Using Vizro MCP: Vizro Create Dashboard is McKinsey’s open source Python toolkit

Vizro It is an open source Python toolkit McKinsey This makes it easy to build beautiful production data visualization applications. With just a few lines of configuration (via JSON, YAML, or Python dictionary), you can create multi-page dashboards that will usually take thousands of lines of code.

Built on top of Plotly, Dash and Pydantic, Vizro combines open source flexibility with best practices for design and scalability. It will be customizable to power users very quickly and is powerful enough to move from prototypes to seamless production.

In this tutorial, we will use Vizro MCP Server Create dashboards directly from Claude Desktop.

Set up dependencies

UV package manager

To run a Vizro server, we will need a UV package manager. Install it with the following command:

For Mac/Linux:

curl -LsSf  | sh

For Windows:

powershell -ExecutionPolicy ByPass -c "irm  | iex"

After downloading UV rays, run the following command to get the location of UVX

For Mac/Linux:

For Windows:

Keep UVX convenient locations we will need it as a configuration file for Claude

Claude Desktop

You can download Claude Desktop from next and open claude_desktop_config.json Use any text editor that is located in the Claude installation directory. If the file does not exist, you can create it manually. Once opened, enter the following code:

MAC/Linux:

{
  "mcpServers": {
    "vizro-mcp": {
      "command": "/placeholder-path/uvx",
      "args": [
        "vizro-mcp"
      ]
    }
  }
}

Windows:

{
  "mcpServers": {
    "vizro-mcp": {
      "command": "placeholder-path//uvx",
      "args": [
        "vizro-mcp"
      ]
    }
  }
}

Use UVX path instead of placeholder path

Run the server

Once the file is configured, the Vizro MCP server should be visible in the server list.

Vizro also comes with some sample datasets. You can try the following tips to get started:

“Create Vizro Dashboards with Tips Datasets”

Claude will use Vizro-MCP To generate a dashboard via Pycafe and open it in your browser, show interactive charts such as Tip vs Total Bill, average tips by daily bills, average tips by gender, and tips by party size, and daytime, gender, gender, and smoker status.

I am a civil engineering graduate in Islamic Islam in Jamia Milia New Delhi (2022) and I am very interested in data science, especially neural networks and their applications in various fields.