An MCP server that allows searching and exploring datasets from the AWS Open Data Registry.

• search_datasets: Search for datasets by name, description, or tags with fuzzy matching
• get_dataset: Get detailed information about a specific dataset
• Automatic caching of the AWS Open Data Registry for fast lookups

npm install

npm start

The server will be available at http://localhost:3000/mcp using the streamable HTTP transport. You can install this in Claude Code with: claude mcp add aws-open-data --transport http http://localhost:3000/mcp. You'll then need to restart Claude Code to pick up the new MCP.

Search for datasets matching a query. If no query is provided, returns all datasets.

Arguments:

• query (string, optional): Search query to match against dataset names, descriptions, and tags. Empty query returns all datasets
• limit (number, optional): Maximum number of results to return (default: 25)
• detail (enum, optional): Detail level - 'nameOnly' (just names), 'minimal' (name and description), or 'full' (complete dataset info). Default: 'minimal'

Example:

{
  "query": "satellite",
  "limit": 5,
  "detail": "minimal"
}

Get detailed information about a specific dataset.

Arguments:

• filename (string, required): The filename of the dataset (must end with .yaml)

Example:

{
  "filename": "sentinel-1.yaml"
}