vscode-antigravity-cockpit
VS Code extension for monitoring Google Antigravity AI quotas. Features Webview dashboard, QuickPick mode, and quota grouping.
VS Code extension that shows your remaining Google Antigravity AI model quotas as a live dashboard with countdown timers, progress bars, and status bar alerts so you always know how much of each model limit is left.
Antigravity Cockpit is a Visual Studio Code extension for monitoring AI model usage quotas from Google Antigravity. When you use AI models, most services impose limits on how many requests you can make within a time window. This extension shows you how much of each model's quota remains so you can see at a glance which models are running low and when the limits reset.
The extension offers two viewing modes. The first is a Webview dashboard, a panel inside VS Code that displays each model as a card showing the remaining quota as a percentage, a countdown timer, and a progress bar. Models that share a quota pool can be grouped together, and you can drag cards to reorder them. The second mode is QuickPick, which uses VS Code's built-in text dropdown interface and works in environments where the visual panel cannot load, or for users who prefer keyboard navigation.
A status bar item sits at the bottom of the VS Code window and shows the quota state for whichever model you have pinned. You can choose from several display formats: just an icon, a colored dot, a percentage, or a combination. Multiple models can be pinned to the status bar simultaneously. Notifications fire when a model's quota drops below thresholds you configure, such as a yellow warning at 30% and a red alert at 10%.
An auto wake-up feature lets you schedule timed requests to AI models so they run before your daily quota resets, making use of otherwise idle time. It supports multiple accounts via an authorization flow and stores credentials in VS Code's encrypted secret storage.
The extension is available through the Open VSX marketplace and supports 16 languages, following VS Code's language setting automatically. The README is written in Chinese with an English version linked separately. The license is MIT, with a note that the project is intended for personal learning and research rather than commercial use.
Where it fits
- Pin your most-used AI model to the VS Code status bar to see remaining quota at a glance
- Set threshold alerts to get a warning when a model drops below 30% or a critical alert at 10%
- Schedule auto wake-up requests to run before your daily quota resets to use otherwise idle capacity
- Group models that share a quota pool together in the card dashboard and drag them to preferred order