From an idea to the App Store,
and every day after.
You describe it, or hand it over
New app: one page in plain English about what it is, who it is for and what done looks like. Existing app: connect the code, and the engine reads it the way a new engineer would.
It plans
The engine, the part of Studio that does the work, breaks the app into milestones and small, checkable pieces, sizes them, and keeps refining the plan as the app grows.
Builders work in parallel
Several builder agents take pieces at once. Tests run first; a second reviewer agent must approve every meaningful change before it lands. Nothing sloppy gets through.
Only what needs you, waits
Spending, anything the public will see, and open questions can park for your tap, in one queue. You answer that queue from any browser, and from the Studio app on your phone. Everything else keeps moving.
It ships
TestFlight and Google Play testing tracks, and instant updates between store builds. When you say go, Studio submits the same build to the App Store and Google Play for real. A QA agent taps through the real app before each release. Every change is tested and reviewed by a second AI before it can join a release.
It gets the app through review
A human at Apple reading your app is where most releases stall, so Studio owns that part too. It submits the build, watches for the verdict, and when Apple pushes back it reads the notes and turns each reason into a fix. All it needs from you is the code Apple texts you, and one tap before the app goes back.
The loop keeps turning.
Once an app is in people’s hands, the same engine listens, fixes, grows and reports, without you forwarding a single email.
It hears your users
Crashes and reports from people using the app arrive in one inbox. A bug reported at midnight can be a fix on TestFlight by morning.
It grows the app
Ads and posts drafted from the real, running app, in the brand you approved. Nothing goes public without your OK.Posts and store listings drafted from your app, in the brand you approved. Nothing goes public without your OK.
It watches the numbers
Installs, ratings, crashes and revenue across every app. A dip becomes a ticket, not a chart you have to notice.
It runs on your schedule
Wake at six, build until the queue is clear, ship what passed, sleep. Or react to an event, like a new report.
One app, everyone on it.
Five members on every plan, with the access you choose. A co-founder, a designer and a client all watching the same app at the same time.
Decide how much you want to be asked.
Ask first
Every piece of work is proposed to you first. Good for your first week, or your first app.
Build on its own
Work runs unasked; anything touching money or the public, any store submission, and any open question, waits for you.
Spend and publish without asking
An optional switch under Advanced: money, public work and generated pictures clear without a tap, inside the caps you set. Store submissions, visual sign-offs and agent questions still wait for you.
For the engineers: what is actually running.
Skip this if you just want an app. Read it if you want to know what you are trusting. This is the path every change takes; none of it is optional.
Your repo, your history
Expo / React Native projects. Real git commits on real branches; the engine never rewrites history. Bring one repo or a whole portfolio, each with its own engine and its own supervision level.
Gates, not vibes
Lint and tests before review. An adversarial reviewer on every non-trivial diff. A baseline test gate that blames only new failures. Per-change and daily spend caps you set. Tiny, docs-only diffs skip review; nothing else does.
Your accounts, your keys
Builds go through your Apple, Google and Expo accounts; the AI runs on your own Claude subscription. Nothing is metered and nothing leaves your name. Generated artwork is the one thing we supply, and your own always replaces it.