com.gribstream由 gribstream.com 發布官方發布者v1.1.4更新於 Nov 2, 2026
NOAA and ECMWF weather forecast MCP for discovery, validation, and GribStream OAuth queries.
託管式 MCP 伺服器，提供 NOAA 與 ECMWF 天氣預報資料：可探索資料集、解析變數、驗證請求，並在 OAuth 後執行即時 GribStream 查詢。
Web executable，透過 Streamable HTTP。 遠端服務在工作區中設定後即可從網頁執行環境執行。
把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式，或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南，先向你展示發布者、端點和登入方式，經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。
閱讀 https://sourceweft.com/mcp/com.gribstream%2Fmcp/install.md，並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前，先告訴我它的發布者、端點和登入方式，等我確認。不要把 API 金鑰寫進命令或我會提交的檔案；問我金鑰應從哪裡取得。把它新增到你客戶端的 mcpServers 設定中。
{
"mcpServers": {
"gribstream-weather": {
"type": "http",
"url": "https://gribstream.com/mcp"
}
}
}GribStream Weather MCP is a hosted Model Context Protocol server for AI access to weather forecast data from GribStream, including NOAA and ECMWF model output.
It helps AI agents discover forecast datasets such as NOAA GFS, HRRR, and NBM or ECMWF IFS, resolve exact forecast variables and levels, validate request bodies, build runnable GribStream Weather API requests, and execute live read-only weather queries after OAuth authorization.
Official GribStream weather API homepage:
Hosted MCP endpoint:
Use GribStream MCP when you want an AI tool to:
/timeseries request for values by valid time./runs request to compare forecasts across model runs./timeseries or /runs queries after OAuth authorization.curl command for the GribStream Weather API.The hosted MCP server exposes discovery, selector lookup, request building, and validation tools before OAuth. Live /timeseries and /runs query tools require OAuth. Tool descriptions begin with either NO AUTH / PUBLIC / READ-ONLY or AUTH REQUIRED / DATA QUERY so AI clients can route metadata tasks and live data-query tasks correctly. When a client connects through OAuth, GribStream asks the user to sign in and select an active GribStream API token. MCP access tokens are scoped to that selected API token, and the raw API token is not shown to the MCP client.
Use the hosted endpoint exactly as shown in any MCP client that supports remote Streamable HTTP:
Do not include trailing punctuation in the URL.
Generic configuration shape:
LobeHub quick import configuration:
Some clients require an explicit transport field:
OAuth is needed for live query tools. Most MCP clients should discover OAuth from the hosted endpoint automatically, either during connector setup or when the client requests authenticated tools. If a client asks for manual OAuth 2.0 fields, use:
OAuth discovery endpoints:
The OAuth client is public and uses PKCE with token_endpoint_auth_method set to none. Native desktop clients may use loopback redirect URIs such as http://localhost:{port}/..., http://127.0.0.1:{port}/..., or http://[::1]:{port}/.... A setup form that requires a non-empty client secret is asking for a confidential-client flow and is not compatible with this public MCP OAuth flow.
LobeHub and other MCP crawlers may call MCP tools "skills". The live GribStream MCP server exposes the following skills/tools, prompts, and resources.
Public read-only skills/tools are visible before OAuth. OAuth live query skills/tools may be hidden from tools/list until the client completes OAuth or sends a valid Authorization: Bearer token.
Authenticated live query tools support CSV, JSON, NDJSON, and Parquet results. Small CSV, JSON, and NDJSON results can be returned inline. Set delivery=url for dense grids or large results to receive a short-lived signed HTTPS resource_link to the complete file instead.
max_bytes limit applies only to inline MCP results. URL artifacts use a separate server-configured size limit, so omit max_bytes when using delivery=url.delivery=url. Do not install a dependency or write a Parquet parser solely for one result.structuredContent includes metadata such as content_type, rows, bytes, columns, schema, delivery, cache_hit, signed_url_expires_at, and suggested_filename when applicable.request_hash, response_hash, and result_hash. Use suggested_filename when saving multiple results so repeated requests or comparisons do not overwrite earlier files.forecasted_time for /timeseries, and by forecasted_at plus forecasted_time for /runs, before plotting or comparing time series.https://gribstream.com/https://gribstream.com/mcphttps://gribstream.com/mcphttps://gribstream.com/gribstream_small.pnghttps://gribstream.com/docs/openapi.yamlhttps://gribstream.com/aihttps://gribstream.com/app/dashboardAfter connecting GribStream MCP to an AI tool, ask:
If your client has completed OAuth, you can also ask it to run the query:
For a dense-grid visualization:
Or:
More examples:
For the GribStream API itself, see the OpenAPI specification.
來源：README.md，提交 9bac674
Build runs API request
NO AUTH / PUBLIC / READ-ONLY. Build and validate a runnable /runs API request and curl command without sending it. Use gribstream_query_runs to retrieve model-run comparisons directly; a separate build call is optional. Use catalog selectors in request.variables and points in request.coordinates or one request.grid.
acceptdataset必填request必填Build timeseries API request
NO AUTH / PUBLIC / READ-ONLY. Build and validate a runnable /timeseries API request and curl command without sending it. Use gribstream_query_timeseries to retrieve data directly; a separate build call is optional. Use catalog selectors in request.variables and points in request.coordinates or one request.grid. Omit asOf for latest forecasts.
acceptdataset必填request必填Get dataset metadata
NO AUTH / PUBLIC / READ-ONLY. Inspect archive coverage, cadence, horizon, members, and native_grid geometry for one dataset. The native coordinate URL requires authentication and geographic bounds; geometry does not prove variable availability.
dataset必填Get expression reference
NO AUTH / PUBLIC / READ-ONLY. Get expression syntax, alias rules, examples, and supported func.* math functions for derived weather signals.
