Examples
Real-world prompt patterns and tool chains for the SmarterWeather MCP Server. These examples show how AI assistants compose multiple tools to answer weather questions.
Basic Weather Check
User Prompt
“What's the weather in Austin, TX this weekend?”
Tool Chain
geocode_locationget_weatherHow It Works
The AI resolves 'Austin, TX' to coordinates, then fetches the full weather overview. Daily forecasts for Saturday and Sunday are extracted from the response.
Trip Planning Comparison
User Prompt
“I'm deciding between a beach trip to Myrtle Beach or hiking in Asheville this Saturday. Which has better weather?”
Tool Chain
geocode_location (×2)compare_locationsHow It Works
Two locations are geocoded, then compare_locations provides a side-by-side comparison of temperature, precipitation probability, wind, and cloud cover for the same time window.
Best Time for Outdoor Activity
User Prompt
“When's the best 3-hour window to go running in Denver this week? I don't want rain or strong winds.”
Tool Chain
geocode_locationget_best_periodHow It Works
get_best_period searches the next 72 hours for a 3-hour window with low precipitation probability and moderate winds, filtered to daylight hours.
Snow Accumulation Analysis
User Prompt
“How much snow could we get in Boulder this storm? What's the range?”
Tool Chain
geocode_locationget_snow_forecast_rangeHow It Works
The probabilistic snow tool returns NBM ensemble percentiles (10th, 25th, 50th, 75th, 90th), exceedance thresholds, and the snow level altitude. The AI can communicate the range and confidence level.
Severe Weather Briefing
User Prompt
“Is there any severe weather risk for Oklahoma City today? Give me the full picture.”
Tool Chain
geocode_locationget_weatherget_spc_outlooksget_severe_weather_probabilitiesget_area_forecast_discussionHow It Works
A comprehensive severe weather briefing using multiple tools: current conditions and alerts, SPC convective outlook risk level, probabilistic hail/tornado/wind fields, and the local NWS forecast discussion narrative.
Aviation Weather Check
User Prompt
“What are the flight conditions at Centennial Airport for the next 12 hours?”
Tool Chain
geocode_locationget_aviation_conditionsHow It Works
Returns visibility, ceiling heights, and flight category probabilities (VFR/MVFR/IFR/LIFR) hour-by-hour for the next 12 hours at the airport location.
Precipitation Totals Over Multiple Days
User Prompt
“How much rain is expected in Houston over each of the next 5 days?”
Tool Chain
geocode_locationget_period_totalsHow It Works
get_period_totals aggregates QPF (quantitative precipitation forecast) over 5 separate 24-hour periods, returning totals for each labeled day.
Forecast Confidence Analysis
User Prompt
“How confident is the temperature forecast for Chicago next week?”
Tool Chain
geocode_locationget_forecast_confidenceHow It Works
Returns ensemble standard deviation for temperature at each hour. Higher values indicate more model disagreement, which the AI translates into a plain-language confidence assessment.
Tips for Better Results
- Be specific about location. Include city and state to avoid ambiguity — “Portland, OR” vs. “Portland, ME”.
- Specify time windows. “This weekend”, “next 6 hours”, or “Tuesday through Thursday” helps the AI choose appropriate parameters.
- Ask for confidence levels. Questions like “how certain is this forecast?” trigger probabilistic tools that provide ensemble-based confidence intervals.
- Request severe weather context. For storm situations, ask for “the full picture” to get SPC outlooks, mesoscale discussions, and probabilistic hazard data combined.