Vega-Lite
Free使用 Vega-Lite 格式和渲染器从获取的数据生成可视化图表。
FreeFree tier
About Vega-Lite
A Model Context Protocol (MCP) server that provides an interface for visualizing data using Vega-Lite syntax. It offers two core tools: save_data to persist a table of data aggregations, and visualize_data to render the data as Vega-Lite specifications, supporting both text and PNG output via the MCP ImageContent container. Designed for integration with Claude Desktop and other MCP-compatible clients.
Key Features
save_data tool for persisting tabular data aggregations
visualize_data tool that generates Vega-Lite charts from saved data
Supports output as text (with Vega-Lite specification artifact) or PNG (base64 encoded image)
Uses Model Context Protocol (MCP) for LLM integration
Open source, written in Python (86%) with Dockerfile support
Easy configuration with Claude Desktop via claude_desktop_config.json
Pros & Cons
Pros
- Open source and free to use
- Leverages the powerful Vega-Lite grammar for rich visualizations
- Integrates seamlessly with Claude Desktop and other MCP clients
- Supports both human-readable text output and rendered PNG images
- Simple two-tool interface makes it easy for LLMs to use
Cons
- Limited to data that has been explicitly saved via the save_data tool
- Cannot edit or animate visualizations after creation
- Requires an MCP-compatible client (e.g., Claude Desktop) to operate
- Documentation is minimal beyond the README
Best For
Generating data visualizations from aggregated data using natural language via an LLM agentCreating Vega-Lite charts for dashboards and reports within an MCP ecosystemTeaching and prototyping data visualization with Vega-Lite syntax
FAQ
What is the mcp-vegalite-server?
It is an open-source MCP server that allows LLMs to visualize data using Vega-Lite syntax. It provides save_data and visualize_data tools.
How do I configure it with Claude Desktop?
Add the server to your claude_desktop_config.json with the command 'uv --directory /path/to/mcp-vegalite-server run mcp_server_vegalite --output_type png' (or text).
What output formats are supported?
Two output types: text (returns a success message with a Vega-Lite specification artifact) and png (returns a base64 encoded PNG image).