Check remaining Claude Code or OpenRouter quota without logging into each website.
Get notified before your AI service usage runs out so you are not caught off guard.
Monitor account balances across multiple AI providers in dollars or yuan.
Track both 5-hour and weekly usage windows for AI coding tools.
| mgks/aimonitor | aakarsh-goyal/macjuice | alan13367/opensnapx | |
|---|---|---|---|
| Stars | 1 | 1 | 1 |
| Language | Swift | Swift | Swift |
| Setup difficulty | easy | easy | moderate |
| Complexity | 2/5 | 1/5 | 3/5 |
| Audience | developer | general | developer |
Figures from each repo's GitHub metadata at analysis time.
Download a disk image and drag to Applications, then bypass a one-time macOS security warning since the app is not Apple Developer Program signed.
AIMonitor is a small macOS app that lives in your menu bar and tracks how much usage you have left across several AI services. Instead of logging into each provider's website to check whether you are about to hit a limit, you click one icon and see current quotas, remaining balances, and service health for all your configured providers at once. The app supports seven providers: Claude Code, Codex, Kimi, MiniMax, Z.ai (GLM), DeepSeek, and OpenRouter. Some providers use OAuth login through their official command line tools, so you run the provider's setup once and AIMonitor reads the credentials automatically, refreshing tokens when they expire. Others require an API key that you paste into the preferences screen. MiniMax and Z.ai offer a choice between international and China endpoints. The information displayed varies by provider: some show five hour and weekly usage windows with plan tier details, while others show account balances in dollars or yuan. A few practical details stand out. The app refreshes every 60 seconds by default, and you can change that interval. It sends notifications when quota drops below 20 percent, 10 percent, or when it runs out or resets. Each provider appears as a card with progress bars for both the five hour and weekly windows. Your API keys are stored in a local file on your machine rather than in the macOS Keychain, which means you will not see repeated system password prompts. Installing involves downloading a disk image file from the project's releases page and dragging the app to your Applications folder. Because the project is open source and not part of Apple's paid developer program, macOS will show a security warning on first launch. The README walks through a one time process of confirming you want to open the app anyway through System Settings. Developers can build from source using Swift and a Makefile, and the README documents how to add new providers. The project is MIT licensed with no analytics and no backend. The full README is longer than what was shown.
A macOS menu bar app that shows remaining usage and balances for multiple AI services like Claude, OpenRouter, and DeepSeek in one place.
Mainly Swift. The stack also includes Swift, macOS, Makefile.
Use freely for any purpose, including commercial use, as long as you keep the copyright notice.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
double-check against the repo, no cap.