Data Novascotia

What Data Novascotia MCP server does

The Data Novascotia MCP server gives MCP-compatible agents access to datasets published in Nova Scotia’s Open Data catalog at data.novascotia.ca. It is designed for workflows that begin with finding a relevant dataset, continue with examining its fields, and finish by retrieving selected records.

The service works with Socrata datasets. Search results include each dataset’s resource ID, name, description, category, and update date. The resource ID is then used to request metadata or execute a query against that specific dataset.

How it works

The Data Novascotia MCP server exposes three pack-specific tools. datasets searches the catalog by keyword. metadata returns a dataset’s schema and related details, including columns, data types, row count, category, and last-updated information. query sends a Socrata SoQL request and returns matching rows as JSON.

Queries can include where, select, group, and order clauses, along with limit and offset. The SoQL clause values are supplied without the leading dollar sign. A typical workflow is to search for a subject, use the returned resource ID to inspect the schema, and then construct a query using the available column names and types.

Setup and configuration

The Data Novascotia MCP server can be connected through the Pipeworx hosted endpoint:

{
  "mcpServers": {
    "data-novascotia": {
      "url": "https://gateway.pipeworx.io/data-novascotia/mcp"
    }
  }
}

It can also run locally over standard input and output with no Pipeworx account:

{
  "mcpServers": {
    "data-novascotia": {
      "command": "npx",
      "args": ["-y", "@pipeworx/mcp-data-novascotia"]
    }
  }
}

Run npx -y @pipeworx/mcp-data-novascotia directly to verify that the local process starts. The hosted service also exposes HTTP tool calls, including POST https://gateway.pipeworx.io/v1/tools/data_novascotia_datasets with a JSON query value.

Tools and capabilities

• datasets: Find Nova Scotia Open Data datasets by keyword.
• metadata: Inspect schema, column types, row count, category, and update information.
• query: Run filtered, selected, grouped, ordered, paginated SoQL queries.
• Return dataset search results and query rows as structured JSON.

Limitations and notes

The hosted pack endpoint lists the three Data Novascotia tools together with shared Pipeworx meta-tools. Its exact tool list can change, and the connection’s initialize response is authoritative. The local package exposes only the pack-specific tools and does not provide the gateway’s shared routing tools.

The hosted endpoint and the full Pipeworx gateway use the same underlying gateway; they differ in which tools are listed directly. The ask_pipeworx meta-tool can route questions across the wider catalog when using a gateway connection. The standalone package does not provide that routing. No environment variables are documented for this server, and the project is licensed under MIT.