Skip to main content
Glama

Get METAR Weather Observations

aviation_get_metar
Read-onlyIdempotent

Get current or historical METAR observations for 1–10 named stations, a bounding box, or a US state or DC. Named lookups keep all observations that AWC drew within the window. Surveys keep the latest observation per station in the draw, with optional flight_category and limit. Results include decoded weather, wind, visibility, ceiling and flight category plus the raw METAR. Use aviation_find_stations to resolve or verify station identifiers.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "aviation_get_metar"
  }
}

Input Schema

NameRequiredDescriptionDefault
bboxNoGeographic bounding box for an area observation survey.
as_ofNoHistorical UTC reference as YYYY-MM-DDTHH:mm:ssZ or YYYY-MM-DDTHH:mm:ss.sssZ. AWC retains up to the previous 30 days. Omit to use the current time.
hoursNoLookback window in hours (1–720), ending at as_of or the current time when omitted, and not a row limit. With station_ids every observation drawn inside the window is returned, so a station reporting more often than hourly yields more than one row per hour: at the default of 1, half-hourly stations return two observations and SPECI-issuing stations can return more, so budget rows by the station reporting interval, never one per station. With bbox the result is reduced to the latest observation per station, but the window still shapes the draw the upstream row cap applies to — a wide window spends that cap on repeat readings from busy stations instead of on more stations, so lower it before narrowing the box.
limitNoMaximum stations to return from offset after selection. The served maximum is 400 when limit is supplied. Omit to return all remaining matches.
stateNoTwo-letter US state or DC code, case-insensitive. Returns observations available through the AWC state selector.
offsetNoZero-based start among matching survey stations ordered by station ID.
station_idsNoICAO station IDs to query. 1–10 stations per call. Returns all drawn observations inside hours, subject to the upstream cap. Use bbox or state to search an area.
flight_categoryNoIn a bbox or state survey, filter after selecting the latest observation per station and before applying limit. Stations with unknown category do not match.

Output Schema

NameRequiredDescriptionDefault
capNoThe upstream row maximum that was applied to the draw. Present only on a truncated result.
errorNoPresent when the call failed. Absent on success.
shownNoDistinct stations shown after any survey reduction, filter and limit. Also emitted on a capped named lookup.
noticeNoGuidance for a capped draw, narrowed survey, or named lookup with missing stations. Surveys returning more than 40 stations with no limit also get size guidance.
limitedNoWhether the caller limit withheld survey matches from the draw. Present when the survey supplied a limit.
matchedNoStations in the draw after any flight_category filter and before paging. Present when offset was supplied or limit withheld matches. Upstream truncation can exclude observations from this count.
missingNoRequested station IDs absent from the result. Absent when none are missing.
partialNoTrue when a requested station produced no observation. False affirms the result covers every requested station, so full coverage is distinguishable from a short batch rather than being inferred from the count. Present only on a station_ids lookup.
returnedNoDistinct station IDs that produced at least one observation. Counted per station, not per row — with hours > 1 a station reporting six times still appears once. Present only on a station_ids lookup.
surveyedNoDistinct stations reached before the survey flight_category filter. A capped result counts only the upstream draw.
requestedNoStation IDs this call asked for, in the order given. Present only on a station_ids lookup — a bbox survey asks for an area, not a list.
truncatedNoTrue when AWC reached the row cap. False is not a completeness claim: AWC can also thin dense bbox draws by station priority. Named lookups emit this flag only when capped.
nextOffsetNoOffset for the next page of the currently available draw. Present only while matches remain.
observationsNoDrawn observations. Named stations can have multiple entries. Surveys by bbox or state return the latest observation for each matching station.
upstreamRowsNoObservations AWC returned before the reduction to one row per station. Present on a capped named lookup or when a capped survey was reduced to fewer distinct stations.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed20 schema fields changed
    • addedInput schema / properties / as_of
      Added value: +{
      +  "description": "Historical UTC reference as YYYY-MM-DDTHH:mm:ssZ or YYYY-MM-DDTHH:mm:ss.sssZ. AWC retains up to the previous 30 days. Omit to use the current time.",
      +  "maxLength": 24,
      +  "type": "string"
      +}
    • changedInput schema / properties / flight_category / description
      Previous value: -"Return only the stations whose flight category is one of these (e.g., [\"IFR\", \"LIFR\"] to find where conditions are poor). Applied in the tool after the draw is reduced to the latest observation per station and before limit, so it changes which stations come back, not what AWC draws or the upstream row cap. A station AWC could not rate (flight_category \"unknown\") never matches. Belongs to the bbox mode; supplying it alongside station_ids is rejected, since each named station already reports its own category. Optional."New value: +"In a bbox or state survey, filter after selecting the latest observation per station and before applying limit. Stations with unknown category do not match."
    • changedInput schema / properties / hours / description
      Previous value: -"Lookback window in hours (1–12), not a row limit. With station_ids every observation inside the window is returned, so a station reporting more often than hourly yields more than one row per hour: at the default of 1, half-hourly stations return two observations and SPECI-issuing stations can return more, so budget rows by the station reporting interval, never one per station. With bbox the result is reduced to the latest observation per station, but the window still shapes the draw the upstream row cap applies to — a wide window spends that cap on repeat readings from busy stations instead of on more stations, so lower it before narrowing the box."New value: +"Lookback window in hours (1–720), ending at as_of or the current time when omitted, and not a row limit. With station_ids every observation drawn inside the window is returned, so a station reporting more often than hourly yields more than one row per hour: at the default of 1, half-hourly stations return two observations and SPECI-issuing stations can return more, so budget rows by the station reporting interval, never one per station. With bbox the result is reduced to the latest observation per station, but the window still shapes the draw the upstream row cap applies to — a wide window spends that cap on repeat readings from busy stations instead of on more stations, so lower it before narrowing the box."