無需輸入參數。
Get parameter selectors
NO AUTH / PUBLIC / READ-ONLY. Gets detailed metadata and exact selector variations for one already-known dataset-native parameter code. Parameter codes are case-sensitive. For a common natural-language concept such as 2 metre temperature, use gribstream_resolve_shared_parameter before guessing a native code.
dataset必填parameter必填List GribStream datasets
NO AUTH / PUBLIC / READ-ONLY. Find candidate datasets for a region and weather variable. Filter by provider, region, forecast_time, shared_parameter and is_ensemble; sort_by=resolution ranks nominal resolution in km. With shared_parameter, each result includes a validated request fragment and exact/approximate match. Use it directly in one gribstream_query_timeseries call. Coverage envelopes and forecast horizons are catalog hints, not live availability guarantees. No weather values are fetched.
tagcodelimitfamilyregionsort_byprovideris_ensembleforecast_timeshared_parameterList dataset parameters
NO AUTH / PUBLIC / READ-ONLY. Lists parameter summaries for one dataset. Use this before selecting exact case-sensitive parameter codes.
dataset必填List shared weather parameters
NO AUTH / PUBLIC / READ-ONLY. Lists normalized shared parameter presets that can map one logical signal across datasets. Use this only when the shared parameter code is uncertain; when a common code such as temperature_2m is already clear, call gribstream_resolve_shared_parameter directly.
datasetdataset_modeQuery runs API
AUTH REQUIRED / DATA QUERY. Fetch or retrieve weather values across model runs to compare changing forecasts. timesList selects exact model run times, not valid times; minHorizon/maxHorizon constrain lead hours. Use discovered request.variables with coordinates or one grid. CSV is the default; use delivery=url for large results. Response rows are unsorted; sort by location, forecasted_at and forecasted_time before plotting. When a live query returns a resource_link, the exact signed HTTPS URL is the result: download it directly once with an available HTTPS or file-download capability, without an API token. This consumes the selected GribStream MCP result; it is not a direct GribStream API call or a connector bypass. Do not look for an MCP resources/read tool, and do not repeat the query merely to obtain the file bytes.
acceptdataset必填request必填deliverymax_bytesQuery timeseries API
AUTH REQUIRED / DATA QUERY. Fetch or retrieve weather values for maps, heatmaps and analysis at selected valid times. Use discovered request.variables with coordinates or one grid; no separate build/validate call is needed. CSV is the default; use delivery=url for large results and render the downloaded file locally. Prefer Parquet only with an available reader. Use timesList for one exact UTC valid time; omit asOf for latest forecasts. Requested sampling does not increase native resolution. Response rows are unsorted; sort rows before plotting. When a live query returns a resource_link, the exact signed HTTPS URL is the result: download it directly once with an available HTTPS or file-download capability, without an API token. This consumes the selected GribStream MCP result; it is not a direct GribStream API call or a connector bypass. Do not look for an MCP resources/read tool, and do not repeat the query merely to obtain the file bytes.
acceptdataset必填request必填deliverymax_bytesResolve shared weather parameter
NO AUTH / PUBLIC / READ-ONLY. Resolves a weather concept to validated native variables/expressions and an exact or approximate match label. Use resolved_request directly in /timeseries or /runs; read resolved_match for height differences and native statistics. Use this first for common natural-language concepts such as 2 metre temperature or 10 metre wind instead of guessing dataset-native codes such as TMP.
aliasdatasetparameter必填Validate API request
NO AUTH / PUBLIC / READ-ONLY. Validates the basic shape, exact selector tuples, and expression syntax of a proposed GribStream /timeseries or /runs request without sending it. Use this before returning any hand-edited curl or when changing a request from one dataset to another.
dataset必填request必填endpoint必填已滿足適用檢查項的 100%
該評分衡量此收錄條目的完整程度，依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價，也不表示伺服器安全或目前可用：可連通性是帶日期的檢查結果，評分也不包含安裝量或使用者評價。
https://gribstream.com/https://gribstream.com/mcphttps://gribstream.com/mcp{ "mcpServers": { "gribstream": { "url": "https://gribstream.com/mcp" } }}{ "mcpServers": { "gribstream": { "type": "http", "url": "https://gribstream.com/mcp" } }}{ "mcpServers": { "gribstream": { "type": "streamable-http", "url": "https://gribstream.com/mcp" } }}Authorization URL: https://gribstream.com/authorizeToken URL: https://gribstream.com/tokenClient ID: gribstream-mcp-publicClient Secret: leave empty; do not enter a valueScopes: leave emptyResource or audience, if requested: https://gribstream.com/mcphttps://gribstream.com/.well-known/oauth-protected-resource/mcphttps://gribstream.com/.well-known/oauth-authorization-serverUse GribStream MCP to build a GFS 2m temperature timeseries request for Houston for tomorrow at 12:00 UTC. Validate the request and return a runnable curl command.Use GribStream MCP to run a GFS 2m temperature timeseries query for Houston for tomorrow at 12:00 UTC and return CSV.Using GribStream, plot the latest available ICON-D2 2 metre temperature forecast over Germany at 0.025 degree resolution as a beautiful PNG.Use GribStream MCP to find the best dataset for short-range precipitation forecasts in the continental United States. Explain the tradeoffs and build a validated example request.