MCP Plots

Plots MCP Server

A Model Context Protocol (MCP) server for data visualization. It exposes tools to render charts (line, bar, pie, scatter, heatmap, etc.) from data and returns the plot as image/base64 text/mermaid diagram.

Why MCP Plots?

• Instant, visual-first charts using Mermaid (renders directly in MCP clients like Cursor)
• Simple prompts to generate charts from plain data
• Zero-setup options via uvx, or install from PyPI/Docker
• Flexible output formats: mermaid (default), PNG image, or text

Quick Usage

• Ask your MCP client: "Create a bar chart showing sales: A=100, B=150, C=80"
• Default output is Mermaid, so diagrams render instantly in Cursor

Quick Start

PyPI Installation (Recommended)

pip install mcp-plots

mcp-plots # Start the server

For Cursor Users

1. Install the package: pip install mcp-plots
2. Add to your Cursor MCP config (~/.cursor/mcp.json):

   {

"mcpServers": { "plots": { "command": "mcp-plots", "args": ["--transport", "stdio"] } } }

   Alternative (zero-install via uvx + PyPI):

   {

"mcpServers": { "plots": { "command": "uvx", "args": ["mcp-plots", "--transport", "stdio"] } } }

3. Restart Cursor

1. Ask: *"Create a bar chart showing sales: A=100, B=150, C=80"*

Development Installation

uvx --from git+https://github.com/mr901/mcp-plots.git run-server.py

[Documentation →](docs/README.md) | [Quick Start →](docs/quickstart.md) | [API Reference →](docs/api.md)

MCP Registry

This server is published under the MCP registry identifier io.github.MR901/mcp-plots. You can discover/verify it via the official registry API:

curl "https://registry.modelcontextprotocol.io/v0/servers?search=io.github.MR901/mcp-plots"

Registry metadata for this project is tracked in server.json.

Install with Smithery

This repository includes a smithery.yaml for easy setup with Smithery.

• File: smithery.yaml
• Docs: https://smithery.ai/docs/config#smitheryyaml

Example install using the Smithery CLI (adjust --client as needed, e.g. cursor, claude):

npx -y @smithery/cli install \

https://raw.githubusercontent.com/mr901/mcp-plots/main/smithery.yaml \ --client cursor

After installation, your MCP client should be able to start the server over stdio using the command defined in smithery.yaml.

Project layout

src/

app/ # Server construction and runtime server.py capabilities/ # MCP tools and prompts tools.py prompts.py visualization/ # Plotting engines and configurations chart_config.py generator.py

Requirements

• Python 3.10+
• See requirements.txt

Setup Routes

uvx (Recommended)

The easiest way to run the MCP server without managing Python environments:

# Run directly with uvx (no installation needed)

uvx --from git+https://github.com/mr901/mcp-plots.git run-server.py

Or install and run the command

uvx --from git+https://github.com/mr901/mcp-plots.git mcp-plots

With custom options

uvx --from git+https://github.com/mr901/mcp-plots.git mcp-plots --port 8080 --log-level DEBUG

Why uvx?

• No Environment Management: Automatically handles Python dependencies
• Isolated Execution: Runs in its own virtual environment
• Always Latest: Pulls fresh code from repository
• Zero Setup: Works immediately without pip install
• Cross-Platform: Same command works on Windows, macOS, Linux

PyPI (Traditional Installation)

1. Install dependencies

pip install -r requirements.txt

1. Run the server (HTTP transport, default port 8000)

python -m src --transport streamable-http --host 0.0.0.0 --port 8000 --log-level INFO

1. Run with stdio (for MCP clients that spawn processes)

python -m src --transport stdio

Local Development (from source)

git clone https://github.com/mr901/mcp-plots.git

cd mcp-plots pip install -e . python -m src --transport stdio --log-level DEBUG

Docker

docker build -t mcp-plots .

docker run -p 8000:8000 mcp-plots

Environment variables (optional):

• MCP_TRANSPORT (streamable-http|stdio)
• MCP_HOST (default 0.0.0.0)
• MCP_PORT (default 8000)
• LOG_LEVEL (default INFO)

Tools

• list_chart_types() → returns available chart types
• list_themes() → returns available themes
• suggest_fields(sample_rows) → suggests field roles based on data samples
• render_chart(chart_type, data, field_map, config_overrides?, options?, output_format?) → returns MCP content
• generate_test_image() → generates a test image (red circle) to verify MCP image support

Cursor Integration

This MCP server is fully compatible with Cursor's image support! When you use the render_chart tool:

• Charts appear directly in chat - No need to save files or open separate windows
• AI can analyze your charts - Vision-enabled models can discuss and interpret your visualizations
• Perfect MCP format - Uses the exact base64 PNG format that Cursor expects

The server returns images in the MCP format Cursor requires:

{

"content": [ { "type": "image", "data": " ", "mimeType": "image/png" } ] }

Example call (pseudo):

render_chart(

chart_type="bar", data=[{"category":"A","value":10},{"category":"B","value":20}], field_map={"category_field":"category","value_field":"value"}, config_overrides={"title":"Example Bar","width":800,"height":600,"output_format":"MCP_IMAGE"} )

Return shape (PNG):

