SecureSlice v4 - SGX-Protected 3D Slicing Pipeline

A production-ready Gramine SGX project for secure 3D model slicing with remote attestation, encrypted model protection, and cryptographic deletion certificates.

Overview

SecureSlice v4 implements a complete secure slicing pipeline running Slic3r → gcode2png → POV-Ray entirely within a single SGX enclave. The system provides:

• Remote Attestation: DCAP-based attestation with fresh nonce binding
• Encrypted Models: AES-256-GCM encryption with ECDH-derived session keys
• Iterative Refinement: Preview-approve workflow with parameter adjustments
• Deletion Certificates: Cryptographic proof of secure deletion with SGX quotes
• Blockchain Logging: Immutable audit trail for deletion certificates

Quick Start (5 Minutes)

1. Setup

# Create working directories
mkdir -p host_io/in host_io/out host_io/work

# Copy test model
cp test_models/cube.stl host_io/in/model.stl

2. Configure Slicing Parameters

cat > host_io/in/params.json << 'PARAMS'
{
  "--layer-height": "0.2",
  "--fill-density": "20%",
  "orientation": {
    "rotate_x": 45,
    "rotate_y": 0,
    "rotate_z": 0,
    "scale": 1.0
  }
}
PARAMS

3. Run the Orchestrator

# Standalone mode (no SGX required for testing)
STANDALONE_MODE=1 python3 orchestrator.py

4. Review Preview (in another terminal)

# Check the generated preview
ls -la host_io/out/preview.png
# View with any image viewer

5. Approve or Iterate

# To approve and finalize:
echo '{"approve": true}' > host_io/in/control.json

# OR to reject and adjust:
echo '{"approve": false, "params": {"--layer-height": "0.15"}}' > host_io/in/control.json

6. Collect Results

ls -la host_io/out/
# final.gcode         - Ready for 3D printer
# preview.png         - Visual preview
# deletion_cert.json  - Cryptographic proof
# orchestrator.log    - Processing log

Complete Installation Guide

Prerequisites

System Requirements

• Ubuntu 22.04 LTS (or compatible Linux)
• Python 3.8 or newer
• Perl 5 (for gcode2png)
• Docker (optional but recommended)
• 4GB RAM minimum
• 10GB free disk space

For SGX Hardware Mode (Optional)

• Intel CPU with SGX support (6th gen Core or newer)
• SGX enabled in BIOS
• Intel SGX drivers installed
• DCAP libraries configured

Installation Steps

# 1. Clone the repository
git clone https://github.com/your-repo/SecureSlice-v4.git
cd SecureSlice-v4

# 2. Install system dependencies
sudo apt-get update
sudo apt-get install -y \
    python3 python3-pip \
    perl \
    imagemagick \
    build-essential

# 3. Install Python packages
pip3 install cryptography requests

# 4. Verify vendored binaries
ls -la vendor/
# Should show:
# - slic3r (20MB) - C++ slicer with rotation fix
# - gcode2png (11KB) - Perl visualization script  
# - povray (2.9MB) - Ray tracer

# 5. Test installation
./vendor/slic3r --help
perl vendor/gcode2png --help

Detailed Usage Examples

Example 1: Simple Cube Slicing

# Setup
cp test_models/cube.stl host_io/in/model.stl

# Basic parameters
echo '{
  "--layer-height": "0.2",
  "--fill-density": "20%"
}' > host_io/in/params.json

# Run
STANDALONE_MODE=1 python3 orchestrator.py &
sleep 10
echo '{"approve": true}' > host_io/in/control.json

# Result
ls -la host_io/out/final.gcode

Example 2: Testing Rotation

# Test different orientations
for angle in 0 45 90; do
    echo "Testing rotation: $angle degrees"
    
    # Set rotation
    echo "{
      \"orientation\": {
        \"rotate_x\": $angle,
        \"rotate_y\": 0,
        \"rotate_z\": 0
      }
    }" > host_io/in/params.json
    
    # Run pipeline
    rm -rf host_io/work/* host_io/out/*
    STANDALONE_MODE=1 timeout 30 python3 orchestrator.py &
    sleep 10
    
    # Save preview
    cp host_io/out/preview.png preview_${angle}deg.png
    echo '{"approve": true}' > host_io/in/control.json
    sleep 2
done

# Compare results
ls -la preview_*.png

Example 3: Iterative Refinement

# Start with coarse settings
cat > host_io/in/params.json << 'EOF'
{
  "--layer-height": "0.3",
  "--fill-density": "10%"
}