    • changedInput schema / properties / hours / maximum
      Previous value: -12New value: +720
    • changedInput schema / properties / limit / description
      Previous value: -"Maximum stations to return, applied after the draw is reduced to the latest observation per station, after any flight_category filter, and ordered by station ID ascending — so the same query with the same limit returns the same stations. It bounds the response without changing the area searched, which a smaller bbox would, and it keeps an alphabetical slice rather than a region or a condition. Distinct from the 400-row upstream cap: a limited result examined every station it counted and withheld some, while a capped one never drew the rest. Belongs to the bbox mode; supplying it alongside station_ids is rejected, since that mode already names the set. Omit to return every station in the box. Optional."New value: +"Maximum stations to return from offset after selection. The served maximum is 400 when limit is supplied. Omit to return all remaining matches."
    • removedInput schema / properties / limit / maximum
      Removed value: -400
    • addedInput schema / properties / offset
      Added value: +{
      +  "description": "Zero-based start among matching survey stations ordered by station ID.",
      +  "maximum": 9007199254740991,
      +  "minimum": 0,
      +  "type": "integer"
      +}
    • addedInput schema / properties / state
      Added value: +{
      +  "description": "Two-letter US state or DC code, case-insensitive. Returns observations available through the AWC state selector.",
      +  "pattern": "^[A-Za-z]{2}$",
      +  "type": "string"
      +}
    • changedInput schema / properties / station_ids / description
      Previous value: -"ICAO station IDs to query. 1–10 stations per call, and every observation inside the hours window is returned for each. Supplying it alongside bbox is rejected — use bbox to ask about an area rather than a named set."New value: +"ICAO station IDs to query. 1–10 stations per call. Returns all drawn observations inside hours, subject to the upstream cap. Use bbox or state to search an area."
    • changedOutput schema / properties / error / properties / data / properties / reason / description
      Previous value: -"Machine-readable failure mode. Declared by this tool: `no_stations_found`: No station returned METAR data — none of the requested IDs, or no station inside the bounding box. `missing_location`: Neither station_ids nor bbox was provided. `conflicting_location`: Both station_ids and bbox were provided. `invalid_bbox`: The bounding box is inverted — minLat > maxLat or minLon > maxLon. `conflicting_limit`: limit was provided together with station_ids, where the caller has already named the set. `conflicting_flight_category`: flight_category was provided together with station_ids, where the caller has already named the set. Other values are possible when a failure originates below the handler."New value: +"Machine-readable failure mode. Declared by this tool: `invalid_state`: The state code is outside the 50 US states and DC. `future_as_of`: The historical reference time is in the future. `outside_retention`: The historical reference time predates the AWC retention window. `no_stations_found`: No station returned METAR data — none of the requested IDs, or no station inside the bounding box. `missing_location`: No location selector was provided. `conflicting_location`: More than one location selector was provided. `invalid_bbox`: The bounding box is inverted — minLat > maxLat or minLon > maxLon. `conflicting_offset`: offset was supplied for a named station lookup. `conflicting_limit`: limit was provided together with station_ids, where the caller has already named the set. `conflicting_flight_category`: flight_category was provided together with station_ids, where the caller has already named the set. Other values are possible when a failure originates below the handler."
    • changedOutput schema / properties / error / properties / data / properties / reason / examples
      Previous value: -[
      -  "no_stations_found",
      -  "missing_location",
      -  "conflicting_location",
      -  "invalid_bbox",
      -  "conflicting_limit",
      -  "conflicting_flight_category"
      -]New value: +[
      +  "invalid_state",
      +  "future_as_of",
      +  "outside_retention",
      +  "no_stations_found",
      +  "missing_location",
      +  "conflicting_location",
      +  "invalid_bbox",
      +  "conflicting_offset",
      +  "conflicting_limit",
      +  "conflicting_flight_category"
      +]
    • changedOutput schema / properties / limited / description
      Previous value: -"True when the requested limit withheld stations that matched — the caller asked to see fewer of them. False affirms the limit did not bite, so every station drawn is here. Present only when a bbox survey supplied a limit. It never states anything about the upstream cap: a limited result examined every station it counted, while a truncated one never drew the rest."New value: +"Whether the caller limit withheld survey matches from the draw. Present when the survey supplied a limit."
    • changedOutput schema / properties / matched / description
      Previous value: -"Stations the draw held before the limit selected from them — after the flight_category filter, where one was supplied. Present only on a limited result. Where the result is also truncated this counts the capped draw and not the box — the observations the cap dropped were never examined, so no count can include them."New value: +"Stations in the draw after any flight_category filter and before paging. Present when offset was supplied or limit withheld matches. Upstream truncation can exclude observations from this count."
    • addedOutput schema / properties / nextOffset
      Added value: +{
      +  "description": "Offset for the next page of the currently available draw. Present only while matches remain.",
      +  "maximum": 9007199254740991,
      +  "minimum": -9007199254740991,
      +  "type": "integer"
      +}
    • changedOutput schema / properties / notice / description
      Previous value: -"Guidance for whichever disclosures fired, in this order: the levers that narrow a truncated draw; what the flight_category filter kept of the stations surveyed, or that it emptied the result; what a requested limit withheld; and — on a survey of more than 40 stations from a call that set no limit — that limit bounds the response but keeps an alphabetical slice, not a region. Or, on a station_ids lookup, the station IDs that returned nothing. The survey statements share one notice and the text keeps them apart; the reconciliation cannot join them, since only bbox reaches the row cap or takes a filter or a limit, and only station_ids reconciles a request."New value: +"Guidance for a capped draw, narrowed survey, or named lookup with missing stations. Surveys returning more than 40 stations with no limit also get size guidance."
    • changedOutput schema / properties / observations / description
      Previous value: -"Weather observations. With station_ids, one per station/time pair — multiple entries per station when the hours window holds more than one observation. With bbox, one per station: the latest observation inside that window."New value: +"Drawn observations. Named stations can have multiple entries. Surveys by bbox or state return the latest observation for each matching station."
    • changedOutput schema / properties / shown / description
      Previous value: -"Stations in this result, counted after the reduction to one observation per station, after any flight_category filter, and after any limit. Present only on a bbox survey."New value: +"Distinct stations shown after any survey reduction, filter and limit. Also emitted on a capped named lookup."