{

"status": "success", "content": [{"type":"image","data":" ","mimeType":"image/png"}] }

Configuration

The server can be configured via environment variables or command line arguments:

Server Settings

• MCP_TRANSPORT - Transport type: streamable-http or stdio (default: streamable-http)
• MCP_HOST - Host address (default: 0.0.0.0)
• MCP_PORT - Port number (default: 8000)
• LOG_LEVEL - Logging level: DEBUG, INFO, WARNING, ERROR, CRITICAL (default: INFO)
• MCP_DEBUG - Enable debug mode: true or false (default: false)

Chart Settings

• CHART_DEFAULT_WIDTH - Default chart width in pixels (default: 800)
• CHART_DEFAULT_HEIGHT - Default chart height in pixels (default: 600)
• CHART_DEFAULT_DPI - Default chart DPI (default: 100)
• CHART_MAX_DATA_POINTS - Maximum data points per chart (default: 10000)

Command Line Usage

With uvx (recommended):

uvx --from git+https://github.com/mr901/mcp-plots.git mcp-plots --help

Examples:

uvx --from git+https://github.com/mr901/mcp-plots.git mcp-plots --port 8080 --log-level DEBUG uvx --from git+https://github.com/mr901/mcp-plots.git mcp-plots --chart-width 1200 --chart-height 800

Traditional Python:

python -m src --help

Examples:

python -m src --transport streamable-http --host 0.0.0.0 --port 8000 python -m src --log-level DEBUG --chart-width 1200 --chart-height 800

Docker

Build image:

docker build -t mcp-plots .

Run container with custom configuration:

docker run --rm -p 8000:8000 \

-e MCP_TRANSPORT=streamable-http \ -e MCP_HOST=0.0.0.0 \ -e MCP_PORT=8000 \ -e LOG_LEVEL=INFO \ -e CHART_DEFAULT_WIDTH=1000 \ -e CHART_DEFAULT_HEIGHT=700 \ -e CHART_DEFAULT_DPI=150 \ -e CHART_MAX_DATA_POINTS=5000 \ mcp-plots

Cursor MCP Integration

Quick Setup for Cursor

The Plots MCP Server is designed to work seamlessly with Cursor's MCP support. Here's how to integrate it:

1. Add to Cursor's MCP Configuration

Add this to your Cursor MCP configuration file (~/.cursor/mcp.json or similar):

{

"mcpServers": { "plots": { "command": "uvx", "args": [ "--from", "git+https://github.com/mr901/mcp-plots.git@main", "mcp-plots", "--transport", "stdio" ], "env": { "LOG_LEVEL": "INFO", "CHART_DEFAULT_WIDTH": "800", "CHART_DEFAULT_HEIGHT": "600" } } } }

2. Alternative: HTTP Transport

For HTTP-based integration:

{

"mcpServers": { "plots-http": { "command": "uvx", "args": [ "--from", "git+https://github.com/mr901/mcp-plots.git@main", "mcp-plots", "--transport", "streamable-http", "--host", "127.0.0.1", "--port", "8000" ] } } }

3. Local Development Setup

For local development (if you have the code cloned):

{

"mcpServers": { "plots-dev": { "command": "python", "args": ["-m", "src", "--transport", "stdio"], "cwd": "/path/to/mcp-plots", "env": { "LOG_LEVEL": "DEBUG" } } } }

4. Verify Integration

After adding the configuration:

1. Restart Cursor
2. Check MCP connection in Cursor's MCP panel
3. Test with a simple chart:

   Create a bar chart showing sales data: A=100, B=150, C=80

MERMAID-First Approach

This server prioritizes MERMAID output by default because:

• ✅ Renders instantly in Cursor - No external viewers needed
• ✅ Interactive - Cursor can analyze and discuss the diagrams
• ✅ Lightweight - Fast generation and display
• ✅ Scalable - Vector-based, works at any zoom level

Chart Types with Native MERMAID Support:

• line, bar, pie, area → xychart-beta format
• histogram → xychart-beta with automatic binning
• funnel → Styled flowchart with color gradients
• gauge → Flowchart with color-coded value indicators
• sankey → Flow diagrams with source/target styling

Available Tools

render_chart

Main chart generation tool with MERMAID-first approach.

Parameters:

• chart_type - Chart type (line, bar, pie, scatter, heatmap, etc.)
• data - List of data objects
• field_map - Field mappings (x_field, y_field, category_field, etc.)
• config_overrides - Chart configuration overrides
• output_format - Output format (mermaid [default], mcp_image, mcp_text)

Special Modes:

• chart_type="help" - Show available chart types and themes
• chart_type="suggest" - Analyze data and suggest field mappings

configure_preferences

Interactive configuration tool for setting user preferences.

Parameters:

• output_format - Default output format (mermaid, mcp_image, mcp_text)
• theme - Default theme (default, dark, seaborn, minimal)
• chart_width - Default chart width in pixels
• chart_height - Default chart height in pixels
• reset_to_defaults - Reset all preferences to system defaults

Features:

• Persistent Settings - Saved to ~/.plots_mcp_config.json
• Live Preview - Shows sample chart with current settings
• Override Support - Use config_overrides for one-off changes

Do