MCP ECharts

The ECharts MCP Server is a lightweight, fully local solution that lets AI assistants such as Claude generate rich, interactive visualizations using Apache ECharts on demand. By exposing a Model Context Protocol interface, the server allows models to request chart configurations and receive rendered images or SVGs without any external rendering service. This eliminates latency, removes privacy concerns, and guarantees that every visual output is produced within the user’s own environment.

At its core, the server accepts a JSON payload describing an ECharts option object—data series, axes, themes, and styling. It validates the payload against the ECharts schema, ensuring that subsequent AI generations produce syntactically correct chart definitions. Once validated, the server renders the chart to a PNG or SVG file and returns either a Base64 string or a public URL when MinIO object storage is configured. The optional MinIO integration means that large visual assets can be stored efficiently, shared via links, and cached for repeated use, which is especially useful in collaborative data‑analysis workflows.

Developers benefit from the server’s seamless integration with existing MCP clients. Whether an AI assistant is running inside a desktop IDE, a chat application, or a web service, the MCP interface abstracts away transport details. The server supports standard , Server‑Sent Events (SSE), and a streamable protocol, giving teams flexibility to choose the most efficient channel for their environment. Because it is written in Node.js with zero external dependencies, adding the MCP server to a project requires only a single npm install and no heavyweight runtime.

Typical use cases include:

• Automated reporting – A model can produce dynamic dashboards from raw data, embedding charts directly into PDFs or HTML reports.
• Data‑driven conversations – In chat interfaces, the assistant can ask for clarification, generate a chart, and display it inline without leaving the conversation.
• Rapid prototyping – Front‑end developers can test visual layouts by feeding the server sample data and receiving instant renderings, speeding up iteration cycles.
• Secure internal analytics – Because all rendering happens locally, sensitive visualizations never leave the organization’s network.

The server’s standout features are its complete ECharts support, including theme switching and advanced visual options, and the validation layer that guarantees correct syntax across multiple AI turns. Combined with optional MinIO storage, the ECharts MCP Server delivers high‑performance, privacy‑preserving chart generation that fits naturally into any AI‑augmented development workflow.