Quick Configuration
Choose the setup that matches your environment.
Before You Start
- Confirm you have the account, endpoint, or API key required for Chart.
- Start with minimum scopes and read-only access where possible.
- Keep secrets in environment variables instead of hardcoding them in JSON.
Last verified: 2026-06-13
Environment: Node.js 18+, @antv/mcp-server-chart over stdio (AntV chart generation)
Install command:
npx -y @antv/mcp-server-chartLocal config
JSON Local config json
{
"chart": {
"command": "npx",
"args": ["-y", "@antv/mcp-server-chart"]
}
}After copying
Check the Chart JSON before running it in OpenClaw, then verify the server with a real prompt.
Verify prompt
Generate a line chart of monthly revenue for Jan-Jun with sample values and return the image URL.
Expected success output
You should get a generated chart (image URL or data) for the requested chart type.
Common Pitfalls & Fixes
- ⚠️ Watch out: Data format requirements and chart type selection.
- 🔑 Always store API keys in environment variables, never hardcode them in JSON.
- 🛡️ Start with read-only scopes if available to verify connection safely.
Security checklist
- Chart rendering may call an external service by default; set VIS_REQUEST_SERVER to self-host if data is sensitive.
- Do not send confidential data into a public rendering endpoint.
Common errors
| Error | Cause | Fix |
|---|---|---|
Chart tools not found | The host was not restarted after adding the server. | Reload OpenClaw and confirm the generate_*_chart tools are listed. |
Request failed / network | The default render service is unreachable. | Check egress, or set VIS_REQUEST_SERVER to a self-hosted renderer. |
Invalid data | The chart data shape is wrong. | Give the model clean, typed data points matching the chart type. |
Example Prompts
Once connected, try these prompts to test capabilities:
- Generate a bar chart showing monthly sales data.
- Create a pie chart for market share distribution.
- Visualize time-series data as a line chart.
Verification Checklist
- Run 1-2 real prompts to confirm Chart returns usable data.
- Check that error messages are clear enough for troubleshooting.
- Document the required scopes, dependencies, and env vars for future reuse.