MacMind

AI-powered Activity Monitor for macOS. Detects resource spikes, analyzes them using local LLM, and offers one-click remediation.

Features

• Real-time Monitoring: Native libproc integration for efficient CPU/memory tracking
• Smart Anomaly Detection: MAD-based statistical detection + static thresholds
• LLM Analysis: Ollama integration for intelligent diagnostics
• Menu Bar UI: SwiftUI-based, minimal footprint
• Safe Actions: Kill/throttle processes with protection for system processes

Requirements

• macOS 13.0+
• Swift 5.9+
• Ollama (optional, for AI analysis)

Quick Start

# Build
swift build

# Run
swift run MacMind

# Test
swift test

With Ollama (optional)

# Install Ollama
brew install ollama

# Pull model
ollama pull llama3.2:1b

# Start Ollama server
ollama serve

# Run MacMind
swift run MacMind

Architecture

Monitor (libproc) → Detector (MAD + thresholds) → Analyzer (Ollama) → UI (SwiftUI)
                                                         ↓
                                                  ActionExecutor

Project Structure

mac-mind/
├── Sources/MacMind/
│   ├── App/           # Main app + ViewModel
│   ├── Monitor/       # Process monitoring (libproc)
│   ├── Detector/      # Anomaly detection (MAD)
│   ├── Analyzer/      # LLM integration (Ollama)
│   ├── Actions/       # Process control
│   └── UI/            # SwiftUI views
├── Tests/
└── docs/              # Specs and research

License

MIT