Test the apps your Antigravity agents build without grading their own work
Antigravity agents build your app and grade it with their own screenshots. Kane CLI gives an independent verdict instead: it drives the running app in a real browser and returns a true pass or fail. Free to install.
or read the documentation
Why test apps built with Antigravity
Google Antigravity is Google's agent-first IDE, where a Manager view runs up to five agents in parallel across the editor, terminal, and Chrome. One task scaffolds a feature, launches the dev server, and lands a lot of code.
Antigravity's answer to trust is Artifacts: screenshots and recordings the agent produces. But those are the same model judging its own work, and a model that shipped a broken OAuth redirect is no judge of whether login holds.
Kane CLI is the independent check. Describe the journey in plain English and it drives the running app in a real Chrome browser, signs in through the OAuth callback, and flags where it breaks. Agents run it via the --agent flag.
What Kane CLI tests in your Antigravity app
The running app, the sign-in flow, and the new feature, verified independently of the agent that built them.
The running app the agent just built
Antigravity scaffolds a feature and spins up the dev server. Kane CLI drives that running app in a real browser and confirms the whole feature works, not just that the agent's screenshot looked right.
Sign-in and the OAuth callback
Walk the full Google sign-in journey, follow the redirect after the OAuth callback, and confirm the session holds. AI-generated auth fails most here: a post-callback redirect that lands the user back on the sign-in screen.
Routes, handlers, and navigation
Click through the routes the agent generated and confirm each resolves. Catch the invalid route handler export and the build break that follows a Next.js upgrade before the deploy goes red.
The new feature's exact flow
An agent ships one feature per task. Describe that flow once, from entry point to saved result, and verify the new code does what the task asked, end to end in a real browser.
Layout the agent patched but never re-ran
Antigravity often spots a misaligned button, patches the CSS, and screenshots once. Kane CLI re-runs the real interaction so the fix holds across the flow, not just in that single frame.
Regression on the next parallel agent
With up to five agents editing in parallel, the next task can quietly break a flow another agent shipped. Lock in known-good journeys and re-run them on every change so regressions surface immediately.
Trust the running app, not the agent's screenshot
Start in your terminal
Validate on the cloud
Release with confidence
Independent eyes on what the agent built
Kane CLI and KaneAI share the same automation engine and dashboard.
From a passing Artifact to a passing flow
Antigravity agents plan and write the app across editor, terminal, and browser. Kane CLI proves the running app works for a real user before the feature reaches anyone, closing the gap between a passing Artifact and a passing flow.
Independent verification, not the agent grading itself
Antigravity Artifacts are screenshots and recordings the same model produced and judged. You, or the agent through the --agent flag, point Kane CLI at the running app for an independent real-browser pass or fail, so a model that wrote a broken redirect cannot also sign off on it.
Evidence you can share
Every run produces a persistent video, a step trace, and a replay link you can drop into a PR, a bug report, or a team message, separate from the agent session it came out of.
Test your Antigravity app in three steps
Install Kane CLI
Run npm install -g @testmuai/kane-cli and sign in with your TestMu AI account. It lives in the Antigravity terminal alongside the dev server your agent launched.
Point your agent, or run it yourself
Point your Antigravity agent at the Kane CLI guide so it runs flows with the --agent flag against the running app, or run them yourself from the terminal against the local dev server or your deployed domain.
Describe the flow and verify
Write the journey in plain English, like sign in with Google and confirm the project saves. Kane CLI drives a real browser, verifies each step, and returns a pass or fail with evidence.
Get Started With Kane CLI
🎉 Launch offer: Bonus credits for the first 3 months on paid plans
Choose the right plan for you
Local test authoring via CLI
Auto-heal & vision
View test cases on UI
Test Manager
Free
$0
/month
200 Credits
Resets in every
30 days
Starter
$19
/month
2000 Credits
Launch: 4,000 Credits (+100%)
Bonus for first 3 months
Pro
$99
/month
10,000 Credits
Launch: 15,000 Credits (+50%)
Bonus for first 3 months
Enterprise
Get access to solutions built on Enterprise-Grade Security, Privacy, and Compliances.
Need more credits?
Got a bigger use case in mind?
Let’s talk
Choose the right plan for you
Free
$0
/month
200 Credits
Resets in every
30 days
Starter
$19
/month
2000 Credits
Launch: 4,000 Credits (+100%)
Bonus for first 3 months
Pro
$99
/month
10,000 Credits
Launch: 15,000 Credits (+50%)
Bonus for first 3 months
Enterprise
Get access to solutions built on Enterprise-Grade Security, Privacy, and Compliances.
Need more credits?
Got a bigger use case in mind?
Let’s talk
Get the technical rundown
Documentation
Everything you need to install, configure, and run Kane CLI in under 2 minutes.
Frequently asked questions
Install Kane CLI, then describe the journey in plain English, for example "open the app, sign in with Google, create a project, and confirm it appears in the Manager dashboard." Kane CLI drives the running app in a real Chrome browser, checks each step, and returns a clean pass or fail with a video recording. Run it yourself from the Antigravity terminal against the dev server the agent spun up, or let the agent run it through the --agent flag, so you get a result you can trust instead of a screenshot it graded itself.
Antigravity runs up to five agents in parallel across the editor, terminal, and browser, shipping a lot of code while you are away from the keyboard. Its screenshots and browser recordings are Artifacts, but that is the agent grading its own homework: the model that wrote a buggy OAuth callback is not the best judge of whether login works on a second run. AI-built apps frequently ship with auth and redirect issues after the OAuth callback, route handler and build breaks, and CSS the agent patched but never re-checked end to end. Kane CLI is an independent real-browser pass or fail on the running app, not the model's own verdict.
Yes, and this is the accurate framing for an editor-and-terminal agent. Antigravity agents have direct terminal access, so point yours at the published guide at testmuai.com/kane-cli/agents.md and it will install Kane CLI, run flows with the --agent flag against the dev server it launched, read the structured results, and fix the failing route or redirect before it hands you the task. The agent builds the feature, then drives a real browser to prove it works, in the same run.
The full sign-in journey including the OAuth callback redirect that AI-generated auth commonly mishandles, the running app the agent just scaffolded with its dev server live, the flow the new feature added, route handlers and navigation that break after a framework upgrade, console errors, and the layout the agent patched but never re-verified. It checks each step, not just the final screenshot, and flags the exact point where the app breaks so the next agent run can fix it.
Yes. Antigravity Artifacts live inside the agent session, but your CI needs a hard gate. Authenticate with your TestMu AI credentials, pass --headless and --timeout, and gate your pipeline on the exit code: 0 on pass, 1 on fail, 2 on setup or auth errors, and 3 on timeout. Run your sign-in and core-feature flows on every deploy, so a redirect loop or a broken route never reaches production even when an agent ran out of context mid-task.
The CLI is free to install and use. Local runs are free; cloud runs on the TestMu AI grid are billed against your TestMu AI plan. Start on the free tier and verify your Antigravity app end to end without a credit card.
Teach your Antigravity agent to verify what it ships
Antigravity agents have direct terminal access. Point yours at the Kane CLI guide and it will install, authenticate, and verify the running app's sign-in and new-feature flows in a real browser on its own, before it hands you the task.
Point your agent to: testmuai.com/kane-cli/agents.md