090°Open data
Open datasets, fully documented — searchable here, and readable by any LLM.
1 datasets
metro_weather_forecast_intel
Forward-looking weather features for 20 US/Canadian metros (15 US + 5 CA, ICAO join keys shared with metro_degree_days and metro_daylight_intel): today + 6 days x 20 metros = ~140 (forecast_date, metro) rows per daily snapshot. US rows from the NWS gridpoint 7-day forecast; CA rows from ECCC citypageweather-realtime. Per row: temp_max_c / temp_min_c, precip_prob_max (0-100), precip_type (rain/snow/mixed/none), wind_max_kmh, short summary, lead_days (0-6), flags (heat >= 35C, deep freeze <= -15C, heavy precip PoP >= 70, snow, high wind >= 60 km/h, severe = any), and demand_disruption_score (0-100 = min(100, round(30*heat_i + 30*cold_i + 25*precip_i + 15*wind_i)) with documented ramps). Units: degC, %, km/h, 0/1 flags, 0-100 score. Caveats: forecasts, not observations — a later same-day re-run can legitimately differ; wind is honest null when the provider reports none.
Once today's free allowance is used up, AI features can run on your own provider account.
Kept in this browser tab only (cleared when you close it) and sent with each AI request. Our servers use it for that request and never store or log it.