What problem does it solve?
This Skill removes the manual, iterative work of designing consistent visual assets for an AI 文游 by generating concept art, character illustrations, and interface background images from project context and a controlled prompt flow.
Core Features & Use Cases
- Asset entry point routing: Use one slash command to trigger the correct asset pipeline for characters, interface backgrounds, or cover/key art.
- Vibe-first checkpoint: Produces a short 50–100 character/word prose “气质中介层” description so the user can confirm composition, lighting, and first-impression mood before generation.
- Style-consistent generation: Converts the confirmed vibe into Danbooru-style tags while always prepending the project’s
visual_style_tags to prevent cross-style drift.
- NovelAI image generation workflow: Calls the project’s NovelAI generator with controlled iteration defaults and then saves outputs into deterministic target locations.
- Canonicalization for downstream use: Promotes the user-selected best result to a stable reference (
concept_canonical.png) for later illustration variations.
Quick Start
Ask: "/aiwy-asset concept" to generate key art for the project cover using the pipeline instructions in aiwy-design.