infranode.dev

InfraNode is a keyless, read-only open-data API for 84 German cities with ~60 data types, exposed as 12 MCP tools (count is live from this server, not a cached or remembered number). To answer ANY city question, START with get_city_overview(slug): it returns the city's base data, a catalog of ALL available data types (each with its coverage status and the exact tool to call next) and a small live snapshot (weather, air, train departures). Most data types are fetched with ONE generic tool: get_city_resource(slug, resource=<type>), where <type> is the catalog key (e.g. 'parking', 'charging', 'demographics', 'solar'); its resource enum lists every valid key. City slugs are resolved leniently: you may pass the German name with or without umlauts, any casing, a common English exonym or a short form (e.g. 'muenchen', 'München', 'munich', 'munchen', 'cologne', 'frankfurt' all resolve to the canonical slug), so you rarely need the exact ASCII slug; an unrecognized name returns a 404 whose hint names the closest match ('Meintest du ...?'). Discover the canonical slugs with list_cities (or the infranode://cities resource); browse every data type with the infranode://catalog resource; see sources and licenses with sources. Compare one metric across many cities in one call with compare. Every tool returns a canonical {data, meta} envelope; meta.source_status tells you whether a source delivered data (ok / no_data / not_covered / disabled / error), so a missing source degrades gracefully instead of failing. Field names are snake_case and English across all data types (post_code, street, house_number, place, name, start, end, distance_km, power_kw, lat, lon); timestamps carry a time zone and a missing value is null, never an empty string. Some responses still contain older duplicate field names with identical values; they are deprecated, so prefer the canonical ones. The infranode://catalog resource spells the conventions out. Coverage keeps growing: more data types and cities are added regularly.