In September, 2025, we launched the Knowledge Commons Mannequin Context Protocol (MCP) server to offer a normal approach for AI brokers to work together with Knowledge Commons information natively. This was adopted by the Gemini CLI extension, which streamlined the setup course of for Gemini CLI customers.
Whereas these instruments efficiently opened the door for pure language information exploration, they relied on native Python environments, which created some challenges. Putting in and operating open-source instruments was not appropriate with high-security environments; and internet hosting an area MCP server lacked the scalability wanted for builders who wish to publish their very own question brokers.
At this time, we’re taking the following step to make it simpler for everybody to make use of datacommons.org with an LLM by launching the hosted Knowledge Commons MCP service in Google Cloud Platform. Now you’ll be able to simply connect with the free service and let Google handle the remaining – no want to fret about Python environments, useful resource administration, model releases, safety compliance and many others.
Why use it?
The MCP server offers a standardized approach for AI brokers to devour Knowledge Commons information natively. Analysts can create insights by asking high-level questions in pure language with the MCP returning information from trusted sources, whereas builders can simply create AI brokers personalized for his or her wants. For instance, you will get statistical solutions to questions like “What’s the correlation between unemployment ranges and weight problems charges in U.S. states?” or “Rank-order the GDP of each japanese European nation”.
Join right now!
In case you’re already utilizing the Knowledge Commons Gemini CLI extension, you needn’t do something. The following time you run Gemini CLI, the extension will robotically replace itself to connect with the server over the online as a substitute of beginning up an area server occasion. If you have not used the extension earlier than, see Use the Gemini CLI extension to get began.
In case you’re utilizing Gemini CLI with out the extension, or every other agent, you may must get a free Knowledge Commons API key and replace your configuration to level to the hosted server. For full particulars, see Use Gemini CLI.
"mcpServers": {
"datacommons-mcp": {
"httpUrl": "https://api.datacommons.org/mcp",
"headers": {
"X-API-Key": "YOUR DC API KEY"
}
}
}JSON
Utilizing Knowledge Commons MCP instruments with Google Antigravity
And as at all times, in case you’re considering sharing your cool utilization of Knowledge Commons or have suggestions, please e-mail us at assist@datacommons.org.
Observe: The hosted MCP server can solely be used to question datacommons.org. In case you’re operating your individual Customized Knowledge Commons occasion, you may nonetheless must run your individual MCP server. See Run MCP instruments for extra data.
