provchart-mcp
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@provchart-mcp Generate an SVG line chart of p50/p95 latency for a GitHub README."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
provchart-mcp
MCP server for ProvChart — generate pure-CSS charts (HTML/CSS or SVG) from any MCP-aware agent. No Chart.js, no client-side chart runtime required to paint.
What it does
Three tools, exposed over stdio:
Tool | Returns | Use for |
|
| Web pages, dashboards, anywhere you can inject markup |
|
| READMEs, docs, Markdown, static exports |
| Integration notes | No API call — quick reference for injection, ranges, runtime, errors |
Charts compile to real CSS (clip-path, custom properties) or real <svg> markup. Nothing runs client-side unless you opt into the optional provchart-runtime for hover tooltips and reveal animations.
Related MCP server: Charted
Install
npm install -g provchart-mcpOr run it directly without installing, via npx (see config below).
Setup
You'll need an API key from the ProvChart dashboard → Developer API.
Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"provchart": {
"command": "npx",
"args": ["-y", "provchart-mcp"],
"env": {
"PROVCHART_API_KEY": "pc_live_xxxxxxxx"
}
}
}
}Other MCP clients
Any client that supports stdio servers works the same way — point it at provchart-mcp (or node src/index.js if running from a local clone) and set PROVCHART_API_KEY in the environment.
You can also pass apiKey per-call as a tool argument instead of using the env var, if a given call needs to use a different key.
Usage
Once connected, just ask your agent for a chart:
Generate a line chart of monthly signups: Jan 120, Feb 450, Mar 3200, Apr 8900, May 15400. Give me the SVG for my README.
The agent calls provchart_generate_svg, gets back real SVG markup, and can write it straight to a file.
Chart types
line, area, bar, stackedbar, hbar, scatter, combo, gauge.
Value ranges
Charts auto-scale to your data by default — no need to pre-normalize anything to 0–100. Pass min / max if you want the axis pinned instead. Ask the server directly:
Use provchart_explain with topic "ranges"
Example payload
{
"type": "bar",
"series": [
{ "name": "2025", "color": "#8b7bff", "points": [40, 55, 48, 70] },
{ "name": "2026", "color": "#4fd8c4", "points": [35, 50, 62, 58] }
],
"axisX": ["Q1", "Q2", "Q3", "Q4"],
"theme": "midnight"
}Errors
Code | Meaning |
| Key missing, revoked, or malformed |
| Plan inactive |
| Plan quota used up for the month |
HTML and SVG generation share the same monthly quota.
Links
License
MIT
Available Tools
3 toolsprovchart_explainA
Return short integration instructions for ProvChart (inject HTML/CSS, SVG usage, runtime, ranges). No API call.
Inputs: topic.
provchart_generateA
Generate a ProvChart pure CSS chart (HTML + CSS). No Chart.js. Inject css into a tag and html into a container. Prefer this for web pages and dashboards.
Inputs: type (required), series (required), max, min, +12 more.
provchart_generate_svgA
Generate a ProvChart SVG chart for README/docs/Markdown. Returns svg markup and optional dataUri. Prefer committing .svg files over huge data URIs on GitHub.
Inputs: type (required), series (required), max, min, +12 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 3 tools
Each tool has a clearly distinct purpose: HTML/CSS chart generation, SVG chart generation, and usage instructions. Even though generate and generate_svg are similarly named, their output formats and target use cases are explicitly different.
All tool names use a consistent provchart_ prefix followed by a simple snake_case verb or verb_noun pattern. The naming scheme is predictable and easy to navigate.
Three tools is appropriately minimal for a focused chart-generation server. Each tool covers a distinct need and none feel redundant or unnecessary.
The server covers both major ProvChart output formats (HTML/CSS for web and SVG for docs) plus integration guidance. For its stated scope, there are no obvious missing operations or dead ends.
Maintenance
Related MCP Connectors
Create, inspect, manage, and render charts and data visualizations as SVG/PNG or interactive embeds.
Diagrams, badges, charts and QR codes as plain image URLs you can paste into Markdown.
Generate production-ready chart code (Recharts, Chart.js, ECharts, Plotly) from a prompt.
Renders interactive Chart.js charts and dashboards inline in AI conversations.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables AI agents to generate safe, runnable HTML chart pages from structured JSON data using Apache ECharts. It provides tools for chart type recommendation, page generation, validation, and patching with controlled, deterministic output.6 npmMIT
- AlicenseAqualityCmaintenanceZero-dependency charts for Python and AI agents. SVG and PNG, 15 chart types, a one-line MCP server, and a Claude skill.4294 PyPI17MIT
- AlicenseNot gradedqualityNot gradedmaintenanceEnables AI agents to generate beautiful, presentation-ready charts (SVG + PNG) with zero setup, supporting various chart types and styling options.25 npmMIT
- AlicenseAqualityBmaintenanceRenders word-sized charts — sparklines, bars, bullets and 100+ more types — to self-contained SVG with generated alt text, small enough to sit inline in a model's reply or a table cell. Three tools: find the right chart for a question, get its wiring, and render it.33,919 npm209MIT