{"slug": "antigravity-usage-intelligence-track-tokens-quotas-cache-reasoning-and-agent", "title": "Antigravity Usage Intelligence: Track Tokens, Quotas, Cache, Reasoning, and Agent Activity", "summary": "A developer built Antigravity Usage Intelligence, an open-source usage tracker and analytics dashboard for Google's Antigravity agentic development environment. The tool breaks down token consumption into fresh input, cached context, and reasoning tokens, tracks quota usage via Antigravity's local language server with local rolling-window fallbacks, and visualizes activity through heatmaps and model-level analytics. The developer says the goal is to make the behavior of AI development workflows visible rather than to serve as a billing calculator.", "body_md": "I’ve been using Antigravity heavily for development, and one thing kept bothering me:\n\n**I could see the agent doing a lot of work, but it was hard to understand how much was actually being used.**\n\nHow many tokens were going through my sessions?\n\nHow much of that was fresh input versus cached context?\n\nHow much reasoning was happening?\n\nWhich projects were using the most?\n\nHow often was I hitting the quota?\n\nAnd what did my usage actually look like over time?\n\nThat led me to build **Antigravity Usage Intelligence**, an open-source usage tracker and analytics dashboard designed around those questions.\n\nGitHub: [https://github.com/Nir-Bhay/antigravity-usage-intelligence](https://github.com/Nir-Bhay/antigravity-usage-intelligence)\n\nAntigravity has become much more agent-oriented. With Antigravity 2.0, Google introduced a standalone desktop experience alongside features such as subagents, scheduled tasks, agent management, CLI support, and other tools for longer and more autonomous workflows.\n\nThat changes the way we think about usage.\n\nA simple prompt counter is not enough anymore.\n\nAn agent can perform multiple turns, call tools, read files, generate output, use reasoning, and reuse cached context during a single task.\n\nThe official Antigravity documentation also explains that quota limits are tied to the amount of work performed by the agent, which can vary significantly from one task to another.\n\nThat makes visibility much more useful than simply knowing whether a request succeeded.\n\nThe dashboard breaks usage into several parts instead of showing one mysterious number.\n\nThe dashboard separates:\n\nThat distinction matters.\n\nFor example, two agent sessions can both look like large workloads, while their actual token behavior is very different because one relies heavily on cached context.\n\nThe goal is not to turn token usage into a billing calculator.\n\nThe goal is to make the behavior of your AI development workflow visible.\n\nOne of the most useful parts is quota visibility.\n\nThe extension can query Antigravity’s local language server and surface information such as:\n\nWhen live server data is unavailable, the project can fall back to local rolling-window calculations.\n\nThat distinction is important because **local estimates and server-reported quota are not the same thing**.\n\nI would treat the server-side value as the authoritative quota signal and the local calculations as a useful approximation for understanding recent activity.\n\nNumbers become much easier to understand when you can see them over time.\n\nThe activity heatmap gives a quick visual view of when you were actually using Antigravity.\n\nInstead of opening individual sessions, you can quickly spot:\n\nIt is basically a contribution graph for your agent activity.\n\nThat sounds small, but it changes the experience.\n\nAfter a few weeks, you start seeing how your development habits actually look.\n\nThe dashboard also shows usage over time.\n\nThis helps answer questions like:\n\n**Am I gradually using more AI every week?**\n\n**Which days have the highest token consumption?**\n\n**Did one project cause a sudden spike?**\n\n**Am I relying more on reasoning-heavy models?**\n\nThose are much more useful questions than simply asking how many messages you sent.\n\nAntigravity can work with multiple models, and the project identifies model activity so you can see which models are actually being used during your workflows.\n\nFor example, the current Antigravity model lineup includes Gemini and Claude reasoning models, with availability depending on the user's plan.