错层照片带合影 · Staggered Photo Ribbons
将照片主体转为错层照片带合影，宽幅照片带有限错位，人物连接仍可辨。保留可辨身份与主要空间关系。
A group photograph cut into offset horizontal image ribbons on restrained paper space. Offset a few wide source-photo ribbons by small amounts while keeping complete faces, body connections and overall group order readable.
本方法改编自 VigoZhao 的 MIT 授权说明。原作者归属、来源与改动见 provenance.json，完整上游许可及其版权声明见 LICENSE-THIRD-PARTY.txt。
Adapted from VigoZhao's MIT-licensed instructions. provenance.json records the original author, pinned source and modifications; LICENSE-THIRD-PARTY.txt retains the complete upstream license and copyright notices.
原始归属 · Original attribution: Copyright (c) 2026 VigoZhao Adapted by MuziGeek from VigoZhao; original-post attribution is retained where the collection provides it.
输入分析与实际保留
- 读取用户提供的原图；若没有可用图片，先请求图片。不要用本仓库示例代替用户输入。
- 观察原图主体、身份、数量、连接、遮挡、光线和画幅。依据本 Skill 的保留约束与几何边界，区分需要保留及明确允许转化的部分。样片说明不能当作用户输入的事实。
- 使用支持参考图编辑的可用工具，把原图作为编辑输入。先说明观察到的主体，再结合下方机制及 recipe.yaml 按顺序组合指令；工具名称不是必须安装的依赖。
- 默认沿用原图比例；仅按用户要求及本 Skill 明确写出的构图或几何转化调整。配方的 4:5 是目录样片比例，不授权裁去必须保留的主体。
- 对照输入检查结果；只修正发现的问题，不为了增强风格改变身份或添加无关景物。区分已完成的视觉检查与尚未验证的效果。
Input analysis and preservation
- Inspect the supplied image; request an image only when none is available. Never substitute a repository example for the user's input.
- Inspect the real subjects, identity, counts, attachments, overlaps, light and framing. Distinguish preserved attributes from explicitly permitted transformations using this Skill’s preservation and geometry rules; sample descriptions are not facts about user input.
- Use an available reference-image editing tool. Combine the observed facts with the ordered mechanisms in recipe.yaml; its model-family label does not require an installation script.
- Keep the input ratio by default; change composition or geometry only as requested by the user and expressly permitted below. The recipe’s 4:5 ratio is a catalog-sample default, not permission to crop protected subjects.
- Compare output with input and correct observed failures. Report only visual checks actually completed; stronger styling must not override identity or scene integrity.
风格机制 · Style mechanism
- 将照片主体转为错层照片带合影，宽幅照片带有限错位，人物连接仍可辨。保留可辨身份与主要空间关系。
- 识别特征：宽幅照片带、有限位置错移、纸片浅投影。
- Medium anchor: A group photograph cut into offset horizontal image ribbons on restrained paper space.
- medium — A group photograph cut into offset horizontal image ribbons on restrained paper space.
- composition — Offset a few wide source-photo ribbons by small amounts while keeping complete faces, body connections and overall group order readable.
- texture — Use crisp cut edges and shallow paper shadows; no added duplicate heads or unrelated photographs.
- palette — Use a controlled palette appropriate to the requested medium; retain identifying source colors whenever the mechanism permits.
- lighting — Keep illumination internally consistent across the entire edited image, with contact shadows appropriate to the requested medium.
- detail — Retain source subject count, essential openings and connected anatomy; remove no identifying detail unless the permitted transformation requires it.
- finish — Deliver one finished image at the input aspect ratio, without comparison panels, new lettering, logos, watermarks or an external frame.
保留约束 · Preservation constraints
在实际输入观察结果之上应用以下约束；清单不授权凭空添加图片中不存在的元素。
Apply these constraints to observed content; they never authorize inventing absent subjects.
- Source subject count, recognizable identity, main pose and connected anatomy, subject to the explicitly permitted transformation.
- Essential openings, attachments and foreground-to-background relationships.
- The complete main subject within the input frame; no accidental clipping.
适用范围与几何边界 · Applicability and geometry
- 适用：合影。
- Applies to: Group.
- 允许宽幅照片带之间小幅错位；面部保持完整，身体连接与群组次序仍可辨。
- Permit small offsets between wide photo ribbons; keep faces complete and retain readable body connections and group order.
- 未明确允许的结构变化均不属于此风格；不把这些方法解释为复刻某位艺术家、品牌或现成作品的授权。
- Structural changes not explicitly allowed above are outside this method. This is not permission to reproduce a named artist, brand or existing artwork.
排除项 · Avoid
- Invented people, extra limbs, duplicate main subjects or disconnected anatomy.
- Unrequested logos, decorative writing, watermarks or fabricated technical claims.
- A generic filter that fails to implement the specified physical or optical mechanism.
验收 · Acceptance
- 宽幅照片带有限错位，人物连接仍可辨。 Use crisp cut edges and shallow paper shadows; no added duplicate heads or unrelated photographs.
- 允许宽幅照片带之间小幅错位；面部保持完整，身体连接与群组次序仍可辨。 Permit small offsets between wide photo ribbons; keep faces complete and retain readable body connections and group order.
- 主体清楚可辨，连接自然；效果必须体现具体机制，不能只是通用滤镜。 The subject remains legible with coherent connections; the requested physical or optical mechanism is visibly implemented.
- 只交付一张完整成图，无无关文字、品牌、边框或虚构测量数据。 Return one finished image without unrelated lettering, brands, frames or fabricated measurements.
配方、来源与许可 · Recipe, provenance and license
- recipe.yaml: structured image-editing recipe / 结构化图像编辑配方。
- provenance.json: original authorship, method context and example status / 原创归属、方法背景与示例状态。
- LICENSE: MIT applies to MuziGeek's original contributions / MIT 适用于 MuziGeek 的原创贡献。
- LICENSE-THIRD-PARTY.txt: VigoZhao · MIT. Adapted material retains these terms and attribution / 改编材料保留原许可和归属。
- 示例中的前后图均为 AI 合成，不是真人摄影或真实制作工艺的证据。结果会随模型与输入变化，不承诺相同输出。
- Example before/after images are AI-generated, not photographs of real people or evidence of physical craft production. Results vary by model and input.
- MIT 不授予用户输入、第三方作品、人物、商标或生成平台条款之外的权利；示例许可单独见 provenance。
- MIT grants no rights to user inputs, third-party works, likenesses or trademarks, and does not override generation-service terms; example terms are recorded separately in provenance.
示例状态 · Example status
ready：原图与效果图的真实生成及视觉审核记录齐全；COS 外链示例另保留上传回读凭证，不在仓库保存图片。
ready: the source and effect have recorded real generation and approved visual reviews. COS-hosted examples also retain upload readback receipts; their image bytes are not bundled.
| AI 合成原图 · Source | 通过审核的效果 · Approved effect |
|---|---|