Multi-outfit concepts often lose character identity and motion continuity between images and video. This skill creates linked prompts, layouts, transition timelines, and consistency constraints.
Review the Skillstore skill "female-outfit-director" from https://skillstore.io/skills/liyue-aigc-female-outfit-director.md and its manifest at https://skillstore.io/api/skills/liyue-aigc-female-outfit-director/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.
エージェントは引き続き計画を提示し、セキュリティポリシーで必要な確認を求める必要があります。
エージェントが読めるリソース
AI エージェント、クローラー、スクリプトがページ全体ではなく整理されたコンテキストを必要とする場合は、これらのリンクを使ってください。
「female-outfit-director」を使用しています。 Create a cool blue Hanfu collage with five looks and gentle transitions.
期待される結果:
Locked direction: one adult Chinese woman, five coordinated blue and white Hanfu looks, a 9:16 studio collage, and a fixed high camera. The timeline uses figure overlay and sleeve-cover transitions at four defined beats, with continuous posture and natural fabric motion.
「female-outfit-director」を使用しています。 Keep my modern lookbook layout but make every transformation feel faster.
期待される結果:
Option one uses full-figure sticker overlays for clear visual causality.
Option two uses page-flip silhouettes for stronger collage styling.
Option three uses step and hand-gesture beats for direct rhythmic changes.
「female-outfit-director」を使用しています。 Use this reference image to create four additional outfits.
期待される結果:
The result preserves visible facial features, hair, accessories, body proportions, lighting, and composition. It adds four compatible outfits, marks uncertain details as system-completed, and links every outfit to a timed transition.
All eight static findings are false positives. Backticks only delimit Markdown code spans or a fenced example, while the high-entropy files are readable Chinese Markdown without obfuscation.
liyue-aigc. (2026). female-outfit-director security audit report (audit version 2) [Author version 1.3]. Skillstore. https://skillstore.io/skills/liyue-aigc-female-outfit-director/audits/2
BibTeX形式の引用
@techreport{liyue-aigc-liyue-aigc-female-outfit-director-2026,
author = {liyue-aigc},
title = {female-outfit-director security audit report (audit version 2)},
institution = {Skillstore},
year = {2026},
number = {2},
url = {https://skillstore.io/skills/liyue-aigc-female-outfit-director/audits/2},
note = {Author version 1.3}
}
CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "female-outfit-director security audit report (audit version 2)"
version: "1.3"
type: report
authors:
- name: "liyue-aigc"
date-released: "2026-09-09"
url: "https://skillstore.io/skills/liyue-aigc-female-outfit-director/audits/2"
identifiers:
- type: other
value: "skillstore:liyue-aigc-female-outfit-director:audit:2"
description: "Skillstore immutable audit report identifier"
Skillstore スコア
このスコアの理由証拠の信頼度: 中
41
アーキテクチャ
85
保守性
85
コンテンツ
65
コミュニティ
83
仕様準拠
作成できるもの
Plan a Fashion Lookbook
Turn five modern outfits into a consistent collage prompt and an eight-second transformation timeline.
Extend Character Reference Art
Extract a character appearance and layout from an uploaded image, then plan coordinated outfit variations.
Direct Hanfu Transitions
Select sleeve, fan, ribbon, or figure-overlay transitions with fabric physics and precise timing.
これらのプロンプトを試す
Create a Basic Outfit Sequence
Create a five-outfit collage and an eight-second transformation script for one adult female character. Use a fixed 9:16 camera and recommend missing details.
Compare Style Directions
Recommend five distinct parameter sets for an adult female character. Include outfit logic, palette, collage layout, camera, movement style, transition mechanism, and timing.
Extend a Reference Image
Analyze the uploaded reference image. Preserve the visible adult character, hair, accessories, clothing details, and layout while creating four compatible outfit variations and a linked video timeline.
Redesign an Existing Transition Script
Keep the character, outfits, camera, duration, layout, and timing from my script. Replace only the transition mechanism with three suitable alternatives and explain continuity requirements.
ベストプラクティス
Provide the character concept, outfit count, visual style, duration, and aspect ratio when available.
State which details are locked and allow the skill to complete only low-impact missing parameters.
Use clear reference images showing the face, hair, accessories, clothing, and body outline.
回避
Do not use this skill for a single portrait without a multi-outfit transformation goal.
Do not combine complex camera movement with several elaborate transitions in a short timeline.
Do not request inconsistent identity, body structure, or camera changes between outfit stages.
よくある質問
Does this skill generate images or videos?
No. It produces structured image prompts, video timelines, concise video prompts, and negative constraints for compatible generation tools.
Can it use an uploaded reference image?
Yes, when the host agent supports image understanding. It extracts visible character, outfit, accessory, layout, lighting, and composition details.
Must I provide every parameter?
No. The skill preserves supplied details and completes low-impact gaps with documented defaults.
What outfit styles does it support?
It documents Hanfu, historical Chinese, modern fashion, lookbook, and Eastern fantasy directions. User-specified styles remain locked.
Can I change only the transition method?
Yes. It can preserve the character, outfits, camera, duration, layout, and timing while replacing transition triggers and related actions.
How many outfits can it plan?
The default is five looks. It can adapt the layout for three to six looks when the user specifies another count.