StatBar

A lightweight, native macOS menubar telemetry app that monitors live system hardware stats (CPU E/P-cores, GPU utilization & memory, RAM breakdown) alongside local LLM server metrics (llama-server, Ollama, LM Studio).

When your local LLM server is offline, the LLM telemetry segment is auto-hidden and StatBar operates as a pure system monitor with zero overhead.

⚡ Features

• Multi-Segment Status Bar: Compact menubar glyph displaying real-time CPU, GPU, Memory, and optional LLM metrics.
• System Telemetry Panel: Live breakdown of Efficiency vs. Performance cores, Metal GPU utilization & VRAM, and Active/Wired/Compressed/Available RAM with a multi-color segmented memory bar.
• LLM Server Intelligence: Auto-probes local LLM endpoints. Displays active model name, context token usage, prompt & generation speed (t/s), and total token counts.
• Native Apple Aesthetics: Uses native NSVisualEffectView materials (.popover), dynamically matching Light Mode, Dark Mode, Clear Glass, and Tinted Glass wallpapers.
• Zero Dock Footprint: Operates exclusively as a background status bar item (LSUIElement) with configurable refresh intervals and Launch-at-Login support (SMAppService).
• CLI Snapshot Flag: Run swift run StatBar --once to print a single telemetry JSON/snapshot directly to stdout.

🚀 Installation

Option 1: Via Homebrew (Recommended)

brew install coderhd/tap/statbar

Or tap first:

brew tap coderhd/tap
brew install statbar

Option 2: Via npm / npx

# Launch directly (installs to /Applications automatically)
npx statbar

# Or install globally
npm install -g statbar

Option 3: Build from Source

git clone https://github.com/coderhd/system-stats-menubar-app.git
cd system-stats-menubar-app
./scripts/build-app.sh

This compiles the release executable, assembles build/StatBar.app with icon assets, and installs it directly to /Applications/StatBar.app.

🗑️ Uninstallation

Via Homebrew

brew uninstall statbar

Via npm / npx

npx statbar uninstall
# or
npm uninstall -g statbar

🤖 Local LLM Server Setup

StatBar automatically detects local LLM server metrics. For full context & token speed tracking with llama-server, start the server with --metrics and --slots enabled:

llama serve --metrics --slots

StatBar connects to http://127.0.0.1:8080 by default. You can customize the server host, port, refresh rate, and optional API Key in StatBar Settings (accessible via right-click).

⚙️ Usage & Settings

• Left Click: Open/close the telemetry popover panel.
• Right Click: Open context menu to access Settings... or Quit StatBar.
• CLI Snapshot Mode:

  swift run StatBar --once

🛠️ Development

# Debug build
swift build

# Run unit tests
swift test

# Package release ZIP & SHA256
./scripts/package-release.sh

📄 License

StatBar is open-source software licensed under the MIT License.