Skip to content
MCP Server Directory logo
media-content

Chartpane

MCP App that renders interactive Chart.js charts and dashboards inline in AI conversations, including bar, line, area, pie, scatter, and radar charts.

Reviewed overview

MCP 服务器概览

What is ChartPane? ChartPane is an MCP (Model Connector Protocol) app that renders interactive Chart.js charts and multi-chart dashboards inline inside AI conversations (Claude, ChatGPT, VS Code, Cursor, etc.). It supports common chart types and client-side rendering so chart data is not stored on the server.

How to use ChartPane?

  • Add the MCP endpoint (https://mcp.chartpane.com/mcp) as a custom connector in an MCP-compatible client (Settings > Connectors > Add custom connector).
  • Or run via mcp-remote (Node.js) by configuring mcpServers to point to the ChartPane endpoint.
  • From the AI assistant, call the provided tools: render_chart for a single chart, render_dashboard for multi-chart grids. Provide chart type, title, labels and datasets in the JSON structure shown in the examples.

Key features

  • 9 chart types: bar, line, area, pie, doughnut, polarArea, bubble, scatter, radar (plus stacked/horizontal variants)
  • Multi-chart dashboard grids (up to 4 columns)
  • Custom colors or automatic 12-color palette
  • Client-side rendering; chart data values are never stored server-side
  • Works with any MCP-compatible client (Claude Desktop, ChatGPT, VS Code, Cursor)
  • Live demo and self-hosting support (Cloudflare Workers, KV/D1)

Use cases

  1. Embed interactive visualizations directly in AI-driven conversations and reports.
  2. Rapid exploratory data visualization for analytics, product metrics, or presentations.
  3. Dashboards combining multiple charts side-by-side for comparative analysis.
  4. Prototyping chart outputs for product or data teams without building a UI.

FAQ

  • Which clients are supported?

Any client that supports MCP Apps (Claude Desktop, ChatGPT with MCP, VS Code, Cursor).

  • Is my data stored on the server?

No. Chart data is rendered client-side and data values are never persisted server-side; only minimal request metadata is logged.

  • How do I self-host?

ChartPane runs on Cloudflare Workers. Clone the repo, configure secrets/KV/D1, then npm run deploy (see README for wrangler.jsonc notes).

  • How do I render multiple charts together?

Use render_dashboard with an array of chart configs and a columns setting to control grid layout.

  • What license and support options exist?

MIT license. Support via GitHub Issues and support@chartpane.com.