Creating Dashboards Using Vizro MCP: Vizro is an Open-Source Python Toolkit by McKinsey

Vizro is an open-source Python toolkit by McKinsey that makes it easy to build beautiful, production-ready data visualization apps. With just a few lines of configuration (via JSON, YAML, or Python dictionaries), you can create multi-page dashboards that would normally take thousands of lines of code.

Built on top of Plotly, Dash, and Pydantic, Vizro combines the flexibility of open source with in-built best practices for design and scalability. It’s quick to learn, customizable for advanced users, and powerful enough to move from prototype to production seamlessly.

In this tutorial, we’ll use the Vizro MCP server to create a dashboard directly from Claude Desktop.

Setting up the dependencies

uv package manager

To run the Vizro server, we will need the uv package manager. Install it with the following commands:

For Mac/Linux:

curl -LsSf https://astral.sh/uv/install.sh | sh

For Windows:

powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

Once uv is downloaded, run the following command to get the location of uvx

For Mac/Linux:

For Windows:

Keep the location of uvx handy, we will need it for the config file for Claude

Claude Desktop

You can download Claude Desktop from https://claude.ai/download. Next, open the claude_desktop_config.json file located in the Claude installation directory using any text editor. If the file doesn’t 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"
      ]
    }
  }
}

Replace the placeholder-path with the path of uvx

Running the Server

Once the file is configured, Vizro MCP Server should be visible in the list of servers.

Vizro comes with some sample datasets as well. You can try the following prompt to get started:

“create a vizro dashboard using tips dataset”

Claude will use the vizro-mcp to generate the dashboard and open it in your browser via PyCafe, showcasing interactive charts like tip vs total bill, average tips by day, tip distribution by gender, and tips by party size, along with filters for day, gender, and smoker status for seamless cross-filtering analysis.

I am a Civil Engineering Graduate (2022) from Jamia Millia Islamia, New Delhi, and I have a keen interest in Data Science, especially Neural Networks and their application in various areas.