nws-weather-mcp-server

Nws Get Observation History

nws_get_observation_history

Read-only

Get a station's recent observation history, newest first — the trend behind the single latest reading nws_get_observations returns (is pressure falling, is the wind picking up). NWS keeps about the last 7 days per station, and a busy airport station reports as often as every 5 minutes, so a day can run past 100 observations: narrow with start (inclusive) and end (exclusive), and page back to older observations with nextCursor. Takes a station ID; use nws_find_stations to find one near a coordinate.

Input Schema

Name	Required	Description
`end`	No	Time to stop before (exclusive), in the same format as start. Omit to start from the most recent observation.
`limit`	No	Observations per page, newest first. A page holds at most 100; a larger limit (up to 10000) is served as 100 and reported through requestedLimit and effectiveLimit. When older observations remain in the window, the response carries nextCursor; pass it back as cursor for them.
`start`	No	Earliest observation time to include (inclusive), as an ISO 8601 date-time with seconds and an offset (e.g., "2026-09-30T00:00:00Z" or "2026-09-29T17:00:00-07:00"). Omit to reach back as far as NWS keeps, about 7 days.
`cursor`	No	Opaque continuation token from a previous response's nextCursor. Omit for the first page. It carries that query's start, end, and applied limit, so those inputs apply to the first page only. Each page continues from just before the oldest observation of the one before it, so pages neither repeat nor skip an observation.
`station_id`	Yes	Observation station ID (e.g., "KSEA"), case-insensitive. Use nws_find_stations to discover station IDs near a coordinate.

Output Schema

Name	Required	Description
`cap`	No	The page limit this page holds, when truncated is true
`error`	No	Present when the call failed. Absent on success.
`shown`	No	Number of observations returned in this page
`notice`	No	Guidance when the requested limit exceeded the page cap, when the window held no observations (NWS keeps about the last 7 days), or when older observations remain beyond this page.
`timeZone`	No	Station time zone when known
`stationId`	No	Observation station ID, uppercased
`truncated`	No	True when older observations remain in the window past this page; continue with nextCursor.
`nextCursor`	No	Opaque token for the next, older page — pass it back as cursor. Present only when older observations remain in the window; omitted on the last page.
`stationName`	No	Station name
`observations`	No	Observations in the window, newest first
`effectiveLimit`	No	The page size applied instead of requestedLimit; continued pages keep it. Absent when the limit was within the cap.
`requestedLimit`	No	The limit asked for, when it exceeded the 100-observation page cap; absent otherwise.

Schema Changelog

Changes observed during successful MCP inspections.

Changed8 schema fields changedOct 23, 2026

changedOutput schema / properties / observations / items / properties / dewpointC / description
```
Previous value: -"Dewpoint in Celsius"New value: +"Dewpoint in Celsius, to one decimal"
```
changedOutput schema / properties / observations / items / properties / heatIndexC / description
```
Previous value: -"Heat index in Celsius"New value: +"Heat index in Celsius, to one decimal"
```

addedOutput schema / properties / observations / items / properties / precipitationLast3HoursMm

Added value: +{
+  "description": "Precipitation in the past 3 hours in mm, to 0.1 mm, carried by the hourly reports nearest 03, 09, 15, and 21 UTC. 0 is a trace (under 0.01 in); null means no amount was reported.",
+  "type": [
+    "number",
+    "null"
+  ]
+}

addedOutput schema / properties / observations / items / properties / precipitationLast6HoursMm

Added value: +{
+  "description": "Precipitation in the past 6 hours in mm, to 0.1 mm, carried by the hourly reports nearest 00, 06, 12, and 18 UTC. 0 is a trace (under 0.01 in); null means no amount was reported.",
+  "type": [
+    "number",
+    "null"
+  ]
+}

addedOutput schema / properties / observations / items / properties / precipitationLastHourMm

Added value: +{
+  "description": "Precipitation in the past hour in mm, to 0.1 mm. 0 is a trace (under 0.01 in), not a dry hour; null means no amount was reported, the usual value for a dry hour and on 5-minute reports.",
+  "type": [
+    "number",
+    "null"
+  ]
+}

changedOutput schema / properties / observations / items / properties / temperatureC / description

Previous value: -"Temperature in Celsius"New value: +"Temperature in Celsius, to one decimal (e.g., 8.3)"

changedOutput schema / properties / observations / items / properties / windChillC / description
```
Previous value: -"Wind chill in Celsius"New value: +"Wind chill in Celsius, to one decimal"
```

changedOutput schema / properties / observations / items / required

Previous value: -[
-  "timestamp",
-  "textDescription",
-  "temperatureC",
-  "dewpointC",
-  "windSpeedKmh",
-  "windDirectionDeg",
-  "windGustKmh",
-  "barometricPressurePa",
-  "visibilityM",
-  "relativeHumidityPct",
-  "heatIndexC",
-  "windChillC",
-  "cloudLayers"
-]New value: +[
+  "timestamp",
+  "textDescription",
+  "temperatureC",
+  "dewpointC",
+  "windSpeedKmh",
+  "windDirectionDeg",
+  "windGustKmh",
+  "barometricPressurePa",
+  "visibilityM",
+  "relativeHumidityPct",
+  "heatIndexC",
+  "windChillC",
+  "precipitationLastHourMm",
+  "precipitationLast3HoursMm",
+  "precipitationLast6HoursMm",
+  "cloudLayers"
+]

