👨🎨 We're sharing a formula for making vibe-coded apps look professional.
It's a deconstruction of the "science of design" and we think it will help you coach your agents into mobile app design that brings polish to your apps.
The formula and examples are in the post below ↓
how to improve your /goal by designing the loop before you run it
a poorly designed loop burns tokens and hands you slop fast, that´s why it's important to spend time designing the loop harness
instead of writing a /goal and hoping it works, you run it through LOOPER skill first and it:
> critiques your loop and sharpens it against a library of best practices
> builds a verification rubric, so the loop actually knows what done means
> lets you add an llm as judge, codex and non-claude models included
> sets guardrails and a token budget, tight or generous, so it doesn't run away
> exports the whole thing as a portable artifact you can repeat and edit
it's a design layer that helps you build a loop that produces quality, then hands it to Claude Code or other models
this pairs with the LLM council setup I posted last week. design the loop in looper, run it with a second model checking the work
nice work @KSimback
we wrote playbooks for almost every way a startup can grow:
> reddit marketing
> seo
> llm/chatgpt visibility
> ugc videos
> getting 100k-1m views on launch videos
bookmark this thread and steal the playbooks below 🧵
It’s incredible that /make-interfaces-feel-better has already been installed more than 30,000 times.
It contains a lot of tips that make interfaces feel better, across UI, animations, performance and more.
npx skills add jakubkrehel/make-interfaces-feel-better
apple is literally handing out $200 in free ad credit and most indie devs have no idea
set up apple search ads
get $200 to spend
test if your app actually converts
free money, takes 10 min
stop telling Claude Code/Codex "do this".
stop telling Claude Code/Codex "write code".
stop telling Claude Code/Codex "fix this bug".
you're using a 𝘀𝗲𝗻𝗶𝗼𝗿 𝗔𝗜 like it's a 𝗷𝘂𝗻𝗶𝗼𝗿 𝗶𝗻𝘁𝗲𝗿𝗻.
here are 8 prompts and goals you can copy-paste directly.
Do you know why most beginners optimize for installs?
Because it’s easier.
Nobody wants to learn attribution.
What’s attribution?
What’s SKAN?
What’s a conversion window?
Today you can ask your AI assistant and get an answer in seconds.
The marketers who win aren’t the ones who know everything. They’re the ones who aren’t afraid to learn.
Track IDFA when possible.
Set up Conversion API.
Use attribution tools.
Without data, ads are expensive.
With data, algorithms learn faster and find your customers sooner.
You shipped something. You have 12 followers. Posting on X alone won't move it.
Here's where to get your first users when nobody knows you yet.
1. Directories. A listing gets you a permanent backlink and visitors actively looking for new tools. Start here, it compounds.
2. Niche communities. Reddit, Discord, indie Slack groups. Answer questions, drop your tool when it actually fits.
3. Reply, don't post. Comment under bigger builders in your space. One good reply beats 10 posts no one sees.
4. Launch platforms. @nicklaunches, @MicroLaunchHQ, @Better_Launch, @launch_llama and the 100+ on the directory list.
5. Direct outreach. DM 20 people who have the exact problem you solve. Slow, but your first 10 users are usually hand-picked.
None of this needs a following. It needs reps.
Which one are you sleeping on? Drop your link and I'll point you to the right starting spot 👇
I often use Firebase Anonymous Authentication.
You get a user ID and users can start using the app immediately.
No passwords.
No sign up buttons.
If users need a full account later, you can always ask then.
New screenshots help boost my Conversion Rate to 11.2% from 0.5%. 📈
Your screenshots are probably the problem if you're not getting app downloads...
Fix them using these best practices from @designerants
your churn rate is basically decided by 4 screens of code.
obviously the best way to have high retention is to build a great product that solves a real problem.
but a strong cancellation flow is the second biggest lever and most apps don't even have one.
two apps with the same product and similar users can have wildly different churn based on what happens in the 60 seconds before the user taps confirm cancel.
the apps with the lowest churn use the same pattern:
intercept the in-app cancel button, run a custom flow, and only deep-link to apple's settings if the user still wants out.
inside that flow:
1. a short reason survey (too expensive / don't use it / missing feature / switching to competitor). branches the dissuader.
2. a dissuader matched to the reason. 'too expensive' → 50% off, escalating to 80% if they skip. 'don't use it' → 30-day pause + content recap. 'missing feature' → what's shipping next
3. a 'here's how much value you're missing out on' screen. specific numbers: saves, content finished, progress, features paid for.
4. only then, the deep-link to settings.
real examples:
- Betterme: tiered discounts (50% then 80%), then enumerates lost content before the final step.
- Noom: no-commitment $19/mo plan at ~65% off, pause option, plus a cross-sell to their glp-1 program.
- Strava: switch monthly to annual at a lower effective monthly rate. structural discount, no coupon.
- headspace: surveys 'too expensive / don't use it / other' before any retention offer.
even in the app store world you can still build out custom cancellation flows.
This is an example of an actual prompt I left form one of my PM bots. At the same time I prompted two other PM bots something similar. I then proceeded to go mow my grass and putz in the yard for several hours. They pinged me a couple of times but diligently worked off their task board.
This is how I work every day. I’ll chat with the bot team to scope and discuss features, etc. but mostly they just work autonomously. One of the first things I built was an agent-first kanban board for task tracking. The bots have full control via the built in MCP. So they can just work off the board endless, creating their own tasks, etc.
https://t.co/BljJO4zz2I