gb-climate-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_stationsA | List the weather stations, their location, elevation, date coverage and the variables available. |
| get_daily_weatherA | Daily values for one station between two dates (inclusive, YYYY-MM-DD), at most 366 days per call. Pass |
| climate_trendA | Yearly values and the linear trend per decade for one station and variable. Rain, snow, precipitation hours and evapotranspiration are yearly totals; everything else is a yearly mean. Partial years are skipped. |
| compare_stationsB | Compare all stations on one variable: the average yearly value and the trend per decade, low to high elevation. |
| rain_thresholdsA | What counts as a heavy, extreme or rare rain day at a station, from its 35-year record (percentiles of daily rain). Values are for the ~25 km ERA5 grid cell, which includes surrounding mountains, so they are much higher than valley-floor gauge readings. Use them to judge how unusual a day is, not how much fell in town. |
| flood_risk_outlookA | Live flash-flood screening outlook for the next 1-16 days at a station. Pulls the current Open-Meteo forecast and scores each day against the station's own history: daily rain vs historical percentiles, peak hourly rain (cloudburst signal), and heat-driven melt. Returns a risk level per day (normal, watch, elevated, high, severe) with the reasons. A screening signal, not an official warning. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| methodology |
TDQS
Scored across 6 tools
Most tools target clearly distinct outputs: listing stations, daily data, rain percentiles, flood outlook, and cross-station comparison are all separable. The only mild overlap is between compare_stations and climate_trend, since both surface trends-per-decade, but compare_stations is explicitly cross-station while climate_trend is single-station, so the descriptions keep them apart.
All names use consistent snake_case, which is good. However the verb pattern is mixed: some are verb_noun (compare_stations, list_stations, get_daily_weather) while others are noun phrases (rain_thresholds, climate_trend, flood_risk_outlook), a minor deviation from a single predictable pattern.
Six tools is well-scoped for a climate data server, covering discovery, retrieval, trend analysis, thresholds, and forecasting without redundancy. Each tool earns its place.
The surface covers the full lifecycle well: station discovery, daily historical retrieval, long-term trends, cross-station comparison, rain thresholds, and live flood screening. Minor gaps exist (e.g. no explicit station-metadata lookup beyond list_stations, no sub-daily history), but core workflows are covered.