Data Edmonton

What Data Edmonton does

Data Edmonton exposes Edmonton open data from data.edmonton.ca to MCP-compatible clients. The service works with the Socrata SODA API and supports both common datasets and arbitrary catalogue resources.

Its three dedicated tools cover the main discovery and query flow:

• edmonton_datasets searches the Edmonton catalogue by keyword and returns dataset names, descriptions, and Socrata resource IDs.
• edmonton_recent retrieves newest-first records from supported common datasets using friendly names. The documented names are 311 and permits.
• edmonton_query runs a SoQL query against a resource ID.

The server is intended for questions involving Edmonton civic and municipal open data, such as recent 311 requests, building permits, or records from another dataset found in the catalogue.

How it works

The hosted MCP endpoint is available at https://gateway.pipeworx.io/data-edmonton/mcp. Add that URL to an MCP client to connect over HTTP. A full Pipeworx gateway endpoint is also available at https://gateway.pipeworx.io/mcp; both endpoints reach the same gateway and data sources, while the scoped endpoint lists the Edmonton pack’s tools directly.

A typical workflow is to search for a dataset with edmonton_datasets, take the returned eight-character Socrata resource ID, and pass it to edmonton_query. For the documented common datasets, edmonton_recent avoids requiring a resource ID. Its optional SoQL where filter can narrow the returned records.

The scoped hosted endpoint also exposes shared Pipeworx meta-tools, including ask_pipeworx, discover_tools, search_within, remember, and recall. The exact tool list is reported by the MCP connection’s initialize response. ask_pipeworx can route a plain-language question across the broader Pipeworx catalogue.

Setup and configuration

For a hosted connection, add the following MCP server entry:

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

The npm package @pipeworx/mcp-data-edmonton provides a local stdio bridge that forwards calls to the same hosted endpoint. Configure it with npx:

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

The bridge requires network access and has no offline mode. Anonymous use is capped at 50 calls per day. A free Pipeworx account raises the limit to 200 calls per day; provide its key as PIPEWORX_API_KEY in the client’s environment configuration. Set PIPEWORX_META_TOOLS=1 if the local bridge should list the gateway-wide meta-tools as well as the Edmonton tools.

The tools can also be called through HTTP. For example, send a JSON POST to /v1/tools/edmonton_recent with {"dataset":"311"}. Tool details are available from the corresponding /v1/tools/<tool> endpoint.

Tools and capabilities

edmonton_recent is suited to latest-record lookups for 311 requests and building permits. It returns rows newest first and accepts a where condition.

edmonton_datasets helps locate resources when the user’s request does not match one of the friendly dataset names. Its results provide the resource ID needed by the query tool.

edmonton_query accepts a Socrata resource ID and supports SoQL clauses for where, select, group, order, limit, and offset. This allows filtering, projection, aggregation, sorting, and pagination against Edmonton open-data resources.

Limitations and notes

This is an independent, unofficial integration and is not affiliated with, endorsed by, or published by the upstream data provider. The npm bridge forwards requests to the hosted gateway rather than running an offline data service. Anonymous and account-based usage have different daily call limits, and calls using an API key are metered and billed according to the gateway’s handling of authenticated requests.