Twelve Data

The Twelve Data MCP Server lets you access financial market data from the Twelve Data API through a simple, centralized MCP interface. It supports historical time series, real-time quotes, and instrument metadata for stocks, forex, and crypto, with flexible options for local or remote operation and easy integration with Claude Desktop or VS Code.

How to use

You interact with the Twelve Data MCP Server through an MCP client. Start the server locally or remotely, then point your MCP client to the server’s endpoint. Use the client’s standard MCP workflow to request data such as historical series, real-time quotes, or instrument metadata. You can describe your data needs in natural language when using assisted routing tools, and the system will route to the appropriate Twelve Data endpoints and return formatted results.

How to install

Prerequisites you should have installed on your machine before running the server: Node.js (for UDP/UV-based usage) and Python (for pip-based installation). You can also run the server via Docker if you prefer containerized execution.

# Option 1: Run directly with UV (recommended for quick trials)
uvx mcp-server-twelve-data --help

# Option 2: Install via Python's pip
pip install mcp-server-twelve-data
python -m mcp_server_twelve_data --help

# Option 3: Run with Docker
docker build -t mcp-server-twelve-data .

# Then start the container with your API key and OpenAI key
docker run --rm mcp-server-twelve-data \
  -k YOUR_TWELVE_DATA_API_KEY \
  -u YOUR_OPENAI_API_KEY \
  -t streamable-http

Configuration and usage notes

Configure how you connect Claude Desktop or VS Code to the Twelve Data MCP Server by selecting one of the available setup options shown below. Each option is designed to be straightforward and works with the MCP client you choose.

{
  "mcpServers": {
    "twelvedata": {
      "command": "uvx",
      "args": ["mcp-server-twelve-data@latest", "-k", "YOUR_TWELVE_DATA_API_KEY", "-u", "YOUR_OPEN_AI_APIKEY"]
    }
  }
}

{
  "mcpServers": {
    "twelvedata": {
      "command": "uvx",
      "args": ["mcp-server-twelve-data@latest", "-k", "YOUR_TWELVE_DATA_API_KEY", "-n", "10"]
    }
  }
}

{
  "mcpServers": {
    "twelvedata-remote": {
      "command": "npx",
      "args": [
        "mcp-remote", "https://mcp.twelvedata.com/mcp",
        "--header",
        "Authorization:${AUTH_HEADER}",
        "--header",
        "X-OpenAPI-Key:${OPENAI_API_KEY}"
      ],
      "env": {
        "AUTH_HEADER": "apikey YOUR_TWELVE_DATA_API_KEY",
        "OPENAI_API_KEY": "YOUR_OPENAI_API_KEY"
      }
    }
  }
}

Troubleshooting and debugging

If you run into issues, use the MCP Inspector to diagnose configuration and routing problems. It lets you observe how requests are processed and where to adjust parameters.

npx @modelcontextprotocol/inspector uvx mcp-server-twelve-data@latest -k YOUR_TWELVE_DATA_API_KEY

Docker usage

If you prefer containerized execution, build and run the server in Docker. The following commands build the image and start the container with your API keys and an HTTP stream target.

docker build -t mcp-server-twelve-data .

docker run --rm mcp-server-twelve-data \
  -k YOUR_TWELVE_DATA_API_KEY \
  -u YOUR_OPENAI_API_KEY \
  -t streamable-http

Available tools

u-tool

An AI-powered universal router for the Twelve Data API that understands natural language requests, routes to the right endpoints, and returns formatted results.

inspector

MCP Inspector tool for debugging and troubleshooting MCP server configurations.