Release Date: November 30, 2025 Created by: Jordan Koch
Version 8.0.0 introduces groundbreaking AI-powered features using Apple's MLX framework for on-device intelligence. This major release transforms NMAP Plus Security Scanner into an intelligent security assistant with natural language understanding, automated threat analysis, and comprehensive documentation generation.
AI-powered network security analysis that goes beyond traditional scanning.
Features:
- Comprehensive network threat assessment
- Device-specific vulnerability analysis
- Severity classification (Critical/High/Medium/Low)
- Contextual threat explanations
- Actionable remediation recommendations
- Real-time risk scoring
Use Case: Scan your network and receive AI-generated security analysis explaining exactly what threats exist and why they matter.
Automatically identifies and categorizes unknown devices with AI precision.
Features:
- Automatic device type identification
- Manufacturer and model detection
- Confidence scoring (High/Medium/Low)
- Suggested friendly device names
- Batch classification for multiple devices
- MAC address OUI analysis
- Classification caching for performance
Use Case: No more "Unknown Device" labels - AI identifies what each device is based on ports, services, and network behavior.
Ask questions about your network in plain English.
Features:
- Natural language network queries
- "Show me all IoT devices with open ports"
- "Which devices are offline?"
- "Find devices connected in the last 24 hours"
- Query history tracking
- Contextual device filtering
- Suggested query templates
Use Case: Instead of clicking through menus, simply ask what you want to know about your network.
AI-generated, prioritized security guidance tailored to your network.
Features:
- Comprehensive security assessments
- Priority-based recommendations (Critical → Low)
- Step-by-step implementation guides
- Impact assessment for each recommendation
- Device-specific security guidance
- Network-wide best practices
- Exportable security reports
Use Case: Get a personalized security improvement roadmap with actionable steps prioritized by impact.
AI-powered detection of unusual network behavior with explanations.
Features:
- Network baseline establishment
- New device detection with risk assessment
- Unusual port activity identification
- Missing device alerts
- Network change analysis
- Contextual anomaly explanations
- Severity-based anomaly classification
Use Case: AI learns your network's normal behavior and alerts you to anything suspicious with detailed explanations.
Chat with an AI security expert about your network.
Features:
- Real-time chat interface
- Network-aware responses
- Security best practices guidance
- NMAP and port scanning education
- Conversation history
- Suggested question templates
- Contextual device information
Use Case: Get instant answers to security questions like "How do I secure my IoT devices?" or "What does port 22 being open mean?"
Professional network documentation generated automatically by AI.
Features:
- Comprehensive network documentation
- Executive summaries
- Device inventory with specifications
- Security analysis sections
- Network topology descriptions
- Configurable documentation sections
- Multiple export formats (Markdown, HTML, Plain Text)
- Professional formatting
Use Case: Generate complete, professional network documentation in seconds for compliance, audits, or knowledge management.
Intelligent system capability detection with graceful degradation.
Features:
- Apple Silicon (M1/M2/M3/M4) detection
- MLX Python toolkit availability check
- Virtual environment support
- Status monitoring (Available/Degraded/Unavailable)
- One-click MLX installation
- Helpful error messages
Use Case: Ensures AI features only run on compatible hardware and provides clear guidance for setup.
Core AI engine powering all intelligent features.
Features:
- Phi-3.5-mini model support (2-3GB)
- On-device inference (no cloud required)
- Metal GPU acceleration
- Streaming and batch generation
- Temperature and token control
- Python process management
- Virtual environment integration
Technical: Uses Apple's MLX framework with mlx-lm for efficient on-device LLM inference.
- Apple Silicon Required: M1, M2, M3, or M4 chip
- RAM: 8GB minimum, 16GB+ recommended
- Storage: 4GB free space for MLX models
- macOS: 13.0 (Ventura) or later
- Python: 3.9+ with pip3
- MLX: Install with
pip3 install mlx mlx-lm
- Virtual environment at
/Volumes/Data/xcode/NMAPScanner/.venv - Pre-downloaded Phi-3.5-mini model at
~/.mlx/models/phi-3.5-mini
# Create virtual environment (recommended)
cd /Volumes/Data/xcode/NMAPScanner
python3 -m venv .venv
source .venv/bin/activate
# Install MLX
pip3 install mlx mlx-lm
# Verify installation
python3 -c "import mlx.core; import mlx_lm; print('MLX Ready!')"The Phi-3.5-mini model will be downloaded automatically on first AI feature usage (~2-3GB).
All AI features will be available in their respective tabs/sections.
- Model Size: 2-3GB (Phi-3.5-mini)
- Inference Speed: 10-30 tokens/second (M3 Ultra)
- Memory Usage: 4-6GB during inference
- First Run: 30-60 seconds (model loading)
- Subsequent Runs: < 5 seconds
- 100% On-Device: All AI processing happens locally on your Mac
- No Cloud Calls: Zero network requests to external AI services
- Private Data: Your network scan data never leaves your device
- Offline Capable: Works without internet connection
- Open Source Model: Uses Microsoft Phi-3.5-mini
If MLX is unavailable (Intel Mac, missing toolkit, etc.):
- ✅ Traditional Features: All existing NMAP scanning continues to work
⚠️ AI Features Disabled: AI features show helpful setup instructions- 📝 Clear Guidance: Step-by-step instructions for enabling AI features
- 🔌 Optional Install Button: One-click MLX installation (when possible)
None. Version 8.0.0 is fully backward compatible. All AI features are additive and don't modify existing functionality.
- Apple Silicon Only: MLX requires M1/M2/M3/M4 - Intel Macs not supported
- macOS Only: AI features not available on tvOS (scanner itself works)
- Model Download: First-time use requires ~3GB download
- Token Limits: Large networks (>30 devices) may be summarized
- Python Dependency: Requires Python 3.9+ with pip3
Planned for v8.1.0+:
- Voice-controlled network queries (Siri integration)
- Custom model support (Llama 3, Mistral, etc.)
- Fine-tuned security models
- Multi-language support
- Larger context windows for enterprise networks
- iOS/iPadOS support (when MLX available)
- See AI feature tooltips in-app
- Ask the Security Assistant for help
- Check suggested queries in Natural Language interface
MLXCapabilityDetector.swift- System detectionMLXInferenceEngine.swift- Core inference API- Each feature file contains inline documentation
Developed by: Jordan Koch AI Framework: Apple MLX AI Model: Microsoft Phi-3.5-mini Build Date: November 30, 2025 Version: 8.0.0 (Build 13)
For issues, feature requests, or questions:
- Issues: Report via project issue tracker
- Security: Report vulnerabilities privately
- Feedback: Share your experience with v8.0.0!
Download NMAP Plus Security Scanner v8.0.0 and experience the future of network security analysis powered by on-device AI.
Minimum Version: macOS 13.0, Apple Silicon Recommended: macOS 14.0+, M2/M3/M4, 16GB RAM
Transforming network security scanning with the power of AI.