Turnono Datacommons MCP Server

What turnono-datacommons-mcp-server does

The turnono-datacommons-mcp-server MCP server provides an MCP interface to Data Commons API data. It is intended for clients or agents that need to find statistical indicators and topics, then request observations or related data.

Its supported capabilities include searching the available indicators and topics, retrieving observations, and working with different data formats and chart configurations. The README does not specify individual tool names or provide a complete list of the supported formats, so the exact tool schema should be checked in the installed server or client connection.

How it works

The server can communicate with an MCP client in either of two transport modes:

• HTTP, for a network-accessible service
• stdio, for a process launched and managed by an MCP client

In both modes, the server acts as the MCP-facing layer while obtaining data through the Data Commons API. HTTP mode can bind to a chosen host and port. The documented example uses host 0.0.0.0 and port 8000; stdio mode starts without those network arguments.

Setup and configuration

To install turnono-datacommons-mcp-server from the repository, install the Python requirements and the package itself:

pip install -r requirements.txt
pip install -e .

The server requires the GOOGLE_API_KEY environment variable for Data Commons access. Set it in the environment used to launch the process. The README does not describe key scopes, account setup, or additional configuration variables.

Start an HTTP instance with:

python -m datacommons_mcp.cli serve http --host 0.0.0.0 --port 8000

Start a stdio instance with:

python -m datacommons_mcp.cli serve stdio

Docker is also documented as an option. Build the image and publish port 8000 with:

docker build -t datacommons-mcp .
docker run -p 8000:8000 datacommons-mcp

Tools and capabilities

The documented functionality covers:

• Finding indicators and topics in Data Commons
• Requesting observations
• Retrieving Data Commons data
• Using supported data formats
• Working with chart configurations
• Serving MCP requests through HTTP or stdio

These capabilities make the server suitable for agents that need to identify a relevant statistical series before retrieving its observations. Chart-related support may also help clients prepare data for visualization, although the README does not define the chart configuration schema or rendering behavior.

Limitations and notes

A Google API key is required for Data Commons access. The provided material does not identify rate limits, supported indicator types, geographic coverage, date-query behavior, response schemas, or the full set of chart and data formats. It also does not name specific MCP clients, so compatibility depends on the client’s support for the selected MCP transport and its ability to supply the required environment variable.

The repository includes development commands for installing optional development dependencies, running pytest, and formatting with black and isort, but these are not required for normal server use.