Aura Pro: Test-Driving isDemo=true Before You Buy a $299.99 Android App

A completed Android app is a different purchase from a codebase dump.

When someone buys a full app — not a template, not a repo to finish — the open question
is never "does the code compile?" It is "does the product actually do what the listing
says, on the phone I already own, before money moves?"

That is the gap Aura Pro closes with its demo APK, and it is the part worth walking
through in detail because it changes the risk profile of the whole purchase.

What the listing promises

Aura Pro is an AI prompt-generation app for Android: 11 intent categories (Build, Write,
Design, Learn, Marketing, Business, Personal, Finance, Legal, Debug, Analysis), each
producing a complete structured professional prompt. Core generation runs on-device with
no API key required. Google Play Billing is already wired via BillingClient 6.2.0 with two
subscription products — com.auraprompt.app.premium_monthly and
com.auraprompt.app.premium_annual.

The price is $299.99, one-time.

On its own that is a normal listing. The interesting part is the verification path.

The demo build is the same build, unlocked

The APK ships with a single flag: isDemo=true. With that flag set, every premium
category is unlocked — no paywall, no account, no API key, no license check. You install
it on a phone or an emulator and run the entire feature surface yourself.

That matters because most app purchases are verified the hard way: you clone a repo, wire
up your own signing key, build, deploy, and then find out what the demo video skipped.
Here the artifact you test is the artifact you would be selling. Nothing is stubbed out
for the demo.

Concretely, the sequence is:

1. Download the demo APK.
2. Install it on any Android device or emulator (allow installs from unknown sources).
3. Open it and run each of the 11 categories end to end.
4. Confirm the output is structured the way you would want to publish it.
5. Only then decide on the purchase.

Why "works offline" is the claim that has to be tested first

Of all the claims on the page, the one with the biggest gap between says and does is
the offline behaviour. Plenty of apps describe themselves as local-first and still reach
for a network call on the cold path.

So the test is not "does it work while I have Wi-Fi" — it is: enable airplane mode, force
stop the app, clear it from memory, relaunch, and generate. If the app needs a round trip
to anything, it will fail here and you will know in ten seconds.

Same test for the prompt structure. Run Marketing and Legal back to back — the two are
the least similar categories, and if the output format is identical across them, the
"11 categories" framing is a dropdown, not eleven templates. If the structure actually
changes, the framing holds.

What you actually own after the sale

The Play Store takeover is handled by two scripts: play_takeover.py takes your Play
Developer API key (--key key.json) and does everything the Play Developer API permits —
APK upload, store listing (auto-filled from the app's own content), store images, in-app
products, subscription products, track activation, and publish.

Six steps are manual, because the Play Console requires a human: Data Safety declaration,
content rating, target audience, app category, news app declaration, and policy checklist.
Those are the same six steps any Play listing requires; the API simply does not expose them.

Also worth being explicit about: after the sale the app is yours. The subscription
products live in your Console, the revenue goes to your account, and there is no ongoing
involvement from the seller.

How payment works

PayPal — send to the account email [e-mail] and include a note naming
the product. The seller confirms in the inbox.

Crypto — 0x078fE079A9440631Bc6AD722D7143dC382b4ea4f on Base (ETH / USDC / USDT).

Also accepted: Western Union or MoneyGram.

All three are accepted; there is no preference and no discount attached to any of them.

The decision rule I would use

If you are buying an app to resell, the demo APK is the due diligence. If it does not
survive airplane mode, if the 11 categories collapse into two output shapes, or if the
premium unlock does not hold across a reinstall — none of those are fixable by the
takeover scripts, and they are all discoverable before you pay.

If it does survive all three, you are buying a finished product with a documented
publishing path, which is a materially different thing from buying code.

Live product page with the working demo:
https://reggiebro.github.io/WorkflowAlpha.ai/landing/