企业智能客服。

39,573 Python
aiai-agentchatgpt+17
assafelovicgpt-researcher

An LLM agent that conducts deep research (local and web) on any given topic and generates a long report with citations. Built for the Model Context Protocol to

24,026 Python
agentaiautomation+8
1Panel-devmaxkb

🔥 MaxKB is an open-source platform for building enterprise-grade agents. MaxKB 是强大易用的开源企业级智能体平台。 for the Model Context Protocol. Enhance AI assistants with po

19,062 Python
agentagentic-aichatbot+11

Run your own MCP server? See who uses it and what to fix.

Measure it with TrackMCP