SchreiverJ/cookie-cutter-mcp

🍪 Cookie Cutter MCP Server

An advanced MCP (Model Context Protocol) server that transforms images into 3D printable cookie cutters using computer vision and computational geometry.

🌟 Features

🔍 Advanced Image Processing

• Multi-algorithm thresholding: Adaptive, Otsu, and manual threshold methods
• Original algorithm: Exact replication of the proven Cookie Cutter Monster algorithm
• Robust contour detection: Handles complex images with multiple objects
• Smart shape selection: Automatically picks the best contour for cookie cutters

🎯 Intelligent Shape Analysis

• Quality scoring: Evaluates contours by size, compactness, and position
• Boundary creation: "Draw and detect" method for perfect inner/outer boundaries
• Edge case handling: Gracefully processes challenging images

🛠️ Professional 3D Generation

• Watertight meshes: Uses Trimesh library for proper extrusion (like Three.js ExtrudeBufferGeometry)
• No geometry holes: Professional-quality STL files ready for 3D printing
• Configurable parameters: Control every aspect of cookie cutter design
• Handle integration: Automatic comfortable handle generation

🚀 Quick Start

Installation

# Clone and install
cd cookie-cutter-mcp
uv sync

# Or with pip
pip install -e .

Claude Desktop Setup

Add to your Claude Desktop MCP configuration:

{
  "mcpServers": {
    "cookie-cutter": {
      "command": "uv",
      "args": ["run", "python", "-m", "cookie_cutter_mcp.server"]
    }
  }
}

Basic Usage

Generate a cookie cutter from ~/images/logo.png

Analyze ~/complex-image.jpg to see what shapes can be detected

Create cookie cutters for all images in ~/designs/ using professional settings

🎮 Available Tools

1. analyze_image

Preview processing results and get recommendations before generating STL.

Example:

Analyze the image at ~/images/logo.png to see what cookie cutter shapes can be detected

Features:

• Detailed contour analysis with statistics
• Processing recommendations
• Quality assessment and suggestions

2. generate_cookie_cutter

Create a 3D printable cookie cutter STL file.

Example:

Generate a cookie cutter from ~/images/logo.png and save it as ~/3d-prints/logo-cutter.stl with professional quality settings

Configuration options:

• thickness: Wall thickness (0.5-3.0mm)
• depth: Cutting depth (8-25mm)
• size: Maximum dimension (20-150mm)
• bevel_cutter: Sharp beveled edge
• round_handle: Comfortable rounded handle
• tolerance: Detail level (0.05-1.0)

3. batch_generate_cutters

Process multiple images at once.

Example:

Generate cookie cutters for all images in ~/logos/ and save them to ~/cutters/

4. create_preset_config

Get optimized configurations for different use cases.

Example:

Show me the configuration preset for small detailed cookie cutters

Available presets:

• small_detailed: Intricate designs, fine details
• large_simple: Big shapes, fast printing
• professional: Balanced quality and speed
• miniature: Tiny detailed cutters
• custom: Your own specifications

🧠 Algorithm Overview

1. Image Processing (Exact Original Algorithm)

• Padding: 10px with BORDER_WRAP (not BORDER_CONSTANT)
• Thresholding: Adaptive with block_size=5, C=1 (exact original parameters)
• Contour Detection: RETR_LIST method for comprehensive shape finding
• Area Filtering: 10%-90% of image area to exclude noise and boundaries

2. Boundary Creation ("Draw and Detect" Method)

• 10x Scale Drawing: Draw contour as thick lines at 10x target size
• Dual Thresholding: Binary inversion + second threshold for clean boundaries
• Inner/Outer Extraction: Automatically find cutting boundaries
• Scale Back: Return to target dimensions with proper offsets

3. 3D Geometry Generation

• Trimesh Extrusion: Uses trimesh.creation.extrude_polygon() to extrude polygons

🎛️ Configuration Reference

🖨️ 3D Printing Guidelines

Recommended Settings

• Layer height: 0.15-0.2mm
• Infill: 15-20%
• Print speed: 40-60mm/s
• Supports: None needed
• Bed adhesion: Skirt

Material Recommendations

• PLA: Easy printing, food safe when certified, some can be dishwasher safe if annealed
• PETG: More durable, pptentially food safe

Post-Processing

• Light sanding of cutting edge for sharpness
• Clean with isopropyl alcohol before food use
• Store in dry location to prevent warping
• Try annealing to make your plastic dishwasher resistant, works with some PLAs

🐛 Troubleshooting

No Contours Detected or Too Many Small Contours

• Try preprocessing image for better contrast

🔬 Technical Details

Dependencies

• OpenCV: Image processing and contour detection
• Trimesh: Professional 3D mesh operations and STL export
• Shapely: Robust geometric operations
• NumPy: Numerical computations
• mapbox-earcut: Polygon triangulation

🤝 Contributing

This project demonstrates advanced computer vision and 3D geometry techniques. Key areas for enhancement:

• Additional visualization tools for debugging
• More sophisticated contour selection algorithms
• Advanced mesh optimization techniques
• Support for multi-part cookie cutters
• Integration with slicing software APIs

📜 License

MIT License - Feel free to use and modify for your projects!

🙏 Acknowledgments

Based on the original Cookie Cutter Monster algorithm, reimagined for modern Python tooling and MCP integration. https://github.com/SchreiverJ/CookieCutterMonster#