Smarter Weather
Measured conclusion: Deep-tested. L2 outcome: tools_ok.
Smarter Weather MCP: forecasts, alerts, outlooks, observations, AQI, grids, and map imagery.
| Author | io.github.smarterweather |
|---|---|
| Access | mixed |
| Publisher key required | No |
| Measured state | Deep-tested |
| L1 liveness | alive |
| L1 latency | 169 ms |
| Last measured | 2026-10-24 |
Add to your agent
Copy this into your MCP client config:
{
"mcpServers": {
"Smarter Weather": {
"url": "https://mcp.smarterweather.com",
"transport": "streamable-http"
}
}
}
Measured tools (16)
search_locationsread-onlyResolve a place to ranked candidates with lat/lon: city ("Denver"), "city, ST", ZIP ("50219"), or partial input with fuzzy=true ("bost" -> Boston); venues and addresses where enabled. Rows carry `type` place|zip|address|poi; pass a row's lat/lon to weather tools. `matched_via: "autosuggest"` marks rows found after the exact match missed; `partial` means a section timed out. Weather tools take `location` directly, venues included -- use this to disambiguate ("which Springfield?") or present choices.
Parameters:
query(required) — string — Place query: city, "city, state", ZIP, partial text, or venue/address where enabled.fuzzy— boolean — Autosuggest mode for partial/misspelled input. Default false (exact search).limit— integer — Maximum candidates to return (1-10). Default 5.near_lat— number — Rank venues/addresses near this point; pair with near_lon.near_lon— number
Hints: read-only yes · destructive no · idempotent yes · open-world yes
get_forecastread-onlyComplete weather overview for a location: current conditions, daily forecast (SPC threats, severity; view "detailed" adds CAPE, UV, full day/night periods), active alerts, and outlooks in one call. Pre-aggregated across NBM, HRRR, GFS, RTMA, and SPC and unit-converted server-side. Primary weather tool; reach for lower-level tools only when you need raw observations or a specific dataset. Times are UTC ISO 8601; quote daily[].astro.local to users, not UTC. If forecast.location.timezone_resolved is false, zone unknown: state times as UTC, never as local. Accepts a place name directly. Examples: {"location": "Denver, CO"} or {"location": "Portland, OR", "days": 5} or {"lat": 41.4, "lon": -92.9}.
Parameters:
lat— number — Latitude in decimal degrees (-90 to 90). Most tools also accept a `location` place-name string instead of lat/lon.lon— number — Longitude in decimal degrees (-180 to 180). For continental US use negative values (west of the prime meridian).days— integer — Number of forecast days (1-14). Default 10.view— string — concise (default; REST's is detailed), ~5-10KB: daily has day_precip_probability / night_precip_probability (precip_probability is their max). detailed, ~12-20KB: adds CAPE, UV, full day/night periods, extra hourly fields.units— string — Units of every value in and out: imperial (default; °F, mph, in) or metric (°C, km/h, mm). si is deprecated and will be removed; use metric.include— string — Comma-separated sections: current, daily, hourly, alerts, outlooks. Default "current,daily,alerts,outlooks". Use get_hourly_forecast for hourly detail.location— string — Free-text place: "City, ST" (e.g. "Portland, OR"), US ZIP, venue/address where enabled, or "lat,lon". Use this OR lat+lon, not both.detail_level— string — Deprecated alias of view.
Hints: read-only yes · destructive no · idempotent yes · open-world yes
get_hourly_forecastread-onlyBlended hourly forecast: temperature, feels-like, humidity, wind, precipitation probability/amount, conditions, and icon per hour. hourly[0] is the current or next hour; hourly_coverage.first_time says which. Timestamps are UTC ISO 8601; convert to the local timezone before presenting. Ask for the days you need up front -- one call with days: 4 beats four calls. ALWAYS check hourly_coverage before answering about a specific hour: it reports first_time and last_time (the window the rows actually span), sample_interval_hours (past the first day rows are every 2-3h, not every hour), and truncated: true when upstream returned less than you asked for. If the hour the user cares about is after last_time, say the forecast does not reach that far yet rather than answering from the nearest row you do have. For one stretch of time ask for that stretch with hours_from/hours_to: it comes back hour by hour even where the full range would be sampled. Accepts a place name or coordinates. Examples: {"location": "Portland, OR", "days": 2} or {"lat": 41.88, "lon": -87.63, "hours_from": 36, "hours_to": 48}.