\n\nThat makes model-level analytics particularly useful when you're comparing different workflows or trying to understand where your usage is going.\n\nOne thing I wanted to make visible was reasoning usage.\n\nWith thinking-oriented models, a task may involve significantly more internal reasoning than what you see in the final response.\n\nTreating all tokens as one number hides that difference.\n\nSeparating reasoning tokens gives you another lens into how your agent is working.\n\nIt is not necessarily about using fewer tokens.\n\nSometimes more reasoning is exactly what you want.\n\nThe point is to understand the tradeoff.\n\nWhen a task goes wrong, aggregated statistics are not enough.\n\nThe session inspector lets you drill down into individual sessions and look at things such as:\n\nTool calls are also categorized, so you can see the type of work the agent actually performed.\n\nThat makes it easier to investigate long-running or unexpectedly expensive sessions.\n\nAgent development is not just about model responses.\n\nA large amount of useful work happens through tools.\n\nReading files, searching code, running commands, replacing content, writing files, and other actions all contribute to the overall shape of a session.\n\nTracking tool requests gives another way to understand whether an agent was mostly answering questions or actually doing work inside a codebase.\n\nThis was important from the beginning.\n\nThe project is designed to operate locally.\n\nIt reads local Antigravity data and communicates with the local language server through loopback rather than sending your usage history to a remote analytics service.\n\nThe repository also documents read-only database access and local caching so that active sessions are not unnecessarily disturbed.\n\nThere are no external telemetry calls in the project.\n\nThat means you can inspect your usage without handing your development history to another analytics backend.\n\nThis is probably the most obvious question.\n\nAntigravity already provides quota information, especially through its current settings and model interfaces.\n\nThe difference is **depth**.\n\nA quota indicator tells you where you stand.\n\nAn analytics dashboard can help explain **why** you are there.\n\nThose are different problems.\n\nIf you are doing serious agent-based development, historical context becomes useful:\n\nWhich project consumed the most?\n\nWhich days were the heaviest?\n\nHow much context was cached?\n\nHow many tool calls did the agent make?\n\nWhich sessions were unusually large?\n\nHow does my usage change over time?\n\nThat is the gap this project is trying to fill.\n\nThe biggest lesson is that token usage is not a single number.\n\nA large token count can come from very different workflows.\n\nA session may have:\n\nLooking only at the total hides the interesting part.\n\nOnce the data is separated, you can start thinking about **how you work with agents**, not just how much you use them.\n\nI think it is most useful for developers who:\n\nIt is probably unnecessary if you only use Antigravity occasionally and never care about usage patterns.\n\nYou can install the extension from the VS Code-compatible marketplace or use the repository directly.\n\n```\nbash\ngit clone https://github.com/Nir-Bhay/antigravity-usage-intelligence.git\ncd antigravity-usage-intelligence\nnpm install\n```\n\n", "url": "https://wpnews.pro/news/antigravity-usage-intelligence-track-tokens-quotas-cache-reasoning-and-agent", "canonical_source": "https://dev.to/nirbhay_hiwse/antigravity-usage-intelligence-track-tokens-quotas-cache-reasoning-and-agent-activity-ca1", "published_at": "2026-09-12 06:53:15+00:00", "updated_at": "2026-09-12 07:26:34.704763+00:00", "lang": "en", "topics": ["ai-tools", "ai-agents", "developer-tools", "large-language-models", "ai-products"], "entities": ["Antigravity Usage Intelligence", "Antigravity", "Google", "Gemini", "Claude", "GitHub", "Nir-Bhay"], "alternates": {"html": "https://wpnews.pro/news/antigravity-usage-intelligence-track-tokens-quotas-cache-reasoning-and-agent", "markdown": "https://wpnews.pro/news/antigravity-usage-intelligence-track-tokens-quotas-cache-reasoning-and-agent.md", "text": "https://wpnews.pro/news/antigravity-usage-intelligence-track-tokens-quotas-cache-reasoning-and-agent.txt", "jsonld": "https://wpnews.pro/news/antigravity-usage-intelligence-track-tokens-quotas-cache-reasoning-and-agent.jsonld"}}