mcpField
C
Climate Data Store (CDS) MCP Server

Climate Data Store (CDS) MCP Server

Local
github.com/albertdowData & Analytics1Updated 78w ago
Repository

About Climate Data Store (CDS) MCP Server

Climate Data Store (CDS) MCP Server lets LLM clients browse CDS catalogues, submit data requests, and track or download job results. It is intended for users working with CDS data through the `datapi` API.

Key features

  1. 01Browse catalogue collections
  2. 02Inspect collection details
  3. 03Submit download jobs
  4. 04List jobs by status
  5. 05Download job results
  6. 06Configure credentials with `.env`

Use cases

  1. 01Find available CDS collections
  2. 02Request a dataset download
  3. 03Check job status and retrieve results

MCP support

Exposes CDS catalogue and job-management operations as tools to MCP clients, including collection lookup, job submission, status listing, and result download.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

No install method listed yet

We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.

Repository

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

Climate Data Store (CDS) MCP Server

Overview

A Model Context Protocol (MCP) server implementation that provides the LLM an interface to retrieve CDS catalogue data and job statuses. The underlying API is datapi - docs found here.

Features

  • Tools:

    • get_jobs: find the jobs available, optionally add a filter based on status. Returns a list of job ids.
    • download_job_result: downloads the job result using job id.
    • get_all_collections: gets all available collection ids in the catalogue.
    • get_collection_by_id: fetches information for a specified collection.
    • submit_job: submits a download request.

  • Environment variable support using .env.

Prerequisites

  • Python 3.13 or higher.
  • CDS API Key: here
  • MCP Host/Client: tested on Claude Desktop and the MCP Inspector.

Installation

  • Clone the repository:
git clone [email protected]:albertdow/mcp-datapi.git
cd mcp-datapi
  • Install dependencies (using uv):
uv add "mcp[cli]" datapi python-dotenv
  • Setup CDS API key by creating a .env file and adding the following:
DATAPI_URL=<DATAPI_URL>
DATAPI_KEY=<DATAPI_KEY>

Details on CDS API key setup can be found here.

Usage

Dev Mode with MCP Inspector

Test the server locally:

mcp dev datapi_server.py
Integrate with Claude Desktop
mcp install datapi_server.py --name "DatapiServer" -f .env

Or directly put add to your claude_desktop_config.json:

{
    "mcpServers": {
        "DatapiServer": {
            "command": "uv",
            "args": [
                "--directory",
                "mcp-datapi",
                "run",
                "mcp_datapi/datapi_server.py"
            ],
            "env": {
                "DATAPI_URL": "<DATAPI_URL>",
                "DATAPI_KEY": "<DATAPI_KEY>"
            }
        }
    }
}

Note:

  • I had to specify the path to uv, e.g. /Users/username/.local/bin/uv.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.