End-to-end flows on your local browser, powered by natural language. Instant validation. Deep bug discovery. Automation that ships with confidence.
Joined May 2026
- Tweets221
- Following7
- Followers202
- Likes138
A conversion screen that is slightly off still moves real money. Kane CLI checks the amount, rate and fee on a test transfer against the receipt, on web, iOS Simulator or Android Emulator.
Free to start 🔗 testmuai.com/kane-cli/?utm_s…
Buyers without PayPal check out as guests by card, a path few people watch. Kane CLI pays as a guest and passes only once payment is confirmed, on web, iOS Simulator or Android Emulator.
Free to start 🔗 testmuai.com/kane-cli/?utm_s…
A customer filing a dispute, whether from a banking app or a website, is deciding whether to trust you with the next purchase. That form cannot wait a sprint for updated tests.
When your coding agent changes it, describe the expected result to Kane CLI in natural language: the dispute is filed and a case opens under review. Kane CLI runs it in a browser, iOS Simulator or Android Emulator and replies with one line, pass or fail, which your agent or your team can act on right away.
Free to start 🔗 testmuai.com/kane-cli/?utm_s…
A shopper at the PayPal button has already decided to buy, which makes it the worst place to ship a change nobody has tested. When Claude Code reworks express checkout, your Selenium suite will catch up eventually, but the release decision is needed now.
Kane CLI can pay through PayPal from a natural language objective and pass only once the payment is confirmed on screen. Plenty of PayPal payments start on a phone, which is why Kane CLI can take the same path on iOS Simulator and Android Emulator before release.
Free to start 🔗 testmuai.com/kane-cli/?utm_s…
AI coding agents now ship mobile changes faster than teams can update their Appium scripts, and every one of those changes lands on real phones, with chipsets, OS skins and sensors that no simulator fully reproduces.
Kane CLI now runs on Real Device Cloud, so you can test what your agent just built the moment it lands:
> Real Android and iOS hardware, picked from 10,000+ devices
> The same natural language tests you already run on web and emulators
> A sealed evidence pack for every run, published to Test Manager
Run your first real device test from your terminal 🔗 testmuai.com/kane-cli/?utm_s…
#KaneCLI #TestMuAI
kane-cli 0.9.0 🚀
📱 Real devices from any machine. Live device window.
🧩 Runtime vars: global, smart in {{…}}.
📎 Upload once, reuse as {{file.alias}} all session.
⏱️ Replays wait for the page to settle, not a timer.
Real devices, steadier replays
github.com/LambdaTest/kane-c…
What a shopper picks on the product page should be exactly what lands in the bag, and that is easy to break when a coding agent reworks the page twice in a week. Keeping a Playwright test current through every version costs more than the change itself.
Kane CLI reads the chosen color and size back from the bag, from one natural language objective you can rerun after each edit. Your coding agent can trigger it through the Kane CLI skill, so every version of the page, on the web or in the app on iOS Simulator and Android Emulator, gets a verdict before the next one arrives.
Free to start 🔗 testmuai.com/kane-cli/?utm_s…
Sneaker drops happen on phones as much as on laptops, and launch day is the worst moment to find out that last night's change from your coding agent let signed-out visitors into the draw. Your Selenium suite was written for the old entry flow, and nobody has had time to update it.
Kane CLI takes the rule the way you would explain it to a colleague: signed-out visitors are blocked, and signed-in members can enter. It tries both paths on the web and, for the app, on iOS Simulator and Android Emulator, so whoever approves the release sees both outcomes before saying yes.
Free to start 🔗 testmuai.com/kane-cli/?utm_s…
A saved favorite that disappears on reload costs you the sale the shopper meant to come back for. If your coding agent rebuilt favorites this week, your Selenium tests probably still point at the old page, and someone has to decide whether the new version can go out.
The release decision does not have to wait for those scripts. Kane CLI can favorite an item, reload and confirm it stayed saved from one natural language objective. If your app keeps favorites too, run that objective on iOS Simulator and Android Emulator before you decide.
Free to start 🔗 testmuai.com/kane-cli/?utm_s…
How many times has your coding agent changed the review form since your test suite was last updated? Every version needs the same answer: does a review with a photo actually show up on the listing, where the next buyer will see it?
Write that question once for Kane CLI in natural language and rerun it after each change, with no Playwright script to edit. Shoppers who leave reviews from your mobile app deserve the same answer, and Kane CLI can ask it there through iOS Simulator and Android Emulator.
