Hacker News Reader

Comment by tapeface | Hacker News Reader

Simple utility apps like this are definitely going to be under threat by AI-driven app creation. Unsatisfied with the free options out there, I've been using Claude to build my own workout tracker app with nearly the exact same Live Activity functionality in addition to the other features most apps lock up behind IAP subscriptions. Within a week I had the iOS and macOS apps up and running with CloudKit sync, analytics, custom routines, etc. I'd never made an app before, didn't know Swift/SwiftUI, and barely had to use Xcode.

View on HN

Workout trackers are especially at risk. Every single one I’ve tried is terrible. They are either really poorly thought out, bad UX, or charge 40/mo and still suck.

I built a simple workout tracker/planner in an afternoon using an AI tool that is exactly what I want and free.

I have been working on this app for a while, trying to create a good UI and good integration with Apple Watch.

As well as live activity, haptic feedback, animations, fast local database with search that is super fast (it’s actually really fast with fussy matching algorithm so even if you type the exercise wrong it will find it instantly)

And if you have any feedback I’ll make sure to fix it!

I'm not the right person to give feedback here. I'm not in your target market. For me, social features are a negative and I don't want to use an app that makes them center stage or even promotes them. Live activity, haptic feedback, animations, etc... are not things I want.

The app I built is simple. It has a list of major exercises and then a map of supporting accessory exercises. It will create a plan that is as long as I'd like for as many weeks as I'd like with a specific number of workouts per week alternating between upper and lower body.

Everything else is fluff I don't want. I want an app that is only a workout planner and tracker app. All the rest of it is a distraction and not useful for how I use apps and workout.

For me, the point of building this app is that I do not want to pay a subscription or any fee, and I want exercises / workouts / routines to be simple. The rest of it is noise, for me.

I get that, for Musklr I made signing in optional, social features completely optional. Works great offline. Caches animations and saves everything locally. So can set the app in flight mode. Just use an Apple Watch if you want. And everything will still work.

I have built one that focuses on enabling me to have training systems that are not possible to re-create in most apps. I’ll release it as free, even if I would decide to charge for it, I would probably offer a generous trial and a lifetime subscription to the major version you bought.

If you just have the name of an exercise and type in the amount of kg and reps. That’s just a TODO app++. Easily created with Claude in an afternoon

This app however, with thousands of exercises, each exercise mapped to the correct muscle group and equipment. Well tested integration with Apple Watch. Social features, like posts, follows, likes, public and private accounts. That’s not made that easily with AI.

Most of those extra features sound like the clutter and cruft that I actively avoid and don’t want in my app. Which is why it’s great that AI can make the app that only has what I want

I think this app is a great example of AI-driven app creation!

For sure! What I'm wondering is why anyone would pay a subscription for someone else's AI-created app if they can spin one up for themselves with only the parts they want/need. The tools aren't there yet, but within 3-5 years I could see native iOS support for general users to create apps via Siri.

Do you have a testflight? I'd love to try it!

Not external, sorry - I haven't bothered putting it through app review or anything.

You don't have to put a testflight through app review. Example: https://reverdure.yourstrategy.co

Reply on news.ycombinator.com