Mcp Vegalite Server

Overview of Mcp Vegalite Server

What is Mcp Vegalite Server about?

Mcp Vegalite Server implements an MCP endpoint that lets large language models store tabular data and create visualizations using Vega‑Lite syntax. It exposes two primary tools – save_data for persisting a data table and visualize_data for rendering that table as a chart.

How to use Mcp Vegalite Server?

1. Add the server to Claude Desktop (or any MCP‑aware client) by editing claude_desktop_config.json:

   {
     "mcpServers": {
       "datavis": {
         "command": "uv",
         "args": [
           "--directory",
           "/absolute/path/to/mcp-datavis-server",
           "run",
           "mcp_server_datavis",
           "--output_type",
           "png"  // or "text"
         ]
       }
     }
   }
   

2. Call the save_data tool with a name and an array of objects representing rows.
3. Call the visualize_data tool with the previously saved data_name and a JSON string containing a Vega‑Lite specification.
4. Depending on the --output_type flag, the response will contain either a success message with the full spec (text) or a base64‑encoded PNG image (png).

Key Features

• Two simple tools: save_data and visualize_data.
• Flexible output: textual spec or PNG image.
• MCP‑native integration: works seamlessly with Claude Desktop and other MCP‑compatible clients.
• No external services required: all rendering occurs locally on the server.

Use Cases

• Rapid prototyping: quickly turn AI‑generated insights into charts without leaving the chat environment.
• Data reporting: store intermediate aggregation results and produce visual summaries on demand.
• Educational demos: illustrate how Vega‑Lite specifications translate to visual output for teaching data viz concepts.

FAQ

Q: What formats are accepted for the data field in save_data? A: An array of JSON objects, where each object represents a row and keys correspond to column names.

Q: Can I customize chart dimensions? A: Yes. Include width, height, and other Vega‑Lite properties in the vegalite_specification string.

Q: How do I retrieve the saved data later? A: You can call visualize_data with the same data_name; the server automatically fetches the stored table.

Q: Is there a way to get the raw data back? A: Not directly via the current tools; however, you can embed the data in the Vega‑Lite spec returned when using output_type: text.

Q: What dependencies are needed to run the server? A: The repository includes a pyproject.toml managed by uv. Running the command shown in the configuration snippet will install required packages.

Data Visualization MCP Server

Overview

A Model Context Protocol (MCP) server implementation that provides the LLM an interface for visualizing data using Vega-Lite syntax.

Components

Tools

The server offers two core tools:

• save_data
  • Save a table of data agregations to the server for later visualization
  • Input:
    • name (string): Name of the data table to be saved
    • data (array): Array of objects representing the data table
  • Returns: success message
• visualize_data
  • Visualize a table of data using Vega-Lite syntax
  • Input:
    • data_name (string): Name of the data table to be visualized
    • vegalite_specification (string): JSON string representing the Vega-Lite specification
  • Returns: If the --output_type is set to text, returns a success message with an additional artifact key containing the complete Vega-Lite specification with data. If the --output_type is set to png, returns a base64 encoded PNG image of the visualization using the MPC ImageContent container.

Usage with Claude Desktop

# Add the server to your claude_desktop_config.json
{
  "mcpServers": {
    "datavis": {
        "command": "uv",
        "args": [
            "--directory",
            "/absolute/path/to/mcp-datavis-server",
            "run",
            "mcp_server_datavis",
            "--output_type",
            "png" # or "text"
        ]
    }
  }
}