Loading...
Loading...
Found 2 Skills
Generate actual reference images for characters from any story (novel adaptations, original stories, standalone character designs, all acceptable - no original novel text required): Describe the character in a single paragraph, which will be split into hierarchical fields, supplemented, and then confirmed. First generate a front-facing full-body anchor image; all other views (close-up portrait, 90° side view, back view, detail shots, 45° close-up portrait) only reference this anchor image. Generate images in tiers as needed. Each image has an identifier, can be regenerated individually, and expired images are automatically marked after regeneration. Supports Qwen Image (ComfyUI), codex built-in image generation, OpenAI Images API (GPT Image 2), and custom commands. The main image and derived images can use different models. Outputs asset.json + a set of PNG files + a double-clickable report. Zero dependencies. For prompt-only generation without actual image output, use novel-characters. Use when asked to generate character images, create character reference sheets, generate images for my character, makeup test shots, generate three-view drawings, character reference images.
Use when building any val with a user interface — dashboards, web apps, landing pages, forms, admin tools, anything users see in a browser. Covers JSX/React conventions, Twind/Tailwind styling, React version pinning, the view-source link requirement, and what to avoid (template-string HTML, external assets).