Visualization

A powerful Model Context Protocol (MCP) server for creating interactive data visualizations using matplotlib. This server provides comprehensive visualization tools including relationship graphs, scatter plots, 3D visualizations, and more.

Features

• 🔗 Relationship Graphs: Create network diagrams to visualize connections between entities
• 📊 Scatter Plots: Basic and classification scatter plots with customizable styling
• 🌐 3D Visualizations: Support for 3D scatter plots, surface plots, and wireframes
• 📈 Statistical Charts: Histograms, line plots, and heatmaps
• 💾 Auto-save: Automatically saves high-resolution images to temporary directory
• 🖼️ Live Display: Shows plots in interactive matplotlib windows
• 🎨 Customizable: Extensive styling options for colors, labels, and layouts

Installation

Prerequisites

$ python3 -m venv .venv --upgrade-deps 

$ source .venv/bin/activate

$ pip install "mcp[cli]" httpx

$ pip install   matplotlib numpy pandas networkx

Clone and Run

git clone https://github.com/xlisp/visualization-mcp-server.git
cd visualization-mcp-server
python visualization_server.py

Usage

1. Cladue Desktop Client Config

    "visualization": {
      "command": "/Users/clojure/Desktop/visualization-mcp-server/.venv/bin/python",
      "args": [
        "/Users/clojure/Desktop/visualization-mcp-server/visualization_server.py"
      ]
    }

The server provides several visualization tools that can be called via MCP protocol:

1. Relationship Graph

Create network diagrams to show relationships between entities:

# Example: Show relationships between A, B, C, D
nodes = ["Alice", "Bob", "Charlie", "Diana"]
edges = [["Alice", "Bob"], ["Bob", "Charlie"], ["Alice", "Charlie"], ["Charlie", "Diana"]]

# Call: create_relationship_graph(nodes, edges, "Social Network")

Parameters:

• nodes: List of node names
• edges: List of connections (pairs of node names)
• title: Graph title (optional)
• node_size: Size of nodes (default: 1000)
• font_size: Label font size (default: 12)

2. Scatter Plot

Create basic scatter plots with optional labels and colors:

# Example: Basic scatter plot
x_data = [1, 2, 3, 4, 5]
y_data = [2, 5, 3, 8, 7]
labels = ["Point A", "Point B", "Point C", "Point D", "Point E"]

# Call: create_scatter_plot(x_data, y_data, labels, title="My Scatter Plot")

Parameters:

• x_data: X-axis values
• y_data: Y-axis values
• labels: Point labels (optional)
• colors: Point colors (optional)
• title, x_label, y_label: Chart labels
• size: Point size (default: 50)

3. Classification Scatter Plot

Visualize data points grouped by categories:

# Example: Classification visualization
x_data = [1, 2, 3, 4, 5, 6]
y_data = [2, 3, 1, 5, 4, 6]
categories = ["Type A", "Type A", "Type B", "Type B", "Type C", "Type C"]

# Call: create_classification_plot(x_data, y_data, categories)

Parameters:

• x_data: X-axis values
• y_data: Y-axis values
• categories: Category labels for each point
• title, x_label, y_label: Chart labels

4. 3D Visualization

Create 3D plots with multiple visualization types:

# Example: 3D scatter plot
x_data = [1, 2, 3, 4, 5]
y_data = [2, 4, 1, 5, 3]
z_data = [3, 1, 4, 2, 5]

# Call: create_3d_plot(x_data, y_data, z_data, plot_type="scatter")

Parameters:

• x_data, y_data, z_data: 3D coordinates
• plot_type: "scatter", "surface", or "wireframe"
• title, x_label, y_label, z_label: Chart labels

5. Additional Charts

Histogram

# create_histogram(data, bins=30, title="Distribution")

Line Plot

# create_line_plot(x_data, y_data, line_style="-", color="blue")

Heatmap

# create_heatmap(data_matrix, x_labels, y_labels, colormap="viridis")

Output

Each visualization function:

1. Displays the plot in an interactive matplotlib window
2. Saves a high-resolution PNG file to the system temporary directory
3. Returns the file path where the image was saved

Example output:

"Graph saved to: /tmp/relationship_graph_20250802_143022.png and displayed"

File Management

• Images are saved with timestamps to avoid conflicts
• Files are stored in the system temporary directory:
  • Windows: C:\Users\username\AppData\Local\Temp\
  • macOS/Linux: /tmp/
• High resolution (300 DPI) PNG format

MCP Integration

This server follows the Model Context Protocol standard and can be integrated with MCP-compatible clients. The server runs on stdio transport by default.

Server Configuration

if __name__ == "__main__":
    mcp.run(transport='stdio')

Examples

Social Network Analysis

# Visualize social connections
nodes = ["Alice", "Bob", "Charlie", "Diana", "Eve"]
relationships = [
    ["Alice", "Bob"], 
    ["Bob", "Charlie"], 
    ["Charlie", "Diana"],
    ["Alice", "Eve"], 
    ["Eve", "Diana"]
]
# Creates a network graph showing social connections

Data Classification

# Visualize machine learning results
features_x = [1.2, 2.3, 1.8, 3.1, 2.9, 1.5]
features_y = [2.1, 3.2, 1.9, 4.1, 3.8, 2.2]
predictions = ["Class A", "Class B", "Class A", "Class B", "Class B", "Class A"]
# Creates a classification scatter plot

Scientific Data

# 3D scientific visualization
x_coords = [0, 1, 2, 3, 4]
y_coords = [0, 1, 4, 9, 16]
z_coords = [0, 1, 8, 27, 64]
# Creates a 3D surface or scatter plot

Error Handling

The server includes comprehensive error handling:

• Invalid data format detection
• Missing parameter validation
• Matplotlib rendering error recovery
• File system permission checks

Contributing

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

License

This project is licensed under the MIT License - see the LICENSE file for details.

Dependencies

• matplotlib: Core plotting library
• networkx: Graph/network visualization
• numpy: Numerical computations
• pandas: Data manipulation
• mcp: MCP server framework

Roadmap

• Interactive plot widgets
• Animation support
• Export to multiple formats (SVG, PDF)
• Custom color schemes
• Statistical analysis integration
• Real-time data streaming plots

Support

If you encounter any issues or have questions:

1. Check the Issues page
2. Create a new issue with detailed description
3. Include sample data and error messages

---

Made with ❤️ for the MCP ecosystem