Parameters:
lat— number — Latitude in decimal degrees (-90 to 90). Most tools also accept a `location` place-name string instead of lat/lon.lon— number — Longitude in decimal degrees (-180 to 180). For continental US use negative values (west of the prime meridian).days— integer — Days of hourly data (1-7). Default 2. Widened when hours_to reaches further.view— string — concise (default) samples past 24h; detailed adds CAPE, ceiling, UV, gust and thunderstorm probability for 48h.units— string — Units of every value in and out: imperial (default; °F, mph, in) or metric (°C, km/h, mm). si is deprecated and will be removed; use metric.hours_to— integer — Window end, in hours from now, exclusive. 36 to 48 is hours 36-47.location— string — Free-text place: "City, ST" (e.g. "Portland, OR"), US ZIP, venue/address where enabled, or "lat,lon". Use this OR lat+lon, not both.hours_from— integer — Window start, in hours from now (0 = the current hour).detail_level— string — Deprecated alias of view.
Hints: read-only yes · destructive no · idempotent yes · open-world yes
get_current_conditionsread-onlyCurrent weather right now at a location from two independent sources in one call: the RTMA gridded analysis (exact-point values, updated sub-hourly) and the nearest METAR station observation (ground truth with raw METAR, flight category). Use the analysis for point-accurate values and the station for verification. Analysis fields: temperature_2m, dew_point_2m, relative_humidity_2m (derived here from temperature and dew point; listed in analysis.derived), wind_speed_10m, wind_direction_10m, wind_gusts_10m, surface_pressure (station pressure at ground level, not sea-level pressure), visibility, cloud_cover, cloud_ceiling, in units (labels in analysis.unit_labels). nearest_station is the station's own report, unconverted. For a forecast, use get_forecast. Example: {"location": "Pella, IA"}.
Parameters:
lat— number — Latitude in decimal degrees (-90 to 90). Most tools also accept a `location` place-name string instead of lat/lon.lon— number — Longitude in decimal degrees (-180 to 180). For continental US use negative values (west of the prime meridian).units— string — imperial (default; °F, mph, inHg, mi, ft) or metric (°C, km/h, hPa, km, m); si returns K, m/s, Pa, m. si is deprecated and will be removed; use metric. Not applied to nearest_station.location— string — Free-text place: "City, ST" (e.g. "Portland, OR"), US ZIP, venue/address where enabled, or "lat,lon". Use this OR lat+lon, not both.
Hints: read-only yes · destructive no · idempotent yes · open-world yes
get_alertsread-onlyNWS watches, warnings, advisories. Point (city/ZIP/lat+lon): containing polygons. BBox or US state/DC/CONUS (codes, full names, US/national): intersecting polygons. A city miss is not a statewide all-clear — query the state or a bbox; never say regional inventory is impossible. NY/WA and "New York State"/"Washington State" are states; "New York"/"Washington" stay cities. Omit at for now; at (ISO-8601 UTC) is the snapshot then. Empty = all-clear or purged (~24h). alert_id = detail+geometry, ignores at. Ex: {"location":"WI","events":["Tornado Warning"]}.
Parameters:
at— string — ISO-8601 UTC past instant for the in-effect snapshot. Ignored with alert_id.lat— number — Latitude in decimal degrees (-90 to 90). Most tools also accept a `location` place-name string instead of lat/lon.lon— number — Longitude in decimal degrees (-180 to 180). For continental US use negative values (west of the prime meridian).bbox— object — Bounding box {west,south,east,north}; west/east are negative in the US. Skips geocoding; intersecting polygons.events— array — Optional event-name filter, e.g. ["Tornado Warning"].alert_id— string — Alert identifier for detail mode. When set, location is ignored.location— string — Free-text place: "City, ST" (e.g. "Portland, OR"), US ZIP, venue/address where enabled, or "lat,lon". Use this OR lat+lon, not both.
