I’ve been building Shotluma, an open-source editor for creating iOS App Store screenshots with the Power of Vercels AI-SDK.

The main idea is pretty simple: I didn’t want the AI to generate a flat image that looks nice but becomes annoying the second you want to change something.

In Shotluma, the AI works inside the actual editor. It can add text, gradients, shapes, screenshots and device mockups, but those things stay normal canvas elements afterward. You can click them, move them around, resize them, change the text or colors, or just fix something manually.

It can also build a whole screenshot set instead of generating one screen at a time.

And if one screen is almost right, you don’t have to regenerate everything. You can select that screen and ask the AI to change only that one. For example, you could ask it to make the headline bigger or move the device around, then continue editing the result yourself.

I also wanted the editor itself to work without an account or backend. Projects and uploaded images are stored locally in the browser. You don’t need an API key for normal editing either.

If you want to use the AI features, you bring your own API key. Shotluma currently supports providers including OpenAI, Anthropic, Google, Qwen and Moonshot. The requests go to the provider you choose, so any API costs are between you and that provider.

Once you’re done, you can export the complete set as PNGs in a ZIP. Right now it supports 1290 x 2796 and 1242 x 2688. The project is MIT licensed and the source is on GitHub. It’s still early, so I’m posting it here partly because I’d like to hear from people who actually have to make these screenshots for their apps.

What do you currently use for App Store screenshots? And if you try Shotluma, I’m especially curious where the editor gets annoying or confusing. That’s probably more useful to me right now than compliments

Check it out: https://shotluma.com