Free to start 🔗 testmuai.com/kane-cli/?utm_s…
Your coding agent can add a membership offer to web and mobile checkout in an hour. The first question in review is what happens when a shopper taps No thanks, and your Cypress tests do not even know the offer exists yet.
Kane CLI answers that question before anyone writes a new test. It declines the offer in a browser, iOS Simulator or Android Emulator, confirms the Place order button still works, and reports back in one line that review can act on.
Free to start 🔗testmuai.com/kane-cli/?utm_s…
When a shopper changes their pickup time, the fee and the store in the order summary should change with it. A coding agent can rework that screen several times before your Playwright tests catch up even once.
With the Kane CLI skill installed, the agent can ask Kane CLI to pick a new slot on the website, or in the pickup app on iOS Simulator and Android Emulator, and read the summary back. The agent then decides whether to keep going based on the pass or fail it receives.
Free to start 🔗 testmuai.com/kane-cli/?utm_s…
Your AI agent can build it. But can it prove that it works?
That's the challenge we're bringing to the Kane CLI Online Hackathon by TestMu AI.
Build with your favorite coding agent, whether it's Cursor, Codex, Windsurf, or your own setup. Use Kane CLI to test, verify, and get your project ready for demo day.
🏆 $3,000 in cash prizes
⚡ $100 worth of free credits
🤝 A 1:1 session with TestMu AI founders for the winners
📅 October 26 – November 2
🌐 Fully online | Solo or teams of up to 4
Build something ambitious. Put it to the test. Show us what you can ship.
Ready to take on the challenge?
👉 Register now: luma.com/kanecli-online-hack…
#KaneCLI #Hackathon #AIAgents #TestMuAI
Kane CLI by TestMu AI retweeted
Your next build could make you win some cool swag. 👀
At Kane CLI Hack Day luma.com/kane-cli-hack-day , we're putting some exciting prizes on the line for builders who can turn an idea into something real and prove it works.
🥇 1st: Apple Vision Pro + $3,000 OpenAI credits
🥈 2nd: Apple Mac mini + $2,000 OpenAI credits
🥉 3rd: Apple iPad + $1,000 OpenAI credits
And it doesn't stop at the podium. Every winning team gets a 1:1 with TestMu AI founders, design partner status, a feature on the TestMu blog, and a spot in the Kane CLI showcase. Every builder also gets $250 in OpenAI credits + 3 months of KaneAI Pro.
All you need to do is build with an AI coding agent, verify it with Kane CLI, and ship it. 🚀 Register for Kane CLI Hack Day, supported by OpenAI
📍 San Francisco | 🗓️ October 24
#KaneCLI #TestMuAI #HackDay #Hackathon #AIAgents #BuildWithAI #OpenAI #Developers
Weekly shoppers rely on Reorder to bring back the same basket, quantities included. Suppose Cursor rewrites order history today: your Cypress suite still describes yesterday's page, and updating it would take longer than the rewrite did.
Kane CLI can reorder a past order and compare every quantity with the original from a single natural language request, in a browser or against the mobile app on Android Emulator and iOS Simulator. If the agent changes the page again tomorrow, that request runs again without a single edit.
Free to start 🔗 testmuai.com/kane-cli/?utm_s…
Nobody wants to rewrite a Cypress test for a feature their coding agent might change again tomorrow, yet someone still has to say whether today's version is safe to release.
Take the grocery setting that tells the store to send only the brand a shopper chose. If it stops saving, the customer gets the milk they refused. Kane CLI checks exactly that from one natural language objective: place an order, then confirm the preference survived checkout. Point it at the website, or at the grocery app on Android Emulator and iOS Simulator, and the answer comes back as one line while the Cypress suite catches up at its own pace.
Free to start 🔗 testmuai.com/kane-cli/?utm_s…
kane-cli 0.8.22 🎯
📱 Mobile replay matches web: until loops, settings.
🔗 Imported files see the importer's variables.
🔁 until in a branch exports as a loop there.
💳 balance shows available credits only.
Same test, any platform.
github.com/LambdaTest/kane-c…
Two stores in one Instacart cart means two fees and two delivery windows. Checkout has to keep them apart.
The split, verified on Instacrate:
> "2 stores · each delivers separately" at checkout
> GreenLeaf Market at a $3.99 service fee, BulkBarn Wholesale with its own
> a 3:00pm to 5:00pm window for GreenLeaf, an evening 6:00pm slot for the second store
Each store card is checked on its own. Merged stores or swapped fees would fail it.
12 seconds, 5 steps, PASSED. It's all in the video.
Free to start 🔗 testmuai.com/kane-cli/?utm_s…