Changed12 schema fields changedOct 16, 2026

changedInput schema / properties / cursor / description

Previous value: -"Opaque continuation token from a previous response's nextCursor. Omit for the first page. It carries that query's start, end, and limit, so those inputs apply to the first page only. Each page continues from just before the oldest observation of the one before it, so pages neither repeat nor skip an observation."New value: +"Opaque continuation token from a previous response's nextCursor. Omit for the first page. It carries that query's start, end, and applied limit, so those inputs apply to the first page only. Each page continues from just before the oldest observation of the one before it, so pages neither repeat nor skip an observation."

addedInput schema / properties / end / pattern

Added value: +"^\\s*(?:\\d{4}-\\d{2}-\\d{2}T\\d{2}:\\d{2}:\\d{2}(?:\\.\\d{1,6})?(?:Z|[+-]\\d{2}:\\d{2})\\s*)?$"

changedInput schema / properties / limit / description

Previous value: -"Observations per page (1-100), newest first. When a page fills the limit, pass its nextCursor back as cursor for older observations."New value: +"Observations per page, newest first. A page holds at most 100; a larger limit (up to 10000) is served as 100 and reported through requestedLimit and effectiveLimit. When older observations remain in the window, the response carries nextCursor; pass it back as cursor for them."

changedInput schema / properties / limit / maximum
```
Previous value: -100New value: +10000
```

addedInput schema / properties / start / pattern

Added value: +"^\\s*(?:\\d{4}-\\d{2}-\\d{2}T\\d{2}:\\d{2}:\\d{2}(?:\\.\\d{1,6})?(?:Z|[+-]\\d{2}:\\d{2})\\s*)?$"

addedInput schema / properties / station_id / pattern
```
Added value: +"^\\s*[A-Za-z0-9]+\\s*$"
```

changedOutput schema / properties / cap / description

Previous value: -"The page limit this page filled, when truncated is true"New value: +"The page limit this page holds, when truncated is true"

addedOutput schema / properties / effectiveLimit

Added value: +{
+  "description": "The page size applied instead of requestedLimit; continued pages keep it. Absent when the limit was within the cap.",
+  "type": "number"
+}

changedOutput schema / properties / nextCursor / description

Previous value: -"Opaque token for the next, older page — pass it back as cursor. Present only when this page filled the limit; omitted on the last page."New value: +"Opaque token for the next, older page — pass it back as cursor. Present only when older observations remain in the window; omitted on the last page."

changedOutput schema / properties / notice / description

Previous value: -"Guidance when the window held no observations (NWS keeps about the last 7 days), or when this page filled the limit and older observations may remain."New value: +"Guidance when the requested limit exceeded the page cap, when the window held no observations (NWS keeps about the last 7 days), or when older observations remain beyond this page."

addedOutput schema / properties / requestedLimit

Added value: +{
+  "description": "The limit asked for, when it exceeded the 100-observation page cap; absent otherwise.",
+  "type": "number"
+}

changedOutput schema / properties / truncated / description

Previous value: -"True when this page filled the limit and older observations in the window may remain; continue with nextCursor."New value: +"True when older observations remain in the window past this page; continue with nextCursor."

AddedSep 30, 2026

TDQS

A4.6/5.0

Behavior5/5

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

Annotations only declare readOnlyHint; the description adds substantive behavioral context beyond that: ~7-day server-side retention, station reporting cadence of roughly 5 minutes, the 100-per-day-scale volume, and that nextCursor pages strictly could continue without repeats or gaps. These are data-availability and paging traits an agent cannot infer from annotations or schema.

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?

Front-loaded with purpose, then usage, then parameter guidance, with no filler sentences. It is fairly dense and restates some pagination/limit mechanics the schema already covers, which is minor redundancy rather than waste.

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

Completeness5/5

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

Output schema exists, so return values need not be explained; the description instead covers what the schema cannot — retention window, reporting frequency, volume expectations, and how paging composes the result. Nothing needed to call this correctly is missing.

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 coverage is 100%, so start/end inclusivity, limit caps, and cursor opacity are already documented in the schema; the description mostly corroborates those (inclusive start, exclusive end, nextCursor). It adds the operational reason to narrow a window (100+ observations per busy day) but no new syntax or format meaning.

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 a station's recent observation history, newest first') and explicitly contrasts it with the sibling nws_get_observations ('the trend behind the single latest reading'). An agent can distinguish history-listing from latest-reading without opening either schema.

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

Usage Guidelines5/5

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

Names both alternatives and their selection conditions: nws_get_observations for the single latest reading, nws_find_stations to discover station IDs near a coordinate. It also gives concrete narrowing advice ('narrow with start... and end') when a day exceeds 100 observations.

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

All nws-weather-mcp-server tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.