Transcript
**0:00** · App store screenshots are the slowest and most painful part of releasing an app. So, I built a Claude skill to do it for me. Instead of spending days designing, this thing does it in 15 minutes. It doesn't just generate any old screenshot. It follows the exact formula I've been using to grow my apps to $70,000 per month. In this video, I'm going to show you my exact workflow, how I design high-converting App Store screenshots, and how you can now automate the entire process with Claude code. Let's go. This is how I was designing screenshots before, in Photoshop, pixel by pixel. Not only was I doing the actual design stuff, adding iPhone device frames, and choosing the layouts, but I was also brainstorming each of the headlines for the screenshots, because they need to be descriptive enough to know what the app is about, but not too text-heavy. This could sometimes take me hours or even days to get right. With agentic coding, that can feel like a lifetime. It's now quicker to generate the entire app itself than it is to create those screenshots. So, I set out to automate the entire process. First, I had to figure out exactly what I was doing, and I had to put that in words. It's not enough for a screenshot to look pretty, it needs to actually convert users. Most indie app developers fall for this trap.**1:18** · They think that their screenshots are just a bunch of features. Screen one, here's feature A. Screen two, here's feature B. Screen three, by the way, did I mention feature C? But that's actually the worst thing you can be doing. It's boring, nobody's going to read it, and nobody's going to download your app. To convert more downloads, you need to think about it a little bit differently.**1:40** · You need to tell the user why they should care. They got to do it as quickly as possible. To actually stop the user while they're scrolling the App Store, you need to signal that this is the app that solves their problem, that this is the app that they're looking for. Do that, and you win the App Store.**1:58** · Every screenshot I create starts with one single action verb. Track, search, boost, learn, find. It's big, it's bold, and you cannot miss it, followed by the benefit. Not view prices online, but track trading card prices. Not piano note quiz app, but learn piano notes.**2:19** · The verb grabs the attention, and the benefit sells it. And this also needs to work if you're just scanning the headlines on each screenshot. Learn, read, play. Yes, there is a lot going on here, but all you got to remember is verb, benefit. Pair that with a solid color that complements your app, and you're guaranteed to get attention. This structure alone has done more for my downloads than any amount of time tweaking designs in Photoshop. But it goes deeper than coming up with a compelling headline. Before I even think about the design, I ask myself these three questions. Who is the user? What's the number one reason they'd be downloading this app? And what transformation does this app deliver? A habit tracker doesn't just track habits, it makes you better every single day. A recipe app doesn't just show recipes, it makes cooking at home effortless. You need to nail the transformation. Do that, and the headline text just writes itself. Then, you pair each screenshot with a screenshot from your app that best represents that headline. But if it's that simple, why not just do it?**3:25** · Most indie app developers have difficulty switching from benefits to features, and I get it. We're so caught up in the code, so caught up in the what it does, that we forget to explain the why. Without going just that single step deeper, screenshots just end up looking like an afterthought. I've done the app, now all I got to do is slap on some screenshots and ship it to the App Store. And if that's you, then I've got the Claude skill for you. For the one-time payment of $0 and $0.00, you can now generate high-converting App Store optimized screenshots for your apps without using your precious, precious brain. Because after all, you've got more important things to do.**4:04** · Like working on those three other side projects you're vibe coding in parallel.**4:08** · They're not going to build themselves.**4:10** · Oh, wait. You run the skill, it checks your code, and it asks you some basic questions about your target market.**4:15** · Then, it does the magic, generates the headlines using the verb-benefit formula, and selects the best, most vibrant color that matches your app. But here's where it goes a little bit further. This Claude skill will help you take the screenshots. It provides recommendations for each app screenshot, down to the exact screen it recommends.**4:35** · Then, it assesses the quality of your screenshot and will provide you some guidance and help on taking the best shot. Sure, you could ignore all of this and just get to the juicy part, but seriously, take a few minutes here, this could change the entire trajectory of your app. The final version is generated in around about 15 minutes. This skill learns about your app, its target market, it scaffolds the design, then it sends it over to Nanobana for the high-quality processing. This ensures every screenshot is consistent, works as a set, and not only follows a high-converting formula, but is also compliant with the App Store requirements, down to the exact pixel.**5:13** · And this skill has been saving me hours of work with every new app that I build.**5:18** · So, now I'm releasing it for free, so you can check it out, too. I'll put a link in the description below. Feel free to test it out for yourself and show me what the results look like. I've tested this on my own apps, so it'll be interesting to see what others look like. And drop a link to your app in the comments below, and I'll see what I would change. Screenshots used to be that thing that I dreaded the most about releasing a new app. Now, it's the fastest part. If this helped, I'm working on something even bigger, a Claude skill that builds your entire onboarding flow, taking the formula that works, encoding it as a skill, and letting AI do all the heavy lifting.**5:57** · Subscribe if you want to see that one.