    • changedOutput schema / properties / surveyed / description
      Previous value: -"Stations the survey reached — one per station, the latest observation of each — before the flight_category filter selected from them. Present only when a bbox survey supplied flight_category; the count after the filter is shown, or matched where a limit then withheld stations. Where the result is also truncated this counts the capped draw and not the box."New value: +"Distinct stations reached before the survey flight_category filter. A capped result counts only the upstream draw."
    • changedOutput schema / properties / truncated / description
      Previous value: -"True when the upstream draw hit the AWC row cap, so stations inside the box are missing from this result. False says the draw did not reach that cap, which a count alone cannot establish — the draw is reduced to one row per station, so the returned count sits far below the cap either way. It is not a completeness claim: AWC also thins a dense box by station priority, independently of the cap and with nothing in the response to mark it, so re-run over smaller quadrants and union the results when the box must be complete. Present only on a bbox survey."New value: +"True when AWC reached the row cap. False is not a completeness claim: AWC can also thin dense bbox draws by station priority. Named lookups emit this flag only when capped."
    • changedOutput schema / properties / upstreamRows / description
      Previous value: -"Observations AWC returned before the reduction to one row per station. Present only on a truncated draw the reduction then narrowed, where the station count sits below the cap and so cannot reveal the truncation on its own."New value: +"Observations AWC returned before the reduction to one row per station. Present on a capped named lookup or when a capped survey was reduced to fewer distinct stations."
  2. Changed2 schema fields changed
    • changedInput schema / properties / station_ids / items / description
      Previous value: -"ICAO station ID: 4 uppercase letters or digits (e.g., KSEA, KJFK, K0S9)."New value: +"ICAO station ID: 4 ASCII letters or digits, case-insensitive and normalized to uppercase (e.g., KSEA, KJFK, K0S9)."
    • changedInput schema / properties / station_ids / items / pattern
      Previous value: -"^[A-Z0-9]{4}$"New value: +"^[A-Za-z0-9]{4}$"
  3. Changed8 schema fields changed
    • addedInput schema / properties / flight_category
      Added value: +{
      +  "description": "Return only the stations whose flight category is one of these (e.g., [\"IFR\", \"LIFR\"] to find where conditions are poor). Applied in the tool after the draw is reduced to the latest observation per station and before limit, so it changes which stations come back, not what AWC draws or the upstream row cap. A station AWC could not rate (flight_category \"unknown\") never matches. Belongs to the bbox mode; supplying it alongside station_ids is rejected, since each named station already reports its own category. Optional.",
      +  "items": {
      +    "description": "A flight category: VFR, MVFR, IFR, or LIFR.",
      +    "enum": [
      +      "VFR",
      +      "MVFR",
      +      "IFR",
      +      "LIFR"
      +    ],
      +    "type": "string"
      +  },
      +  "minItems": 1,
      +  "type": "array"
      +}
    • changedInput schema / properties / limit / description
      Previous value: -"Maximum stations to return, applied after the draw is reduced to the latest observation per station and ordered by station ID ascending — so the same query with the same limit returns the same stations. It bounds the response without changing the area searched, which a smaller bbox would. Distinct from the 400-row upstream cap: a limited result examined every station it counted and withheld some, while a capped one never drew the rest. Belongs to the bbox mode; supplying it alongside station_ids is rejected, since that mode already names the set. Omit to return every station in the box. Optional."New value: +"Maximum stations to return, applied after the draw is reduced to the latest observation per station, after any flight_category filter, and ordered by station ID ascending — so the same query with the same limit returns the same stations. It bounds the response without changing the area searched, which a smaller bbox would, and it keeps an alphabetical slice rather than a region or a condition. Distinct from the 400-row upstream cap: a limited result examined every station it counted and withheld some, while a capped one never drew the rest. Belongs to the bbox mode; supplying it alongside station_ids is rejected, since that mode already names the set. Omit to return every station in the box. Optional."
    • changedOutput schema / properties / error / properties / data / properties / reason / description
      Previous value: -"Machine-readable failure mode. Declared by this tool: `no_stations_found`: No station returned METAR data — none of the requested IDs, or no station inside the bounding box. `missing_location`: Neither station_ids nor bbox was provided. `conflicting_location`: Both station_ids and bbox were provided. `invalid_bbox`: The bounding box is inverted — minLat > maxLat or minLon > maxLon. `conflicting_limit`: limit was provided together with station_ids, where the caller has already named the set. Other values are possible when a failure originates below the handler."New value: +"Machine-readable failure mode. Declared by this tool: `no_stations_found`: No station returned METAR data — none of the requested IDs, or no station inside the bounding box. `missing_location`: Neither station_ids nor bbox was provided. `conflicting_location`: Both station_ids and bbox were provided. `invalid_bbox`: The bounding box is inverted — minLat > maxLat or minLon > maxLon. `conflicting_limit`: limit was provided together with station_ids, where the caller has already named the set. `conflicting_flight_category`: flight_category was provided together with station_ids, where the caller has already named the set. Other values are possible when a failure originates below the handler."
    • changedOutput schema / properties / error / properties / data / properties / reason / examples
      Previous value: -[
      -  "no_stations_found",
      -  "missing_location",
      -  "conflicting_location",
      -  "invalid_bbox",
      -  "conflicting_limit"
      -]New value: +[
      +  "no_stations_found",
      +  "missing_location",
      +  "conflicting_location",
      +  "invalid_bbox",
      +  "conflicting_limit",
      +  "conflicting_flight_category"
      +]
    • changedOutput schema / properties / matched / description
      Previous value: -"Stations the draw held before the limit selected from them. Present only on a limited result. Where the result is also truncated this counts the capped draw and not the box — the observations the cap dropped were never examined, so no count can include them."New value: +"Stations the draw held before the limit selected from them — after the flight_category filter, where one was supplied. Present only on a limited result. Where the result is also truncated this counts the capped draw and not the box — the observations the cap dropped were never examined, so no count can include them."
