Feature Request: Player-Facing Image Generation for Custom UI
Oncin ·
Could you expose image generation to Custom UI through useYumina()?
I’d like players to be able to generate visuals directly during gameplay, for things like character portraits, NPCs, locations, important story scenes, and campaign moments, without creating or advancing a story turn.
Ideally, the API would support:
a text prompt
aspect ratio
optional reference image for edits/consistency
player-side mushie billing/confirmation
generation status/progress
returning the finished @asset:id or image URL once complete
Then the generated image could be used normally inside Custom UI through resolveAssetUrl() or another asset helper.
For my use case, I’m building an in-game campaign Gallery where players can attach portraits to characters, images to locations, and illustrations to specific story turns. Custom UI can already present and organize those visuals, but currently there doesn’t seem to be a documented player-runtime way to actually generate them.
Having a supported runtime image-generation API would open up a lot of possibilities for more game-like Custom UIs, especially visual novels, RPGs, simulators, character systems, journals, and illustrated story experiences.
Not asking for unrestricted background generation either. A confirmation step showing the mushie cost before generation would be completely fine, and probably preferable.