Hints: read-only yes · destructive no · idempotent yes · open-world yes
get_observed_precipitationread-onlyHow much rain fell: observed precipitation totals at a location. Use for "did it rain", "how much rain fell", "rain in the last 24 hours / since yesterday". Source: MRMS MultiSensor QPE Pass2 (radar bias-corrected to rain gauges) over 1/3/6/24/48/72 h windows (CONUS; 1 h only in Alaska and Hawaii). Values have about 2 mm (0.08 in) resolution, so 0 means under about 1 mm; data runs about an hour behind real time. Pass `at` (within 33 h) for totals ending at a past time (newest frame up to 3 h before it). With include_nowcast (default) it also returns a next-hour radar nowcast (CONUS): peak intensity (light/moderate/heavy), type and start time, not an amount. Forecast totals past the next hour: get_period_totals (planning); station reports: get_observations (observed).
Parameters:
at— string — RFC3339 end time within the last 33 h. Default: latest.lat— number — Latitude in decimal degrees (-90 to 90). Most tools also accept a `location` place-name string instead of lat/lon.lon— number — Longitude in decimal degrees (-180 to 180). For continental US use negative values (west of the prime meridian).units— string — Units of every value in and out: imperial (default; °F, mph, in) or metric (°C, km/h, mm). si is deprecated and will be removed; use metric.windows— array — Accumulation windows ending at the latest frame (or at `at`).location— string — Free-text place: "City, ST" (e.g. "Portland, OR"), US ZIP, venue/address where enabled, or "lat,lon". Use this OR lat+lon, not both.include_nowcast— boolean — Next-hour radar nowcast (CONUS; ignored with `at`).
Hints: read-only yes · destructive no · idempotent yes · open-world yes
get_forecast_distributionread-onlyNBM probabilistic guidance for one weather aspect: percentiles (p10-p90), exceedance probabilities and ensemble spread. Use this for any question about odds, ranges, potential or confidence ("how much could we get", "range of possible highs", "how sure is this") -- a deterministic forecast value cannot answer one. Reading the percentiles: p50 is the most likely outcome, p90 is the reasonable worst case when the risk is the high end (snow totals, wind, rainfall), and p10 when the risk is the low end (cold, visibility, ceiling). A single percentile is not the forecast -- report the likely value with the tail that matters, and label which is which. Aspects: precip (PoP, QPF + percentiles), snow (accumulation percentiles, >1/2/4in probabilities, snow level, snow-type probability), ice (ice accretion, freezing-rain / ice-pellet probabilities), temperature (`daily` high/low p10/p50/p90: the NBM value -/+ 1.28 stddev; hourly temp + stddev), wind (speed/gust percentiles), severe (thunderstorm probability plus NBM CWASP, a severe-environment index, both in percent: severe_weather_prob_p50 is the median CWASP value and severe_weather_prob_above_75 the probability CWASP exceeds 75; NBM has no tornado, hail or damaging-wind odds; SPC's: get_outlooks (guidance) hazard=severe), aviation (LIFR/IFR/MVFR visibility + ceiling odds), confidence (ensemble stddev; low = settled, high = details still in play). Labels in unit_labels. Example: {"location": "Denver, CO", "aspect": "snow", "hours": 72}.
Parameters:
aspect(required) — string — Distribution family (see description).lat— number — Latitude in decimal degrees (-90 to 90). Most tools also accept a `location` place-name string instead of lat/lon.lon— number — Longitude in decimal degrees (-180 to 180). For continental US use negative values (west of the prime meridian).hours— integer — Forecast hours; default 48-72 by aspect.units— string — imperial (default; °F, mph, in) or metric. si is deprecated and will be removed; use metric.location— string — Free-text place: "City, ST" (e.g. "Portland, OR"), US ZIP, venue/address where enabled, or "lat,lon". Use this OR lat+lon, not both.