    • changedOutput schema / properties / notice / description
      Previous value: -"Guidance for whichever disclosures fired: the levers that narrow a truncated draw, what a requested limit withheld, or the station IDs that returned nothing. The cap and the limit can both fire on one result, and the text keeps them apart; the reconciliation cannot join them, since only bbox reaches the row cap or accepts a limit, and only station_ids reconciles a request."New value: +"Guidance for whichever disclosures fired, in this order: the levers that narrow a truncated draw; what the flight_category filter kept of the stations surveyed, or that it emptied the result; what a requested limit withheld; and — on a survey of more than 40 stations from a call that set no limit — that limit bounds the response but keeps an alphabetical slice, not a region. Or, on a station_ids lookup, the station IDs that returned nothing. The survey statements share one notice and the text keeps them apart; the reconciliation cannot join them, since only bbox reaches the row cap or takes a filter or a limit, and only station_ids reconciles a request."
    • changedOutput schema / properties / shown / description
      Previous value: -"Stations in this result, counted after the reduction to one observation per station and after any limit. Present only on a bbox survey."New value: +"Stations in this result, counted after the reduction to one observation per station, after any flight_category filter, and after any limit. Present only on a bbox survey."
    • addedOutput schema / properties / surveyed
      Added value: +{
      +  "description": "Stations the survey reached — one per station, the latest observation of each — before the flight_category filter selected from them. Present only when a bbox survey supplied flight_category; the count after the filter is shown, or matched where a limit then withheld stations. Where the result is also truncated this counts the capped draw and not the box.",
      +  "type": "number"
      +}
  4. Changed24 schema fields changed
    • addedInput schema / properties / bbox
      Added value: +{
      +  "description": "Geographic bounding box for an area observation survey.",
      +  "properties": {
      +    "maxLat": {
      +      "description": "Northern boundary latitude in decimal degrees.",
      +      "maximum": 90,
      +      "minimum": -90,
      +      "type": "number"
      +    },
      +    "maxLon": {
      +      "description": "Eastern boundary longitude in decimal degrees.",
      +      "maximum": 180,
      +      "minimum": -180,
      +      "type": "number"
      +    },
      +    "minLat": {
      +      "description": "Southern boundary latitude in decimal degrees.",
      +      "maximum": 90,
      +      "minimum": -90,
      +      "type": "number"
      +    },
      +    "minLon": {
      +      "description": "Western boundary longitude in decimal degrees.",
      +      "maximum": 180,
      +      "minimum": -180,
      +      "type": "number"
      +    }
      +  },
      +  "required": [
      +    "minLat",
      +    "minLon",
      +    "maxLat",
      +    "maxLon"
      +  ],
      +  "type": "object"
      +}
    • changedInput schema / properties / hours / description
      Previous value: -"Lookback window in hours (1–12), not a row limit: every observation inside the window is returned, so a station reporting more often than hourly yields more than one row per hour. At the default of 1, half-hourly stations return two observations and SPECI-issuing stations can return more. Budget rows by the station reporting interval, never one per station."New value: +"Lookback window in hours (1–12), not a row limit. With station_ids every observation inside the window is returned, so a station reporting more often than hourly yields more than one row per hour: at the default of 1, half-hourly stations return two observations and SPECI-issuing stations can return more, so budget rows by the station reporting interval, never one per station. With bbox the result is reduced to the latest observation per station, but the window still shapes the draw the upstream row cap applies to — a wide window spends that cap on repeat readings from busy stations instead of on more stations, so lower it before narrowing the box."
    • addedInput schema / properties / limit
      Added value: +{
      +  "description": "Maximum stations to return, applied after the draw is reduced to the latest observation per station and ordered by station ID ascending — so the same query with the same limit returns the same stations. It bounds the response without changing the area searched, which a smaller bbox would. Distinct from the 400-row upstream cap: a limited result examined every station it counted and withheld some, while a capped one never drew the rest. Belongs to the bbox mode; supplying it alongside station_ids is rejected, since that mode already names the set. Omit to return every station in the box. Optional.",
      +  "maximum": 400,
      +  "minimum": 1,
      +  "type": "integer"
      +}
    • changedInput schema / properties / station_ids / description
      Previous value: -"ICAO station IDs to query. 1–10 stations per call."New value: +"ICAO station IDs to query. 1–10 stations per call, and every observation inside the hours window is returned for each. Supplying it alongside bbox is rejected — use bbox to ask about an area rather than a named set."
    • changedInput schema / properties / station_ids / items / description
      Previous value: -"4-letter ICAO station ID (e.g., KSEA, KJFK)."New value: +"ICAO station ID: 4 uppercase letters or digits (e.g., KSEA, KJFK, K0S9)."
    • changedInput schema / properties / station_ids / items / pattern
      Previous value: -"^[A-Z]{4}$"New value: +"^[A-Z0-9]{4}$"
    • removedInput schema / required
      Removed value: -[
      -  "station_ids"
      -]
    • changedOutput schema / anyOf
      Previous value: -[
      -  {
      -    "not": {
      -      "required": [
      -        "error"
      -      ]
      -    },
      -    "required": [
      -      "observations",
      -      "requested",
      -      "returned",
      -      "partial"
      -    ]
      -  },
      -  {
      -    "required": [
      -      "error"
      -    ]
      -  }
      -]New value: +[
      +  {
      +    "not": {
      +      "required": [
      +        "error"
      +      ]
      +    },
      +    "required": [
      +      "observations"
      +    ]
      +  },
      +  {
      +    "required": [
      +      "error"
      +    ]
      +  }
      +]
    • addedOutput schema / properties / cap
      Added value: +{
      +  "description": "The upstream row maximum that was applied to the draw. Present only on a truncated result.",
      +  "type": "number"
      +}
    • changedOutput schema / properties / error / properties / data / properties / reason / description
      Previous value: -"Machine-readable failure mode. Declared by this tool: `no_stations_found`: None of the requested station IDs returned METAR data. Other values are possible when a failure originates below the handler."New value: +"Machine-readable failure mode. Declared by this tool: `no_stations_found`: No station returned METAR data — none of the requested IDs, or no station inside the bounding box. `missing_location`: Neither station_ids nor bbox was provided. `conflicting_location`: Both station_ids and bbox were provided. `invalid_bbox`: The bounding box is inverted — minLat > maxLat or minLon > maxLon. `conflicting_limit`: limit was provided together with station_ids, where the caller has already named the set. Other values are possible when a failure originates below the handler."
