mcp-cbs-cijfers-open-data
About
MCP server for working with CBS Cijfers Open Data
Details
- Author
- dstotijn
- Downloads
- 326
- Categories
- Search
Jump to
- Lists all available CBS data catalogs
- Queries datasets with advanced OData filtering, sorting, and pagination
- Retrieves dataset dimensions and their possible values
- Fetches observations (statistical data) with optional filters
- Supports OData $select, $filter, $orderby, $top, $skip, and $search
- Provides OData service metadata document for the CBS API
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
mcp-cbs-cijfers-open-dataCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
Install Go 1.24, then configure your MCP host (e.g., Claude Desktop, Cursor) by adding the server entry with command: "go" and args: ["run", "github.com/dstotijn/mcp-cbs-cijfers-open-data@latest"]. The server can also be run manually using the mcp-cbs-cijfers-open-data binary with optional flags -http, -sse, and -stdio to choose transport.
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"mcp-cbs-cijfers-open-data": {
"cbs-cijfers-open-data": {
"command": "go",
"args": [
"run",
"github.com/dstotijn/mcp-cbs-cijfers-open-data@latest"
]
}
}
}
}McpServers
{
"cbs-cijfers-open-data": {
"command": "go",
"args": [
"run",
"github.com/dstotijn/mcp-cbs-cijfers-open-data@latest"
]
}
}mcp-cbs-cijfers-open-data
This project provides a Model Context
Protocol (MCP) server for the CBS Open Data
API. It
allows AI tools to interact with the CBS Open Data through MCP
tools.
Features
- Implements a Model Context Protocol (MCP)
server
- Provides tools for interacting with the CBS Open Data:
- List available datasets and catalogs
- Query dataset metadata and dimensions
- Retrieve statistical data with filtering options
- Explore dimension values
Tools
This MCP server provides the following tools for AI assistants to interact with the CBS Open Data API:
get_catalogs
Retrieves all available CBS data catalogs.
Parameters:
- None
query_datasets
Lists available datasets from the CBS Open Data API with advanced filtering, sorting, and pagination options.
Parameters:
- catalog (required): Catalog identifier (use "CBS")
- select (optional): OData $select parameter to choose specific fields
- filter (optional): OData $filter parameter for filtering results
- orderby (optional): OData $orderby parameter for sorting results
- top (optional): OData $top parameter for limiting results
- skip (optional): OData $skip parameter for pagination
- count (optional): OData $count parameter to include count in response
- search (optional): OData $search parameter for free-text search
- expand (optional): OData $expand parameter to include related entities
get_dimensions
Retrieves all dimensions for a specific dataset.
Parameters:
- catalog (required): Catalog identifier
- dataset (required): Dataset identifier
get_dimension_values
Retrieves all values for a specific dimension with filtering and sorting options.
Parameters:
- catalog (required): Catalog identifier
- dataset (required): Dataset identifier
- dimension (required): Dimension identifier
- select (optional): OData $select parameter to choose specific fields
- filter (optional): Additional OData $filter parameter for filtering results
- orderby (optional): OData $orderby parameter for sorting results
- top (optional): OData $top parameter for limiting results
get_observations
Retrieves observations from a specific dataset with optional filters.
Parameters:
- catalog (required): Catalog identifier
- dataset (required): Dataset identifier
- filters (optional): Filters as key-value pairs
- limit (optional): Maximum number of observations to return
query_observations
Queries observations with advanced OData query options to filter, sort, and shape results.
Parameters:
- catalog (required): Catalog identifier
- dataset (required): Dataset identifier
- select (optional): OData $select parameter to choose specific fields
- filter (optional): OData $filter parameter for filtering results
- orderby (optional): OData $orderby parameter for sorting results
- top (optional): OData $top parameter for limiting results
- skip (optional): OData $skip parameter for pagination
- count (optional): OData $count parameter to include count in response
- search (optional): OData $search parameter for free-text search
- expand (optional): OData $expand parameter to include related entities
get_metadata
Retrieves the OData service metadata document for the CBS API.
Parameters:
- None
Requirements
- Go 1.24
Installation
Configuration for common MCP hosts (Claude Desktop, Cursor):
{
"mcpServers": {
"cbs-cijfers-open-data": {
"command": "go",
"args": ["run", "github.com/dstotijn/mcp-cbs-cijfers-open-data@latest"]
}
}
}
Alternatively, you can manually install the program (given you have Go installed
):
go install github.com/dstotijn/mcp-cbs-cijfers-open-data@latest
Usage
$ mcp-cbs-cijfers-open-data --help
Usage of mcp-cbs-cijfers-open-data:
-http string
HTTP listen address for JSON-RPC over HTTP (default ":8080")
-sse
Enable SSE transport
-stdio
Enable stdio transport (default true)
Typically, your MCP host will run the program and start the MCP server, and you
don't need to manually do this. But if you want to run the MCP server manually,
for instance because you want to serve over HTTP (using SSE):
Given you have your PATH environment configured to include the path named by
$GOBIN (or $GOPATH/bin $HOME/go/bin if $GOBIN is not set), you can then
run:
mcp-cbs-cijfers-open-data --stdio=false --sse
Which will output startup logs with the SSE transport URL:
CBS API MCP server started, using transports: [sse]
SSE transport endpoint: http://localhost:8080
---
© 2025 David Stotijn — This project is licensed under the Apache License 2.0
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