Hints: read-only yes · destructive no · idempotent yes · open-world yes
search_catalogread-onlyFind which dataset answers a weather question when no dedicated tool fits: ranks the data catalog (model grids, radar, satellite, outlooks, observations) by the question's words and returns the best datasets with variable, units, cadence, a one-line why and the exact `call` (tool and arguments) to run next. Pass lat/lon, or `domain`, to keep only datasets covering the place; without one, Alaska/Hawaii twins fold into the CONUS dataset. Calls are ready for "now"; for a past time set `time_start`/`time_end` (or `at` where the tool supports it). Free; no network call.
Parameters:
query(required) — string — The question or keywords, in the user's words.lat— number — Latitude of the place (with lon).lon— number — Longitude of the place (with lat).kind— string — Only this dataset kind.limit— integer — Hits to return (1-10, default 5).domain— string — Coverage filter when there is no point; not with lat/lon.family— string — Only this source family (e.g. "nbm", "mrms").
Hints: read-only yes · destructive no · idempotent yes · open-world no
find_toolsread-onlyLists the weather tools this connection does not show (best times, comparisons, forecast totals, air quality; station observations, soundings, hurricane recon; lightning, storm cells, storm reports; outlooks, forecaster text, hurricanes; climate records and normals; forecast accuracy; maps and charts; people at risk; crop heat units), with when to use each and its required arguments. Run one with call_tool, in this session, no reconnect. With `tools`, returns their full input schemas. Free; no network call.
Parameters:
tools— array — Tool names: return their full input schemas instead of the list.
Hints: read-only yes · destructive no · idempotent yes · open-world no
call_toolread-onlyRuns a tool find_tools lists, by name: same price, limits and result as calling it directly. Tools already in your list are called directly, not through this. `arguments` follows that tool's input schema.
Parameters:
tool(required) — string — Tool name, from find_tools.arguments— object — That tool's arguments.
Hints: read-only yes · destructive no · idempotent yes · open-world yes
describe_datasetread-onlyA dataset's variables (standard names, units, display units, categories, bands), horizon, step, cadence, freshness, cost class and an example_query for query_dataset. Example: {"dataset_id": "nbm_conus"}.
Parameters:
dataset_id(required) — string — Pipeline dataset identifier (e.g. "nbm_conus", "mrms_reflectivity_conus"). Discover valid values with list_datasets; inspect variables with describe_dataset.
Hints: read-only yes · destructive no · idempotent yes · open-world yes
get_platform_statusread-onlyCurrent data-freshness status of the weather platform: overall state, per-source states (ok / degraded / outage / no_signal), open incidents with cause attribution (provider outage vs internal processing delay), and active provider advisories. Use this when a user asks whether data is current, when other tools return surprisingly stale data, or before presenting time-critical weather. If a source is degraded or in outage, tell the user their data may be stale rather than presenting it as live. No inputs. Refreshed about every 5 minutes.
Parameters:
include_ok— boolean — true: list every monitored source including healthy ones. false (default): only sources that are not ok, keeping the response compact.
Hints: read-only yes · destructive no · idempotent yes · open-world yes
get_time_contextread-onlyComplete temporal context for a location: local time, timezone, 14-day calendar with day names and Today/Tomorrow offsets, sunrise/sunset/solar times (from the weather pipeline's astro product), and moon phase. Use whenever you need to reason about dates, times, or daylight for a location -- including "what time is sunset?", "is it dark there now?", or "what day of the week is the 4th-day forecast?". If current_time.timezone_resolved is false, zone unknown: state times as UTC, never as local. Accepts a place name directly. Example: {"location": "Seattle, WA"}.
