Browser Integration

Antigravity Track

Module 39

Antigravity Track — Module 39

The Agent Tests Itself: ThreadCo's developer adds a new checkout form field. Instead of opening a browser and manually testing it, she tells the agent: "Test the gift wrapping field end-to-end in the browser." The agent opens Chrome, navigates to localhost:3000, fills the form, submits it, reads the response, and reports back — with a screenshot proving it worked.

Antigravity's Chrome extension lets agents autonomously interact with web applications — navigating, clicking, filling forms, reading DOM state, and taking screenshots — without you touching the browser.

Install Once, Use Always

Install the Antigravity Chrome extension from the Chrome Web Store. It connects automatically to any running Antigravity session. No per-project configuration needed — it works across all your projects.

Full DOM Access

Agents can read any element on the page — text, input values, computed styles, network responses. They use this to verify that UI changes actually work, not just that the code compiles.

Form Interaction

Agents fill text inputs, select dropdowns, tick checkboxes, click buttons, and handle file uploads. They can complete multi-step flows end-to-end — login, navigate, submit, verify result.

Screenshot Artifacts

After every browser action, the agent captures a screenshot saved as a verifiable artifact in the Manager Surface. You can see exactly what the agent saw — and share it as evidence of a working feature.

Agent Testing the Checkout Form

Antigravity — Browser Agent — Chrome

Browser Agent Log
✓ Navigate to localhost:3000
✓ Click "Add to cart"
✓ Click "Checkout"
✓ Fill shipping form
→ Find gift wrap checkbox
○ Tick checkbox
○ Type gift message
○ Submit order
○ Verify £2.50 added to total
○ Take screenshot
Browser preview — localhost:3000/checkout
Order Summary
Sunset Gradient Tee × 1£29.99
Add gift wrapping (+£2.50)
Gift message: "Happy Birthday! 🎂"
Total£32.49

          ✓ Screenshot captured — gift_wrap_test_01.png saved as artifact
        

What Browser Agents Can Do

Action	Example
Navigate	"Go to the product page for SKU #4821"
Click	"Click the Add to Cart button"
Fill forms	"Fill the shipping form with test data"
Read DOM	"What is the total shown in the order summary?"
Verify state	"Confirm the gift wrap checkbox is checked and £2.50 added"
Screenshot	Automatic after every significant step
Multi-step flows	"Complete a full checkout as a guest user"
Error detection	"Report any console errors or network 4xx/5xx responses"

Works with Any Web Framework

The browser agent works with any web app that runs in Chrome — React, Vue, Next.js, plain HTML, Django, Rails. It doesn't care about the stack. As long as it runs on a URL, the agent can interact with it.

← Manager Surface Next: Best Practices →