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 saveddata(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 visualizedvegalite_specification(string): JSON string representing the Vega-Lite specification
- Returns: If the
--output_typeis set totext, returns a success message with an additionalartifactkey containing the complete Vega-Lite specification with data. If the--output_typeis set topng, returns a base64 encoded PNG image of the visualization using the MPCImageContentcontainer.
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_vegalite",
"--output-type",
"png" # or "text"
]
}
}
}
Usage with uv
uv --directory /Users/markomitranic/Sites/mcp/mcp-vegalite-server run mcp_server_vegalite --output-type png
Usage with Docker
docker build -t mcp-server-vegalite .
docker run -i --rm mcp-server-vegalite --output-type png