Gribstream 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:

Use GribStream MCP when you want an AI tool to:

- List public GribStream weather datasets.
- Choose a NOAA, ECMWF, or other forecast model for a region, horizon, or weather task.
- Query GFS, HRRR, NBM, IFS, and other numerical weather prediction model data exposed by GribStream.
- Find exact variable selectors, levels, and aliases.
- Build a/timeseriesrequest for values by valid time.
- Build a/runsrequest to compare forecasts across model runs.
- Validate a GribStream API request before running it.
- Execute live/timeseriesor/runsqueries after OAuth authorization.
- Retrieve dense forecast grids as downloadable files without embedding the complete result in the conversation.
- Generate a copy-pasteablecurlcommand for the GribStream Weather API.

Use the hosted endpoint exactly as shown in any MCP client that supports remote Streamable HTTP:

Do not include trailing punctuation in the URL.

{ "mcpServers": { "gribstream": { "url": "https://gribstream.com/mcp" } } }

{ "mcpServers": { "gribstream": { "type": "http", "url": "https://gribstream.com/mcp" } } }

Some clients require an explicit transport field:

{ "mcpServers": { "gribstream": { "type": "streamable-http", "url": "https://gribstream.com/mcp" } } }

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:

Authorization URL: https://gribstream.com/authorize Token URL: https://gribstream.com/token Client ID: gribstream-mcp-public Client Secret: leave empty; do not enter a value Scopes: leave empty Resource or audience, if requested: https://gribstream.com/mcp

https://gribstream.com/.well-known/oauth-protected-resource/mcp https://gribstream.com/.well-known/oauth-authorization-server

The OAuth client is public and uses PKCE withtoken_endpoint_auth_methodset tonone. Native desktop clients may use loopback redirect URIs such ashttp://localhost:{port}/...,http://127.0.0.1:{port}/..., orhttp://[::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 fromtools/listuntil the client completes OAuth or sends a validAuthorization: Bearertoken.

Authenticated live query tools support CSV, JSON, NDJSON, and Parquet results. Small CSV, JSON, and NDJSON results can be returned inline. Setdelivery=urlfor dense grids or large results to receive a short-lived signed HTTPSresource_linkto the complete file instead.

- URL delivery works with CSV, JSON, NDJSON, and Parquet. Parquet always uses URL delivery.
- Download the exact signed URL once without an API token. This consumes the MCP result; it is not a separate direct GribStream API request.
- The 10 MBmax_byteslimit applies only to inline MCP results. URL artifacts use a separate server-configured size limit, so omitmax_byteswhen usingdelivery=url.
- For dense grids, prefer Parquet when a compatible reader is already installed; otherwise use CSV withdelivery=url. Do not install a dependency or write a Parquet parser solely for one result.
- Use one grid request and only split after an actual request or artifact-size error. If splitting is necessary, split time first, variables second, and space last.
- structuredContentincludes metadata such ascontent_type,rows,bytes,columns,schema,delivery,cache_hit,signed_url_expires_at, andsuggested_filenamewhen applicable.
- Inline tabular results also include a preview plusrequest_hash,response_hash, andresult_hash. Usesuggested_filenamewhen saving multiple results so repeated requests or comparisons do not overwrite earlier files.
- Rows are streamed from parallel extraction and are not guaranteed to be sorted. Sort byforecasted_timefor/timeseries, and byforecasted_atplusforecasted_timefor/runs, before plotting or comparing time series.

- Official GribStream weather API homepage:https://gribstream.com/
- Endpoint:https://gribstream.com/mcp
- MCP page:https://gribstream.com/mcp
- Icon:https://gribstream.com/gribstream_small.png
- OpenAPI:https://gribstream.com/docs/openapi.yaml
- AI setup docs:https://gribstream.com/ai
- API dashboard:https://gribstream.com/app/dashboard

After connecting GribStream MCP to an AI tool, ask:

Use 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.

If your client has completed OAuth, you can also ask it to run the query:

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.

- Client configuration examples
- Weather request workflows

For the GribStream API itself, see theOpenAPI specification.

Connect SkyLink API to Claude Desktop, Cursor, or any MCP client in under a minute. Ask about live flights, weather, NOTAMs, and ADS-B positions — your AI gets real data, not hallucinations.