Data Issy

What Data Issy does

Data Issy exposes Issy-les-Moulineaux Open Data from data.issy.com as MCP tools through the Pipeworx gateway. It is intended for agents that need to locate and inspect municipal datasets before retrieving records. The available subject areas include city services, mobility, environment, and smart-city data.

The integration is independent and unofficial; it is not affiliated with, endorsed by, or published by the upstream data provider.

How it works

An MCP client can connect directly to the hosted endpoint at https://gateway.pipeworx.io/data-issy/mcp. The connection exposes the Data Issy tools and also includes shared Pipeworx gateway meta-tools. These shared tools can route broader questions through the wider Pipeworx catalog.

A local npm bridge is also available as @pipeworx/mcp-data-issy. The client starts the bridge with npx, and the bridge forwards requests to the same hosted endpoint. Network access is required because the bridge has no offline mode.

The normal workflow is to search for a dataset, inspect its metadata to learn the available fields, and then query records using the returned schema.

Setup and configuration

For a hosted MCP connection, add the following endpoint to an MCP client configuration:

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

The npm option uses a local stdio process:

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

Anonymous calls are limited to 50 per day. Adding a Pipeworx account key raises the limit to 200 calls per day. Supply the key as PIPEWORX_API_KEY; requests use it as a Bearer token and are metered in the same way as hosted calls.

Set PIPEWORX_META_TOOLS=1 with the npm bridge to list the shared Pipeworx tools in addition to the Data Issy tools. The default local configuration lists only this pack's tools.

Tools and capabilities

• search_datasets finds Issy-les-Moulineaux datasets by keyword and returns dataset IDs, titles, themes, and record counts.
• dataset_info returns a dataset's fields, schema, themes, and record count. Use it to identify column names before writing a query.
• query retrieves dataset records with ODSQL. It supports where filters, group_by and select aggregation, order_by sorting, and limit/offset pagination.

The tools can also be called over HTTP without an MCP client. For example, the search tool is available through the Pipeworx tools API at https://gateway.pipeworx.io/v1/tools/data_issy_search_datasets.

Limitations and notes

The local bridge depends on network access to the Pipeworx gateway. A pack-specific hosted connection also lists shared gateway meta-tools, so its tool list is larger than the three Data Issy tools. The exact scope is reported by the connection's initialize response.

If the environment uses an HTTPS proxy, Node's built-in fetch does not automatically read HTTPS_PROXY. Newer Node releases can use that variable when NODE_USE_ENV_PROXY=1 is set. The full Pipeworx gateway is another connection option when an agent needs direct access to all catalog packs rather than only the Data Issy pack.