What problem does it solve? Creating a consistent personal IP character, avatar, and matching expression sticker pack from a real photo or character brief normally requires repeated prompt tuning and manual style control. This Skill runs a strict one-question-at-a-time wizard that locks identity anchors, style, traits, and palette before any image is generated, so every output stays visually consistent. ## Core Features & Use Cases - Guided Plan-Mode Wizard: Asks exactly one question per turn to resolve IP type, source material, style, enhanced traits, palette, and prompt-lock approval before generating anything. - Preset Style Catalog: Ships five active style presets (彩绘风, 线条风, 3D风, 插画风, 手办风) with full style fingerprints, plus a task-scoped custom style mode using user-supplied reference images. - Structured Deliverables: Produces one numbered 3x2 six-variant preview, then a 4x3 final board with a formal avatar and nine Chinese-captioned expressions, with optional transparent-background sticker export. - Use Case: A creator uploads two portrait photos, picks the 3D风 preset, confirms the prompt lock, chooses candidate 4 from the preview, and receives a unified avatar plus a nine-expression WeChat sticker board. ## Quick Start Ask the assistant to use the personal-ip-generator skill to create a personal IP avatar and expression pack from your uploaded portrait photo, answering one question at a time.
