Summary
Google Antigravity: is Google's agent platform worth it?
See what Google Antigravity is, how to sign in with a Google account, what it offers, and how to use it alongside other agents in VS Code or the Antigravity IDE.
Marcelo Santiago
Criador do bridgecode
Get AI articles in your inbox
Technical curation on models, agents and coding tools. No spam, unsubscribe anytime.
- What you will learn: what Google Antigravity is, how to access it and how to use it day to day.
- Who it is for: people starting with vibe coding who have a Google account.
- Outcome: Antigravity working and a simple flow to create, review and publish.
- Analysis basis: official documentation, no own performance testing.
Google Antigravity is Google's programming agent platform. You describe what you want, and the agents plan, write and test the code.
According to the official documentation, it appears in several forms: a desktop app, a CLI called agy, extensions for editors like VS Code and JetBrains, and a Python SDK.
What you need before you start
- A Google account.
- A test project in Git.
- Optional: VS Code, if you want to use the extension.
How to access: start with your Google account
Download the app or install the extension from the links in the official documentation. When you open it, sign in with your Google account.
There is an individual free tier and larger tiers tied to Google AI plans. Each has weekly limits and five hours of usage. Check the official page for what applies to your account.
What it offers
- Several models: Gemini, plus Claude and GPT models.
- Subagents: helper agents that work in separate copies of the project, the git worktrees.
- MCP: connection to external tools, such as databases and browsers.
- Skills: custom instructions you teach the agents.
- Permissions: you decide what the agent can do without asking.
Is Google Antigravity good? Is it worth it?
If you have a Google account and want to start for free, it is a great starting point. The free tier lets you test the agents at your own pace.
Usage limits are the point to watch. On large projects, you can hit the limit before the end of the day.
App, extension or CLI: which to use?
The desktop app is the simplest way to start: you chat with the agents in its own window. CLI agy runs in the terminal and works well for those who already work in VS Code. See the step by step in how to use Antigravity CLI in the VS Code terminal.
Using with other agents
Antigravity does not need to work alone. bridgecode is an extension you install in VS Code or Antigravity IDE and that accepts Antigravity as an agent.
This way you combine Antigravity with Claude Code, Codex or OpenCode, each with its own account. In Claude Code, for example, you sign in with a Pro or Max subscription, as shown in our guide on how to install and use Claude Code in VS Code.
Practical flow: create, review and publish
1. Create
Write the request as if you were talking to a colleague:
Request: "Create a one page site for a cafe, with menu and map. Before you change anything, show me the plan."
Read the plan, approve it or ask for adjustments. Then the agent writes the code and runs the project to check.
2. Review
Look at the changes before accepting them. The editor's Source Control tab shows each line changed.
Also ask: "Review what you did and explain in simple language." If something breaks, paste the error in the chat and ask for a fix.
3. Publish
Git is the project's version history. A pull request, or PR, is a request to merge changes into the main project after review.
- Ask: "Create a branch called site-cafeteria and make a commit with a clear message."
- Ask: "Push the branch to GitHub and open the PR."
- Review the PR on GitHub and merge it.
The same flow in bridgecode, with separate roles
In bridgecode, Antigravity can take on one role on the team while other agents handle the rest:
- Planner: reads the request and breaks it into small tasks.
- Executor: writes the code for each task.
- Reviewer: checks everything and flags errors before the PR.
Each agent works on a separate copy of the project, so one does not interfere with the other. You approve the plan at the start and the PR at the end.
For example, Antigravity executes and Claude Code reviews. This also helps split usage across each account's limits.
Common mistakes and how to avoid them
Granting all permissions
Start by asking for approval for commands. Grant access gradually, once you trust the flow.
Ignoring limits
Track how much you've already used. Break large tasks into small parts.
Thinking it's just another editor
Antigravity is focused on agents. Use plan mode and review before accepting.
To understand why agentic environments are growing, see why devs are switching from traditional IDEs to agentic environments.
Frequently asked questions about Google Antigravity
Is Google Antigravity any good?
If you have a Google account and want to start with agents at no cost, it's a solid starting point: it has a free tier, multiple models (Gemini, Claude and GPT) and subagents. The caveat is usage limits on larger projects.
What is the difference between the Antigravity app and CLI?
The app is a dedicated window for chatting with agents. The CLI, invoked with the agy command, runs in the terminal and pairs well with those who already work in VS Code.
What are the Google Antigravity plans?
There is an individual free tier and higher tiers tied to Google AI plans. Each tier has weekly and five hour limits; check the official page for your account.
What is ADE?
ADE stands for Agentic Development Environment, or agentic development environment. It is an environment where AI agents receive a goal, plan, write, test and deliver the result for you to review.
Is Google Antigravity free?
There is an individual free tier with usage limits. Higher tiers depend on Google AI plans.
What is MCP?
Model Context Protocol is a standard that connects AI agents to external tools, such as databases and browsers.
What is a subagent?
It is a helper agent that receives part of the work and operates in a separate copy of the project.
What is a pull request?
It is a request to merge your changes into the main project. It shows everything that changed and allows review before the merge.
A good start for those with a Google account
Antigravity is an easy entry point for agents, with a free tier and several models. With other agents alongside, it delivers even more.
Want to try it? Use Antigravity in an AI agent squad on bridgecode and put planner, executor and reviewer in VS Code or in the Antigravity IDE.
Sources: Google Antigravity official documentation and official bridgecode site
Related articles

How to test AI agents in cloud operations with CloudOpsBench
CloudOpsBench releases 45 cloud operation tasks to test your AI agent. See how to clone, validate and run against an emulator or real AWS.

Tencent's Hy4: 1M context, vision and tool calling
Tencent's Hy4 brings 1 million tokens of context, vision and native tool calling for code agents and long duration tasks.

5 tips to operate desktop with agent-desktop
Vercel Labs launched agent-desktop, a tool that lets AI agents operate desktop applications via the accessibility tree. See how it works.
