Skip to content

Summary

Official AntV MCP server that turns a description of your data into a rendered chart — 26 visualization types from bar and line to sankey, mind-map, fishbone and geographic maps.

Features

  • 26 chart types including sankey, treemap, radar, violin and Venn diagrams
  • Diagramming tools: flowchart, mind-map, fishbone and organization charts
  • Geographic maps (district, pin and path) rendered via AMap
  • No API key required — free default rendering service
  • Private deployment via VIS_REQUEST_SERVER for confidential data
  • DISABLED_TOOLS trims the tool list for smaller models
  • stdio, SSE and streamable HTTP transports; Docker image available

Installation

Set up this MCP server in your favorite AI agent — copy a ready-made configuration below.

Any MCP-compatible agent

Most agents (Claude, Cursor, Windsurf, VS Code, and more) read a standard mcpServers configuration.

  1. Open your agent's MCP configuration file.
  2. Merge the snippet below into it.
  3. Restart the agent — the "AntV Chart MCP" tools will be available.
{
  "mcpServers": {
    "antv-chart-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@antv/mcp-server-chart"
      ]
    }
  }
}

Description

AntV Chart MCP is the official Model Context Protocol server from AntV, the data-visualization team at Ant Group whose G2, G6 and L7 libraries sit behind a large share of China's enterprise dashboards. It gives an assistant a vocabulary of charts rather than a drawing API: you describe the data and the question, and the server returns a rendered image.

The catalogue runs to 26 visualization types and covers considerably more than the usual bar/line/pie trio. Statistical charts include histogram, boxplot, violin, scatter, radar, funnel, treemap, sankey and Venn. Diagramming types — flowchart, mind-map, fishbone, organization chart and network graph — make it useful for structure as well as numbers. Geographic output (district-map, pin-map, path-map) is rendered through AMap and is therefore limited to China. A spreadsheet/pivot-table tool rounds out the set for tabular summaries.

Installation is a single npx -y @antv/mcp-server-chart, with a cmd /c variant on Windows and a Docker image for server deployments. Three environment variables cover the cases teams usually hit: VIS_REQUEST_SERVER repoints rendering at a private deployment when charts contain data that cannot leave your network, SERVICE_ID tags generation records for tracking, and DISABLED_TOOLS trims the tool list so a smaller model is not overwhelmed by 26 near-identical schemas. The server speaks stdio, SSE and streamable HTTP transports.

Written in TypeScript and MIT-licensed. No API key is required — the default public rendering service is free to use, which makes this one of the fastest MCP servers to get a useful result out of, though private deployment is the right call for confidential data.

Related MCP Servers

MCP: Headroom

by Headroom Labs

New

Context-compression MCP server that shrinks tool outputs, logs, files and RAG chunks before they reach the model, cutting 60-95% of tokens on JSON payloads.

MCP: Engram

by Gentleman Programming

New

Persistent memory for coding agents as a single Go binary — SQLite and FTS5 behind MCP, an HTTP API, a CLI and a TUI, with no Node, Python or Docker required.

MCP: agent-device

by Callstack

New

Give a coding agent a live feedback loop on real mobile, TV and desktop apps — inspect, tap, type and capture evidence on iOS, Android and HarmonyOS over MCP or a CLI.

New

The official W&B MCP server: query experiment runs, Weave LLM traces, artifacts and registries in natural language, and write findings back as a W&B report.

Browse all MCP servers →