Parameters:
lat— number — Latitude in decimal degrees (-90 to 90). Most tools also accept a `location` place-name string instead of lat/lon.lon— number — Longitude in decimal degrees (-180 to 180). For continental US use negative values (west of the prime meridian).location— string — Free-text place: "City, ST" (e.g. "Portland, OR"), US ZIP, venue/address where enabled, or "lat,lon". Use this OR lat+lon, not both.
Hints: read-only yes · destructive no · idempotent yes · open-world yes
list_datasetsread-onlyDiscover the gridded datasets (model grids, analyses) available at a location, or every public one when none is given, keyed by dataset_id. At a point: its domain (CONUS/Alaska/Hawaii) plus global grids, with freshness where tracked (latest model run as latest_run, latest_time, data_age_seconds, freshness_status). Use this to find dataset_id values for query_dataset or check whether data is current. Variables, units and standard names come from describe_dataset. Example: {"location": "Anchorage, AK"}.
Parameters:
lat— number — Latitude in decimal degrees (-90 to 90). Most tools also accept a `location` place-name string instead of lat/lon.lon— number — Longitude in decimal degrees (-180 to 180). For continental US use negative values (west of the prime meridian).location— string — Free-text place: "City, ST" (e.g. "Portland, OR"), US ZIP, venue/address where enabled, or "lat,lon". Use this OR lat+lon, not both.include_freshness— boolean — Include latest model run and data age where tracked. Default true.
Hints: read-only yes · destructive no · idempotent yes · open-world yes
query_datasetread-onlyRaw time series of one dataset's variables at a point, with unit labels: any gridded product (NBM, HRRR, RRFS, GFS, RTMA, MRMS, air quality, ...). Time modes: hours (default 24), time_start+time_end (ISO 8601), or latest=true (newest value). Up to 168 frames: when truncated, send next_cursor as cursor with the same arguments. reference_time pins a model run; each series reports its run (reference_time, or reference_times when runs mix); runs older than about 48 h may be gone. For blended forecasts use get_forecast. Examples: {"location": "Denver, CO", "dataset_id": "rrfs_surface", "variables": ["temperature_2m"], "hours": 18} or {"lat": 41.4, "lon": -92.9, "dataset_id": "rtma_conus", "variables": ["temperature_2m"], "latest": true}.
Parameters:
variables(required) — array — Standard names (e.g. temperature_2m, precipitation); see describe_dataset.lat— number — Latitude in decimal degrees (-90 to 90). Most tools also accept a `location` place-name string instead of lat/lon.lon— number — Longitude in decimal degrees (-180 to 180). For continental US use negative values (west of the prime meridian).hours— integer — Hours from now (1-264): ahead, or back on analyses (default: newest 166 frames); from the run start with reference_time.units— stringcursor— string — next_cursor of the previous page.latest— boolean — Only the newest value (analyses like RTMA/MRMS).location— string — Free-text place: "City, ST" (e.g. "Portland, OR"), US ZIP, venue/address where enabled, or "lat,lon". Use this OR lat+lon, not both.time_end— string — ISO 8601 window end (with time_start).dataset_id— string — Default: the NBM dataset for the location (nbm_conus/nbm_alaska/nbm_hawaii). Discover with list_datasets.time_start— string — ISO 8601 window start (with time_end).reference_time— string — Pin a specific model run (ISO 8601). Default: latest run.
Hints: read-only yes · destructive no · idempotent yes · open-world yes
reverse_geocoderead-onlyResolve coordinates to a human-readable place (city, state, county, timezone). Use when you have lat/lon but need a display name or the local timezone. Example: {"lat": 39.74, "lon": -104.99} -> Denver, Colorado, America/Denver.
Parameters:
lat(required) — number — Latitude in decimal degrees (-90 to 90). Most tools also accept a `location` place-name string instead of lat/lon.lon(required) — number — Longitude in decimal degrees (-180 to 180). For continental US use negative values (west of the prime meridian).
Hints: read-only yes · destructive no · idempotent yes · open-world yes
Measured by MCPYP probing. Untested fields are omitted, never estimated. Listing is always free; MCPYP never proxies calls or stores user tokens.