    • changedOutput schema / properties / error / properties / data / properties / reason / examples
      Previous value: -[
      -  "no_stations_found"
      -]New value: +[
      +  "no_stations_found",
      +  "missing_location",
      +  "conflicting_location",
      +  "invalid_bbox",
      +  "conflicting_limit"
      +]
    • addedOutput schema / properties / limited
      Added value: +{
      +  "description": "True when the requested limit withheld stations that matched — the caller asked to see fewer of them. False affirms the limit did not bite, so every station drawn is here. Present only when a bbox survey supplied a limit. It never states anything about the upstream cap: a limited result examined every station it counted, while a truncated one never drew the rest.",
      +  "type": "boolean"
      +}
    • addedOutput schema / properties / matched
      Added value: +{
      +  "description": "Stations the draw held before the limit selected from them. Present only on a limited result. Where the result is also truncated this counts the capped draw and not the box — the observations the cap dropped were never examined, so no count can include them.",
      +  "type": "number"
      +}
    • changedOutput schema / properties / notice / description
      Previous value: -"Recovery guidance naming the missing station IDs. Present only on a partial result. It lists the candidate causes without asserting one — upstream omits the row either way."New value: +"Guidance for whichever disclosures fired: the levers that narrow a truncated draw, what a requested limit withheld, or the station IDs that returned nothing. The cap and the limit can both fire on one result, and the text keeps them apart; the reconciliation cannot join them, since only bbox reaches the row cap or accepts a limit, and only station_ids reconciles a request."
    • changedOutput schema / properties / observations / description
      Previous value: -"Weather observations, one per station/time pair. Multiple entries per station when hours > 1."New value: +"Weather observations. With station_ids, one per station/time pair — multiple entries per station when the hours window holds more than one observation. With bbox, one per station: the latest observation inside that window."
    • changedOutput schema / properties / observations / items / properties / clouds / description
      Previous value: -"All reported cloud layers from lowest to highest. Empty whenever the observation published no layer heights, which covers a clear sky, an obscuration with no determinable height, and an observation that reported no sky condition at all; sky_condition distinguishes them. An empty array is not a clear sky on its own."New value: +"Cloud layers from lowest to highest, as AWC decoded them — at most four, so a fifth or higher reported layer is left out with no marker and raw_metar is the complete source. Empty whenever the observation published no layer heights, which covers a clear sky, an obscuration with no determinable height, and an observation that reported no sky condition at all; sky_condition distinguishes them. An empty array is not a clear sky on its own."
    • changedOutput schema / properties / observations / items / properties / station_id / description
      Previous value: -"ICAO 4-letter station identifier (e.g., KSEA)."New value: +"ICAO station identifier, 4 uppercase letters or digits (e.g., KSEA, K0S9)."
    • changedOutput schema / properties / observations / items / properties / wind / properties / direction_deg / description
      Previous value: -"Wind direction in degrees true. Null when variable."New value: +"Wind direction in degrees true. Null when the observation reported a variable wind (VRB), and also when it carried no wind group at all — speed_kt is a number in the first case and null in the second."
    • changedOutput schema / properties / partial / description
      Previous value: -"True when a requested station produced no observation. False affirms the result covers every requested station, so full coverage is distinguishable from a short batch rather than being inferred from the count."New value: +"True when a requested station produced no observation. False affirms the result covers every requested station, so full coverage is distinguishable from a short batch rather than being inferred from the count. Present only on a station_ids lookup."
    • changedOutput schema / properties / requested / description
      Previous value: -"Station IDs this call asked for, in the order given."New value: +"Station IDs this call asked for, in the order given. Present only on a station_ids lookup — a bbox survey asks for an area, not a list."
    • changedOutput schema / properties / returned / description
      Previous value: -"Distinct station IDs that produced at least one observation. Counted per station, not per row — with hours > 1 a station reporting six times still appears once."New value: +"Distinct station IDs that produced at least one observation. Counted per station, not per row — with hours > 1 a station reporting six times still appears once. Present only on a station_ids lookup."
    • addedOutput schema / properties / shown
      Added value: +{
      +  "description": "Stations in this result, counted after the reduction to one observation per station and after any limit. Present only on a bbox survey.",
      +  "type": "number"
      +}
    • addedOutput schema / properties / truncated
      Added value: +{
      +  "description": "True when the upstream draw hit the AWC row cap, so stations inside the box are missing from this result. False says the draw did not reach that cap, which a count alone cannot establish — the draw is reduced to one row per station, so the returned count sits far below the cap either way. It is not a completeness claim: AWC also thins a dense box by station priority, independently of the cap and with nothing in the response to mark it, so re-run over smaller quadrants and union the results when the box must be complete. Present only on a bbox survey.",
      +  "type": "boolean"
      +}
    • addedOutput schema / properties / upstreamRows
      Added value: +{
      +  "description": "Observations AWC returned before the reduction to one row per station. Present only on a truncated draw the reduction then narrowed, where the station count sits below the cap and so cannot reveal the truncation on its own.",
      +  "type": "number"
      +}
  5. Changed6 schema fields changed
    • changedInput schema / properties / hours / description
      Previous value: -"Hours of observation history to return (1–12). Default 1 returns only the most recent observation per station."New value: +"Lookback window in hours (1–12), not a row limit: every observation inside the window is returned, so a station reporting more often than hourly yields more than one row per hour. At the default of 1, half-hourly stations return two observations and SPECI-issuing stations can return more. Budget rows by the station reporting interval, never one per station."
    • changedOutput schema / properties / observations / items / properties / ceiling_ft / description
      Previous value: -"Ceiling in feet AGL — the lowest broken, overcast, or obscuration layer. Per FAA AIM 7-1-29 the ceiling is the lowest broken or overcast layer, or the vertical visibility into an obscuration; few and scattered layers are never ceilings. Null when the observation reported no such layer."New value: +"Ceiling in feet AGL — the lowest broken, overcast, or obscuration layer. Per FAA AIM 7-1-29 the ceiling is the lowest broken or overcast layer, or the vertical visibility into an obscuration; few and scattered layers are never ceilings. Null in two cases that are not the same: the observation reported no such layer, or it reported an obscuration whose vertical visibility the station could not determine (a VV/// group), where the ceiling exists and only its height is missing. A sky_condition of OVX marks the second."
