A macOS menu bar application designed to monitor AI coding assistant usage quotas, ClaudeBar provides users with a consolidated, at-a-glance view of their consumption across multiple services including Claude, Codex, Gemini, GitHub Copilot, and Antigravity. It aims to help developers stay within their AI API limits by offering real-time visual feedback and timely notifications, thereby preventing unexpected service interruptions or overages.

How It Works

ClaudeBar employs a layered architecture separating App, Domain, and Infrastructure concerns. Key design decisions include rich domain models encapsulating business logic, actor-based concurrency for thread-safe state management, and protocol-based dependency injection to enhance testability without relying on live services. The Infrastructure layer utilizes adapters for third-party CLI tools, ensuring these wrappers are excluded from code coverage. Notably, it bypasses a traditional ViewModel layer, allowing SwiftUI views to directly consume domain models, promoting a cleaner and more direct data flow.

Quick Start & Requirements

• Primary Install: Download the latest release DMG or ZIP from GitHub Releases and drag ClaudeBar.app to Applications. Alternatively, build from source using git clone https://github.com/tddworks/ClaudeBar.git and swift build -c release.
• Prerequisites: macOS 15+, Swift 6.2+, and installed CLI tools for providers to be monitored (e.g., claude, codex, gemini). GitHub Copilot requires credential configuration.
• Running: Execute swift run ClaudeBar or launch the application directly.
• Links: GitHub Releases (implied), docs/RELEASE_SETUP.md for detailed release setup.

Highlighted Details

• Supports monitoring of Claude, Codex, Gemini, GitHub Copilot, and Antigravity.
• Features real-time tracking of session, weekly, and model-specific usage percentages.
• Offers Light, Dark, and a festive, auto-enabled Christmas theme with snowfall animation.
• Provides visual status indicators (Green/Yellow/Red) and system notifications for quota warnings or critical states.
• Includes auto-refresh capabilities at configurable intervals and keyboard shortcuts for dashboard access and refresh.

Maintenance & Community

Releases are automated via GitHub Actions, triggered by version tags. Specific details on contributing new providers are available via an integrated TDD skill. No explicit community channels (Discord/Slack) or roadmap links are provided in the README.

Licensing & Compatibility

The project is licensed under the MIT License, permitting commercial use and integration into closed-source projects.

Limitations & Caveats

Requires macOS 15 or later. Monitoring specific AI providers necessitates the prior installation and configuration of their respective command-line interfaces. Automated release workflows require significant setup of GitHub Secrets, including Apple Developer certificates and App Store Connect API keys.