render_chartLABSCO SUMMARY
Two tools, and the second one is the interesting half.
A single chart is a nice touch. A dashboard — several charts arranged in a grid from one call — is what turns a set of numbers you were discussing into something you can actually read across. Because rendering is client-side, this is also one of the few visualisation options where the values in your chart never sit on someone else's server.
What it is
A hosted MCP app that draws charts inline in the client's UI. The server validates the input and returns structured content; the drawing happens in your browser, so the numbers never live anywhere but the request.
What you get
- render_chart — bar, line, area, pie, doughnut, polar area, bubble, scatter and radar
- Stacked and horizontal bar variants of the same tool
- render_dashboard — several charts laid out in a grid, up to four columns
- Your own colours, or an automatic palette when you do not care
- Rendering entirely on the client, with only request metadata logged
Requirements
Nothing to install: add the hosted endpoint as a custom connector, or bridge to it with mcp-remote if your client needs stdio. It runs on Cloudflare Workers and can be self-hosted, which means creating your own key-value namespace and database.
Setup effort
Paste a URL, then authorize — add the endpoint to your client, then approve the OAuth consent screen
USE CASES
01
Seeing the shape of numbers you have been discussing
Ask for the chart type that fits and get an interactive one in the conversation.
02
Comparing several measures at once
Lay them out as a grid rather than asking for one chart at a time.
03
Showing composition rather than change
Use a pie or doughnut, with each slice coloured from the built-in palette.
04
Matching a chart to a brand
Pass your own colours instead of taking the defaults.
05
Charting figures you would rather not upload
Render client-side, where only the request metadata is logged.
2 tools
Listed in project docs. Your client may show a different list.
render_dashboardBEFORE YOU INSTALL
What to know first
Tools you get
2 tools named in the project's documentation.
Requires
npx on PATH
Where it runs
The documented client connects to a hosted endpoint.
INSTALLATION
Connect your client
Add the endpoint to your client, then approve the OAuth consent screen.
Server setup
ENDPOINT URL
https://mcp.chartpane.com/mcpMCP Clients
Claude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"chartpane": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://mcp.chartpane.com/mcp"
]
}
}
}