    • changedOutput schema / properties / observations / items / properties / clouds / description
      Previous value: -"All reported cloud layers from lowest to highest."New value: +"All reported cloud layers from lowest to highest. Empty whenever the observation published no layer heights, which covers a clear sky, an obscuration with no determinable height, and an observation that reported no sky condition at all; sky_condition distinguishes them. An empty array is not a clear sky on its own."
    • addedOutput schema / properties / observations / items / properties / sky_condition
      Added value: +{
      +  "description": "The sky condition the observation stated when it published no layer heights: CLR, SKC, or CAVOK for a clear or insignificant-cloud report, OVX for an obscuration whose layer carried no height (a VV/// group, and the opposite of clear). Null when clouds carries layers — those are the statement — and also when the observation carried no sky-condition group at all. An empty clouds array beside a null here is an unreported sky, never a clear one.",
      +  "type": [
      +    "string",
      +    "null"
      +  ]
      +}
    • changedOutput schema / properties / observations / items / properties / visibility_sm / description
      Previous value: -"Prevailing visibility in statute miles (e.g., \"10+\", \"3\", \"1/2\")."New value: +"Prevailing visibility in statute miles (e.g., \"10+\", \"3\", \"1/2\"), or the string \"unknown\" when the observation carried no visibility group. \"unknown\" is not a measurement and carries no unit."
    • changedOutput schema / properties / observations / items / required
      Previous value: -[
      -  "station_id",
      -  "name",
      -  "lat",
      -  "lon",
      -  "elevation_ft",
      -  "flight_category",
      -  "metar_type",
      -  "observed_at",
      -  "wind",
      -  "visibility_sm",
      -  "ceiling_ft",
      -  "ceiling_type",
      -  "clouds",
      -  "present_weather",
      -  "temp_c",
      -  "dewpoint_c",
      -  "altimeter_inhg",
      -  "raw_metar"
      -]New value: +[
      +  "station_id",
      +  "name",
      +  "lat",
      +  "lon",
      +  "elevation_ft",
      +  "flight_category",
      +  "metar_type",
      +  "observed_at",
      +  "wind",
      +  "visibility_sm",
      +  "ceiling_ft",
      +  "ceiling_type",
      +  "clouds",
      +  "sky_condition",
      +  "present_weather",
      +  "temp_c",
      +  "dewpoint_c",
      +  "altimeter_inhg",
      +  "raw_metar"
      +]
  6. Changed14 schema fields changed
    • removedOutput schema / properties / observations / items / properties / altimeter_inhg / anyOf
      Removed value: -[
      -  {
      -    "type": "number"
      -  },
      -  {
      -    "type": "null"
      -  }
      -]
    • addedOutput schema / properties / observations / items / properties / altimeter_inhg / type
      Added value: +[
      +  "number",
      +  "null"
      +]
    • removedOutput schema / properties / observations / items / properties / ceiling_ft / anyOf
      Removed value: -[
      -  {
      -    "type": "number"
      -  },
      -  {
      -    "type": "null"
      -  }
      -]
    • addedOutput schema / properties / observations / items / properties / ceiling_ft / type
      Added value: +[
      +  "number",
      +  "null"
      +]
    • removedOutput schema / properties / observations / items / properties / dewpoint_c / anyOf
      Removed value: -[
      -  {
      -    "type": "number"
      -  },
      -  {
      -    "type": "null"
      -  }
      -]
    • addedOutput schema / properties / observations / items / properties / dewpoint_c / type
      Added value: +[
      +  "number",
      +  "null"
      +]
    • removedOutput schema / properties / observations / items / properties / temp_c / anyOf
      Removed value: -[
      -  {
      -    "type": "number"
      -  },
      -  {
      -    "type": "null"
      -  }
      -]
    • addedOutput schema / properties / observations / items / properties / temp_c / type
      Added value: +[
      +  "number",
      +  "null"
      +]
    • removedOutput schema / properties / observations / items / properties / wind / properties / direction_deg / anyOf
      Removed value: -[
      -  {
      -    "type": "number"
      -  },
      -  {
      -    "type": "null"
      -  }
      -]
    • addedOutput schema / properties / observations / items / properties / wind / properties / direction_deg / type
      Added value: +[
      +  "number",
      +  "null"
      +]
    • removedOutput schema / properties / observations / items / properties / wind / properties / gust_kt / anyOf
      Removed value: -[
      -  {
      -    "type": "number"
      -  },
      -  {
      -    "type": "null"
      -  }
      -]
    • addedOutput schema / properties / observations / items / properties / wind / properties / gust_kt / type
      Added value: +[
      +  "number",
      +  "null"
      +]
    • removedOutput schema / properties / observations / items / properties / wind / properties / speed_kt / anyOf
      Removed value: -[
      -  {
      -    "type": "number"
      -  },
      -  {
      -    "type": "null"
      -  }
      -]
    • addedOutput schema / properties / observations / items / properties / wind / properties / speed_kt / type
      Added value: +[
      +  "number",
      +  "null"
      +]
  7. Changed6 schema fields changed
    • changedInput schema / $schema
      Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"
    • addedInput schema / additionalProperties
      Added value: +false
    • changedOutput schema / $schema
      Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"
    • addedOutput schema / anyOf
      Added value: +[
      +  {
      +    "not": {
      +      "required": [
      +        "error"
      +      ]
      +    },
      +    "required": [
      +      "observations",
      +      "requested",
      +      "returned",
      +      "partial"
      +    ]
      +  },
      +  {
      +    "required": [
      +      "error"
      +    ]
      +  }
      +]
    • addedOutput schema / properties / error
      Added value: +{
      +  "additionalProperties": {},
      +  "description": "Present when the call failed. Absent on success.",
      +  "properties": {
      +    "code": {
      +      "description": "JSON-RPC error code for this failure.",
      +      "maximum": 9007199254740991,
      +      "minimum": -9007199254740991,
      +      "type": "integer"
      +    },
      +    "data": {
      +      "additionalProperties": {},
      +      "properties": {
      +        "reason": {
      +          "description": "Machine-readable failure mode. Declared by this tool: `no_stations_found`: None of the requested station IDs returned METAR data. Other values are possible when a failure originates below the handler.",
