Statec Lu

What Statec Lu MCP server does

The Statec Lu MCP server gives an MCP client access to datasets published by STATEC, Luxembourg’s Institut national de la statistique et des études économiques. Its main use is to help an agent find a relevant statistical dataflow, understand the dataset’s schema, and retrieve observations from selected series.

STATEC publishes hundreds of dataflows covering statistical subjects such as population and unemployment. Search results include a dataflow identifier, English name, and short descriptive metadata such as publication date, periodicity, author, and category. The identifier is then used with the structure and data retrieval tools.

How it works

A reliable query normally follows three steps:

1. Call list_dataflows with a keyword such as population or unemployment to locate a dataset.
2. Call dataflow_structure with the selected dataflow ID to inspect its ordered dimensions and valid codes.
3. Call get_data with an SDMX key and, when needed, start and end periods.

The SDMX key contains one position for each dimension, in the exact order returned by dataflow_structure. A blank position acts as a wildcard. For example, a key can select a variable and annual frequency while leaving another dimension unrestricted. Omitting the key, or passing an empty value, requests all series and may return a large result.

Setup and configuration

The Statec Lu MCP server can be connected through the hosted Pipeworx endpoint:

{
  "mcpServers": {
    "statec-lu": {
      "url": "https://gateway.pipeworx.io/statec-lu/mcp"
    }
  }
}

The hosted connection exposes the pack’s tools along with shared Pipeworx gateway tools. Its initialization response is the authoritative source for the tools available at connection time. The gateway can also route questions through ask_pipeworx.

For a local connection, run the package with npx:

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

The local process communicates over stdin and stdout and exposes only this pack’s tools. The hosted endpoint also provides HTTP tool calls for clients that do not use MCP directly.

Tools and capabilities

• list_dataflows: browse or keyword-search STATEC dataflows.
• dataflow_structure: inspect a dataflow’s dimension order and permitted codes.
• get_data: retrieve decoded series and period values using dimension filters and optional date bounds.

Limitations and notes

The Statec Lu MCP server requires schema inspection before constructing a precise get_data key. Dimension positions and valid codes are dataset-specific, so a key copied from one dataflow should not be assumed to work for another. Broad requests can produce large responses, particularly when the key is omitted or uses multiple wildcards.

The hosted pack endpoint includes shared Pipeworx meta-tools in addition to the three listed tools; the standalone package does not. The package is licensed under MIT.