MCP server exposing 4 tools for datacommons.
This URL is a JSON-RPC 2.0 endpoint over HTTP. Issue POST requests with a JSON-RPC body. Browsers and search crawlers land here on GET.
POST https://gateway.pipeworx.io/datacommons/mcp
Content-Type: application/json
{"jsonrpc":"2.0","id":1,"method":"tools/list"}datacommons_find_place — Find the Google Data Commons place id (dcid) for a place name — country, state, county, city, region. Returns candidates with name and place type, e.g. "California" -> geoId/06, "India" -> country/IND. Use the dcid with datacommons_observation or datacommons_child_places. Narrow ambiguous names (Paris, Georgia) with place_type.datacommons_find_variable — Search Google Data Commons statistical variables by plain-language indicator, e.g. "median household income" -> Median_Income_Household, "unemployment rate" -> UnemploymentRate_Person. Returns variable ids with names and a match score. Use the id as `variable` in datacommons_observation.datacommons_observation — Get the value of one Google Data Commons statistical variable for one place: the latest value (default), the value at a date, or the whole time series. Each value names its original publisher (US Census, World Bank, Eurostat, CDC, ...) and lists how alternate sources differ. Example: variable "Count_Person", place "country/USA".datacommons_child_places — Compare one Google Data Commons statistical variable across every child place of a parent — e.g. population of all US states, median income of every county in Texas (geoId/48, County), GDP of every country in Europe. Returns rows sorted by value with names and the publisher of each value.Code samples (curl / TypeScript / one-click client install), schemas, and the live playground are on the pack page:
https://pipeworx.io/packs/datacommons/
Pipeworx is an open MCP gateway connecting AI agents to live data. pipeworx.io