      +          "examples": [
      +            "no_stations_found"
      +          ],
      +          "type": "string"
      +        },
      +        "recovery": {
      +          "additionalProperties": {},
      +          "description": "Actionable next step for the caller.",
      +          "properties": {
      +            "hint": {
      +              "type": "string"
      +            }
      +          },
      +          "required": [
      +            "hint"
      +          ],
      +          "type": "object"
      +        },
      +        "retryable": {
      +          "description": "Whether retrying may succeed.",
      +          "type": "boolean"
      +        }
      +      },
      +      "type": "object"
      +    },
      +    "message": {
      +      "description": "Human-readable description of what went wrong.",
      +      "type": "string"
      +    }
      +  },
      +  "required": [
      +    "code",
      +    "message"
      +  ],
      +  "type": "object"
      +}
    • removedOutput schema / required
      Removed value: -[
      -  "observations",
      -  "requested",
      -  "returned",
      -  "partial"
      -]
  8. Changed8 schema fields changed
    • addedOutput schema / properties / missing
      Added value: +{
      +  "description": "Requested station IDs absent from the result. Absent when none are missing.",
      +  "items": {
      +    "description": "A requested ICAO station ID that produced no observation.",
      +    "type": "string"
      +  },
      +  "type": "array"
      +}
    • addedOutput schema / properties / notice
      Added value: +{
      +  "description": "Recovery guidance naming the missing station IDs. Present only on a partial result. It lists the candidate causes without asserting one — upstream omits the row either way.",
      +  "type": "string"
      +}
    • changedOutput schema / properties / observations / items / properties / ceiling_ft / description
      Previous value: -"Ceiling in feet AGL — the lowest broken, overcast, or obscuration layer. Per FAA AIM 7-1-13 the ceiling is the lowest broken or overcast layer, or the vertical visibility into an obscuration; few and scattered layers are never ceilings. Null when the observation reported no such layer."New value: +"Ceiling in feet AGL — the lowest broken, overcast, or obscuration layer. Per FAA AIM 7-1-29 the ceiling is the lowest broken or overcast layer, or the vertical visibility into an obscuration; few and scattered layers are never ceilings. Null when the observation reported no such layer."
    • changedOutput schema / properties / observations / items / properties / present_weather / anyOf
      Previous value: -[
      -  {
      -    "additionalProperties": false,
      -    "properties": {
      -      "decoded": {
      -        "description": "Plain-English reading of the group (e.g., \"fog\", \"light rain showers\").",
      -        "type": "string"
      -      },
      -      "raw": {
      -        "description": "Weather group exactly as encoded (e.g., \"FG\", \"-SHRA\", \"+RA BR\").",
      -        "type": "string"
      -      }
      -    },
      -    "required": [
      -      "raw",
      -      "decoded"
      -    ],
      -    "type": "object"
      -  },
      -  {
      -    "type": "null"
      -  }
      -]New value: +[
      +  {
      +    "additionalProperties": false,
      +    "properties": {
      +      "decoded": {
      +        "description": "Plain-English reading of each group, joined with \"; \" (e.g., \"fog\", \"light rain showers; mist\"). A group the decoder does not recognize is carried through as its own raw token rather than half-translated, so compare against raw when a reading still looks coded.",
      +        "type": "string"
      +      },
      +      "raw": {
      +        "description": "Weather groups exactly as encoded, space-delimited (e.g., \"FG\", \"-SHRA\", \"VCTS -RA\").",
      +        "type": "string"
      +      }
      +    },
      +    "required": [
      +      "raw",
      +      "decoded"
      +    ],
      +    "type": "object"
      +  },
      +  {
      +    "type": "null"
      +  }
      +]
    • addedOutput schema / properties / partial
      Added value: +{
      +  "description": "True when a requested station produced no observation. False affirms the result covers every requested station, so full coverage is distinguishable from a short batch rather than being inferred from the count.",
      +  "type": "boolean"
      +}
    • addedOutput schema / properties / requested
      Added value: +{
      +  "description": "Station IDs this call asked for, in the order given.",
      +  "items": {
      +    "description": "An ICAO station ID as requested.",
      +    "type": "string"
      +  },
      +  "type": "array"
      +}
    • addedOutput schema / properties / returned
      Added value: +{
      +  "description": "Distinct station IDs that produced at least one observation. Counted per station, not per row — with hours > 1 a station reporting six times still appears once.",
      +  "items": {
      +    "description": "An ICAO station ID present in the result.",
      +    "type": "string"
      +  },
      +  "type": "array"
      +}
    • changedOutput schema / required
      Previous value: -[
      -  "observations"
      -]New value: +[
      +  "observations",
      +  "requested",
      +  "returned",
      +  "partial"
      +]
  9. Changed18 schema fields changed
    • addedOutput schema / properties / observations / items / properties / altimeter_inhg / anyOf
      Added value: +[
      +  {
      +    "type": "number"
      +  },
      +  {
      +    "type": "null"
      +  }
      +]
    • changedOutput schema / properties / observations / items / properties / altimeter_inhg / description
      Previous value: -"Altimeter setting in inches of mercury."New value: +"Altimeter setting in inches of mercury, or null when the observation carried no altimeter group (common at stations reporting sea-level pressure only)."
    • removedOutput schema / properties / observations / items / properties / altimeter_inhg / type
      Removed value: -"number"
    • changedOutput schema / properties / observations / items / properties / ceiling_ft / description
      Previous value: -"Ceiling in feet MSL — lowest BKN or OVC layer base. Null when sky is clear."New value: +"Ceiling in feet AGL — the lowest broken, overcast, or obscuration layer. Per FAA AIM 7-1-13 the ceiling is the lowest broken or overcast layer, or the vertical visibility into an obscuration; few and scattered layers are never ceilings. Null when the observation reported no such layer."
