Back to Blog
Tutorials

Antigravity Browser Extension: The Complete Guide for 2026

Everything you need to know about the Google Antigravity Chrome browser extension. Installation, configuration, browser subagent capabilities, Agent Manager integration, WSL setup, troubleshooting, and advanced tips for web developers.

Antigravity Team
March 8, 2026
10 min read

Antigravity Browser Extension: The Complete Guide for 2026

The Google Antigravity browser extension is one of the most powerful yet underutilized features of the Antigravity IDE. While many developers use Antigravity for code generation and terminal automation, the browser extension transforms it into a true full-stack development partner — one that can see, interact with, and test your web applications just like a human user.

In this guide, we cover everything from installation to advanced use cases, including the latest 2026 updates.

Why the Browser Extension Matters

Traditional AI coding assistants can write code, but they can't see the result. They generate HTML and CSS blindly, hoping it looks right. The Antigravity browser extension changes this fundamentally:

  • Agents can see your app — they take screenshots and analyze the DOM
  • Agents can interact with your app — they click, type, scroll, and navigate
  • Agents can verify their own work — visual regression testing catches bugs automatically
  • Agents can record evidence — video recordings and screenshot sequences document everything

This closed feedback loop (write code → see result → fix issues → verify) is what makes Antigravity's browser integration so powerful.

How to Install the Browser Extension

Automatic Installation (Recommended)

  1. Open Antigravity IDE and start a conversation
  2. When an agent needs browser access, click "Setup" in the progress panel
  3. Click "Add to Chrome" in the Chrome Web Store
  4. Grant the required permissions
  5. Return to Antigravity — a green indicator confirms the connection

Manual Installation

If automatic setup doesn't work:

  1. Open Chrome and search for "Google Antigravity" in the Chrome Web Store
  2. Click "Add to Chrome" on the official extension (publisher: Google LLC)
  3. Restart both Chrome and Antigravity IDE
  4. Click the Chrome icon in the Antigravity toolbar to verify the connection

Reinstalling After Removal

  1. Go to chrome://extensions and remove the existing extension
  2. Restart Chrome completely (close all windows)
  3. Reopen Antigravity IDE
  4. Click the Chrome icon to trigger a fresh installation

Understanding the Browser Subagent

When the extension is active, Antigravity spawns a browser subagent — a specialized AI model optimized for web page interaction. This subagent has access to powerful tools:

Interaction Capabilities

  • Click buttons, links, and interactive elements
  • Type into input fields and text areas
  • Scroll pages vertically and horizontally
  • Select dropdown options and checkboxes
  • Hover over elements to trigger tooltips and menus

Reading Capabilities

  • DOM capture — read the full page structure
  • Screenshot capture — take visual snapshots at any point
  • Markdown parsing — convert page content to readable markdown
  • Console log reading — access JavaScript console output
  • Network inspection — monitor API calls and responses

Recording Capabilities

  • Video recording — capture entire browser sessions
  • Action timeline — log every interaction with timestamps
  • Performance metrics — measure load times and render performance

When the agent is actively controlling the browser, you'll see a glow around the browser window — a visual indicator that Antigravity is in control.

Agent Manager + Browser Extension

One of the most powerful combinations in Antigravity is using the Agent Manager with the browser extension. Here are three workflows to try:

1. Parallel Testing Pipeline

Spawn three agents simultaneously:

  • Agent 1: Tests the registration flow
  • Agent 2: Tests the dashboard functionality
  • Agent 3: Tests the settings page

All three run in parallel, each with their own browser tab, cutting testing time dramatically.

2. Build-Test-Verify Loop

Create a chain of agents:

  • Agent 1: Writes the code and starts the dev server
  • Agent 2: Opens the browser and visually verifies the output
  • Agent 3: Runs automated tests and reports results

3. Cross-Viewport Testing

Use multiple agents to test different screen sizes simultaneously:

  • Agent 1: Tests at 1920x1080 (desktop)
  • Agent 2: Tests at 768x1024 (tablet)
  • Agent 3: Tests at 375x812 (mobile)

Using the Extension with WSL

If you run Antigravity in WSL (Windows Subsystem for Linux), the browser extension works with Chrome on your Windows host:

Requirements

  • WSL 2 (WSL 1 does not support browser integration)
  • Google Chrome installed on Windows
  • The Antigravity extension installed in Windows Chrome

Setup

  1. Install the extension in your Windows Chrome browser
  2. Launch Antigravity from your WSL terminal
  3. If Chrome isn't auto-detected, set the BROWSER environment variable:
export BROWSER="/mnt/c/Program Files/Google/Chrome/Application/chrome.exe"
  1. The agent will open Chrome windows on your Windows desktop

Advanced Tips & Tricks

Tip 1: Use Browser Actions for Verification

Instead of just asking agents to "check if the page works," be specific:

Open http://localhost:3000, click the login button,
enter test@example.com as email and password123 as password,
submit the form, and verify that the dashboard page loads
with the welcome message.

Tip 2: Combine with Screenshot Artifacts

Ask agents to take screenshots at each step of a workflow. This creates a visual timeline you can review at a glance:

Build the new header component, then take screenshots
of the homepage at mobile, tablet, and desktop widths.

Tip 3: Debug with Console Logs

The browser subagent can read console output. Use this for debugging:

Open the app, perform the checkout flow, and check the
browser console for any errors or warnings.

Tip 4: Visual Regression Baseline

Before making changes, have the agent screenshot all key pages. After changes, compare the before/after screenshots automatically:

Take screenshots of every page in the app navigation.
Then after I make my CSS changes, compare them to find
any visual regressions.

Troubleshooting Common Issues

| Issue | Solution | |-------|----------| | Extension not connecting | Restart Chrome and Antigravity IDE. Check chrome://extensions. | | Screenshots failing | Ensure the tab is visible (not minimized). Grant screenshot permission. | | Agent can't click elements | Refresh the page. Check for Shadow DOM or iframes. | | WSL can't find Chrome | Set BROWSER env variable to Windows Chrome path. | | Video recording fails | Check disk space. Close other screen recording apps. | | Extension disappeared | Reinstall from Chrome Web Store. See manual install steps above. |

What's New in 2026

Recent updates to the Antigravity browser extension include:

  • Improved video recording quality — higher frame rate and resolution
  • Better Shadow DOM support — agents can now interact with components inside shadow roots
  • Network request inspection — agents can view API calls and responses in real-time
  • Performance metrics — automatic Lighthouse-style performance auditing
  • Multi-tab management — improved handling of multiple browser tabs from the Agent Manager

Conclusion

The Antigravity browser extension is essential for any web developer using the Antigravity IDE. It transforms AI assistance from code-only to full-stack, giving agents the ability to see, interact with, and verify your web applications.

If you haven't set up the extension yet, take 5 minutes to do it now — the productivity gains are immediate.

Related Reading:

Share this Article