This post is my submission for DEV Education Track: Build Apps with Google AI Studio.
What I Built
I used Google AI Studio’s “Build apps with Gemini” feature to create Arcane Forge, a small generator that creates D&D magic items with full descriptions, lore, mechanics, and an image.
My main prompt told Gemini to:
- generate a magic item based on a chosen type (weapon, armor, wondrous item, potion, etc.)
- adjust the item’s rarity
- include price ranges based on Xanathar’s Guide to Everything
- output a clean description + lore block + mechanical features
- feed a short visual prompt into Imagen to generate an item image
I added a few extra knobs (themes, visual styles, power levels, and optional curses), but the core idea stayed simple: click a button → get a usable magic item.
Demo
Here’s a screenshot from my build:
You can try it here:
→ https://arcaneforge.netlify.app/
My Experience
This was the first time I used AI Studio’s “build an app from a prompt” workflow, and the whole thing went from idea to running code in a couple of minutes. The build wasn’t perfect on the first try, but it gave me a complete starting point: React components, API wiring, and a structure that made sense. From there, it was just normal cleanup and tweaking.
A few things stood out to me:
- Imagen was easier to use than expected. Feeding text from the item description into the image generation prompt produced surprisingly coherent results.
- Studio handled the heavy lifting. I didn’t write scaffolding code — it generated the Vite + React project for me.
- Small prompt changes had big effects. Adding pricing rules from Xanathar’s gave the items more grounding than “AI magic items” normally have.
- The build feature is great for prototyping. Not everything needs a blank repo and a weekend. This was “idea → working UI” almost instantly.
Overall, this was a fun project and a good excuse to experiment with the new tools. I’ll probably keep expanding Arcane Forge — maybe adding monster generators or spellbooks next.

Top comments (0)