WHAT I'M WORKING ON
Adding iPhone Duo support to my Video Compression App
After releasing my Video Compression iOS app last week, I started working on support for iPhone Duo. Just like I can use RocketSim and RocketTrace for my own app now, I’m also able to tune and optimize my SwiftUI Agent Skill.
But I wondered: adding iPhone Duo ánd iPad support, where should I start? Would it be better to make my iPhone resizable first? Or should I add iPad support first, then fine-tune for resizability and iPhone Duo?
I decided to start with iPad support. I used a cheap agent model (Composer 2.5 in Cursor) to walk through all iPad screens using RocketSim’s CLI, and asked it to create a web page with screenshots for each page in dark & light mode, at different text sizes. Each screenshot comes with a text field for feedback and a “copy prompt” to return everything at once to the agent.
Am I going extreme? I don’t know. I do know this has been a quick way for me to get an overview of all the pages in my app while I work on something else in the meantime. Token-wise, it’s also doable, as RocketSim’s CLI is quite efficient. I believe it took about 1.8 million tokens for the complete first setup. After adding iPad support, I’ve also been able to add Duo support more easily.
What’s great about this all is that I can ask the agent for feedback on my SwiftUI Agent Skill. That resulted in several updates over the past weeks, so make sure to run npx skills add https://github.com/avdlee/swiftui-agent-skill --skill swiftui-expert-skill in your terminal, and upgrade your agent’s SwiftUI knowledge to support iPhone Duo.