They fail at the niche. Thirty-five prompts and a twenty-one day sequence for finding a market that already spends money, building products worth keeping, taking payment from anywhere in the world, and running ads that do not get your account restricted.
The problem
Here is what happens to the people who do. The products are thin. The first hundred buyers ask for refunds. The platform flags the account for chargeback ratio. The ad account gets restricted for low-quality landing pages. The business is dead in six weeks.
Volume without a quality floor is not a business model. It is a countdown.
The uncomfortable arithmetic underneath all of this: one product that sells five hundred copies beats two hundred products that sell one each. Volume is for multiplying a proven winner, not for buying lottery tickets. Gold Pocket OS is built around that sentence.
What you get
Each prompt is written to be pasted straight into Claude — role, method, output format, and a standard it is held to. Each one tells you what good output looks like and what to do when it goes wrong.
Nine prompts that turn Claude into a research operator — and tell you honestly when a niche is a dud. The sequence is front-loaded with disqualification, because killing a bad idea on day three is worth more than any idea you have on day one.
Nine prompts that build products at volume with a quality gate you are meant to actually use. Includes the Kill Gate — a fresh-context audit that grades your own product before a buyer does.
Eight prompts that build a fast, honest sales page and wire it to a checkout that can legally take money worldwide. Includes the payment-platform decision written out properly, with the specific constraint that stops sellers in India — and the way around it.
Nine prompts for Meta and Google. Built around the rule that actually kills this category — income-claims policy, enforced by classifiers that read your ad and your landing page. Every prompt here has the constraint wired in, including a pre-flight audit you run before you spend anything.
Included with the complete bundle
Thirty-five prompts without a sequence is a folder. The Playbook is the twenty-one day plan that turns them into a method — what to run each day, what you should have by the end of it, and the gate you must pass before moving on.
Each gate exists because the failure it prevents is expensive and invisible until it is not.
Free, no purchase
A worksheet that gets you to your own number in about fifteen minutes: what it actually costs to build and market your first ten digital products with AI. Three honest budgets, every line item with a real price, a break-even table, and the five specific ways people burn the money.
The short version, if you want it now: the smallest honest start is about $25 and three weeks. Everything above that is optimisation.
One email, the PDF attached, and then roughly one useful thing a week. Unsubscribe in a click.
Check your inbox. The worksheet is on its way — usually within a minute. If it is not there, look in Promotions or Spam and mark it "not spam" so the rest reaches you.
By submitting you agree to receive email from Gold Pocket OS. We do not sell or share your address. Unsubscribe any time.
Pricing
All four packs, plus the Playbook — the twenty-one day sequence that makes them work together. The Playbook is not sold separately.
14-day refund, no questions asked
Start with one if you would rather try before committing. Pick any pack above — most people start with Pack 01, because the niche decision is the one everything else depends on.
Buy three and you have paid more than the bundle
Honesty section
Questions
No, but it helps. The free tier genuinely works for a first product — you will hit usage limits and wait. A Pro plan removes that ceiling. The Budget worksheet covers exactly where this sits in the costs.
Mostly. They are written for Claude — the setup instructions use Claude Projects, and the prompts assume live web search. On other models you will lose the persistent-instruction step and some depth on the research prompts. If you are committed to another model, buy a single pack first and see how it goes.
A course teaches you to do the work. This does the work with you. There are no videos, no modules and no community. It is thirty-five prompts, a sequence, and five gates that tell you when to stop.
Pack 03 covers this properly, because it is where most people get stuck. It walks through merchant-of-record platforms, which handle global tax for you, and it names the specific payout constraint that trips up sellers in India along with the way around it. Not tax advice — check your own position with an accountant — but the operational path is written out.
Nobody can honestly tell you that, and this page will not pretend otherwise. What this removes is the four specific ways it goes wrong for most people: picking a market with no money in it, shipping products too thin to keep, building a checkout that quietly loses buyers, and getting an ad account restricted in week one. Avoiding those four is most of the game. The rest is yours.
A zip file. Inside: the pack as a designed PDF, and a folder of plain .txt files — one per prompt — so you can copy them into Claude without PDF formatting mangling the text. The complete bundle adds the Playbook and the Budget worksheet.
Fourteen days, no questions asked. Reply to your receipt email and it is done. You keep the files — chasing people for downloads of a PDF is not a good use of anyone's afternoon.
Who made this
I run a commercial real estate consultancy in Mumbai. I built this because I wanted to sell digital products alongside it and could not find anything that told me the unglamorous parts — the payment constraint for Indian sellers, the ad policy that gets accounts restricted, the point at which a generated product becomes something a buyer will ask for their money back on.
So the packs are written the way I wanted to be told: with the gates in, the maths shown, and the parts that do not work said out loud.
14-day refund, no questions asked · Instant download