    • addedOutput schema / properties / observations / items / properties / ceiling_type
      Added value: +{
      +  "anyOf": [
      +    {
      +      "enum": [
      +        "measured",
      +        "indefinite"
      +      ],
      +      "type": "string"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "description": "How the ceiling height was determined: \"measured\" for a broken or overcast layer base, \"indefinite\" for vertical visibility into an obscuration (an OVX layer). Null exactly when ceiling_ft is null."
      +}
    • changedOutput schema / properties / observations / items / properties / clouds / items / properties / base_ft / description
      Previous value: -"Cloud base altitude in feet MSL."New value: +"Cloud base altitude in feet AGL."
    • changedOutput schema / properties / observations / items / properties / clouds / items / properties / cover / description
      Previous value: -"Sky cover code: FEW, SCT, BKN, OVC, SKC, CLR."New value: +"Sky cover code: FEW, SCT, BKN, OVC, SKC, CLR, CAVOK, or OVX. OVX is the decoded form of a VVhhh group — the sky is obscured and the base is the vertical visibility into it, not a cloud bottom."
    • addedOutput schema / properties / observations / items / properties / dewpoint_c / anyOf
      Added value: +[
      +  {
      +    "type": "number"
      +  },
      +  {
      +    "type": "null"
      +  }
      +]
    • changedOutput schema / properties / observations / items / properties / dewpoint_c / description
      Previous value: -"Dewpoint in degrees Celsius."New value: +"Dewpoint in degrees Celsius. 0 is a real reading; null means the observation carried no dewpoint, so it is unknown."
    • removedOutput schema / properties / observations / items / properties / dewpoint_c / type
      Removed value: -"number"
    • addedOutput schema / properties / observations / items / properties / present_weather
      Added value: +{
      +  "anyOf": [
      +    {
      +      "additionalProperties": false,
      +      "properties": {
      +        "decoded": {
      +          "description": "Plain-English reading of the group (e.g., \"fog\", \"light rain showers\").",
      +          "type": "string"
      +        },
      +        "raw": {
      +          "description": "Weather group exactly as encoded (e.g., \"FG\", \"-SHRA\", \"+RA BR\").",
      +          "type": "string"
      +        }
      +      },
      +      "required": [
      +        "raw",
      +        "decoded"
      +      ],
      +      "type": "object"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "description": "Present weather at the station, or null when the observation carried no weather group (a dry, unobscured day)."
      +}
    • addedOutput schema / properties / observations / items / properties / temp_c / anyOf
      Added value: +[
      +  {
      +    "type": "number"
      +  },
      +  {
      +    "type": "null"
      +  }
      +]
    • changedOutput schema / properties / observations / items / properties / temp_c / description
      Previous value: -"Temperature in degrees Celsius."New value: +"Temperature in degrees Celsius. 0 is a real reading; null means the observation carried no temperature, so it is unknown."
    • removedOutput schema / properties / observations / items / properties / temp_c / type
      Removed value: -"number"
    • addedOutput schema / properties / observations / items / properties / wind / properties / speed_kt / anyOf
      Added value: +[
      +  {
      +    "type": "number"
      +  },
      +  {
      +    "type": "null"
      +  }
      +]
    • changedOutput schema / properties / observations / items / properties / wind / properties / speed_kt / description
      Previous value: -"Wind speed in knots."New value: +"Wind speed in knots. 0 is calm (a reported 00000KT); null means the observation carried no wind group, so the speed is unknown."
    • removedOutput schema / properties / observations / items / properties / wind / properties / speed_kt / type
      Removed value: -"number"
    • changedOutput schema / properties / observations / items / required
      Previous value: -[
      -  "station_id",
      -  "name",
      -  "lat",
      -  "lon",
      -  "elevation_ft",
      -  "flight_category",
      -  "metar_type",
      -  "observed_at",
      -  "wind",
      -  "visibility_sm",
      -  "ceiling_ft",
      -  "clouds",
      -  "temp_c",
      -  "dewpoint_c",
      -  "altimeter_inhg",
      -  "raw_metar"
      -]New value: +[
      +  "station_id",
      +  "name",
      +  "lat",
      +  "lon",
      +  "elevation_ft",
      +  "flight_category",
      +  "metar_type",
      +  "observed_at",
      +  "wind",
      +  "visibility_sm",
      +  "ceiling_ft",
      +  "ceiling_type",
      +  "clouds",
      +  "present_weather",
      +  "temp_c",
      +  "dewpoint_c",
      +  "altimeter_inhg",
      +  "raw_metar"
      +]
  10. First observed

TDQS

A4.1/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnly/openWorld/idempotent, so safety is covered; the description adds genuinely non-inferable behavior — named lookups retain every observation in the window while surveys collapse to the latest per station, with flight_category and limit applying only in survey mode. It also signals that results carry both decoded fields and the raw METAR, which a caller would not otherwise know before invoking. It omits any statement about what happens with no selector at all.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Four sentences, front-loaded with purpose then mode semantics then the sibling pointer; no filler. The middle sentence is dense and packs two distinct modes into one clause, but every sentence carries information.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With an output schema present, the description rightly spends its budget on selection-mode behavior rather than return formatting, and all eight parameters are documented in the schema. The one gap is that no parameter is required, yet the description never says what an omission of station_ids, bbox and state yields.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the baseline is 3. The description restates window/survey semantics that the `hours` and `station_ids` schema text already spells out in greater detail, adding only the mild clarification that limit and flight_category are survey-scoped options.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Get current or historical METAR observations') plus the three selection modes (named stations, bounding box, US state/DC) in the first sentence. An agent can distinguish this from aviation_get_taf (forecasts) and aviation_find_stations (ID resolution) without opening any schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives explicit routing advice for the ID-resolution sibling ('Use aviation_find_stations to resolve or verify station identifiers') and contrasts the named-lookup mode against the survey mode. It does not state when to prefer bbox over state or what to do when no selector is supplied, so it stops short of full alternative coverage.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.