mcp-vegalite-server

MCP.Pizza Chef: isaacwasserman

Two tools: one stores a table your assistant has already worked out, the other draws it as a picture using the Vega-Lite chart grammar. It reads nothing on its own — no files, no database — so the numbers have to come from the conversation. Be careful: the chart description is run through Python's eval, so text from your conversation is executed as code on your machine. It also refuses to start unless a folder named logs already exists, and the launch command in the README is wrong. Untouched since May 2025, with no licence file.

Unmaintained · No commits in 15 months.
Data
Images/Design

Use This MCP server To

Chart the sales figures I just pasted into the chat Turn a summary table into a bar chart image Compare two columns of numbers as a scatter plot Save a small table now and chart it later

README

MseeP.ai Security Assessment Badge

Data Visualization MCP Server

smithery badge

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"
        ]
    }
  }
}

mcp-vegalite-server FAQ

Will it work straight out of the box?
Probably not. It stops on its very first line unless a folder called logs already exists wherever it was launched from, and the startup option shown in the README is spelled with an underscore while the code only accepts a hyphen. Expect to fix both by hand.
Is there a safety concern?
Yes, a serious one. The chart description is handed straight to Python's eval, which executes it as code. Anything your assistant puts in that field — including text it copied off a web page — runs on your computer with your permissions.
Can I use this at work?
Not safely. The repository carries no licence file at all, so no permission to use it has been granted, and the code-execution problem above makes it a poor fit for a work machine.
Does it read my spreadsheet or database?
No. It only charts numbers that have already been typed into it during the conversation. You or your assistant must supply every value first.
Do I need a key?
No key, no account, no sign-up. Everything runs on your own machine using a local chart-rendering library.
Which apps does it work in?
It was written for Claude Desktop and is listed on Smithery, but the Smithery launch command repeats the same misspelled option, so that route fails in the same way.
Is it still maintained?
No. There have been no code changes since May 2025, and the launch instructions have been wrong that entire time.
What does it actually give me?
Either a PNG picture of the chart shown in the conversation, or the chart definition as text, depending on which mode it was started in.