Nuyoah Xiezhen Prompt skill
南鸢写真提示词
Portrait photography in Chinese: prompts, series, gacha keywords and reshoots with a new subject from the same plan.
- 786 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- On a direct request the skill generates images through Codex Image and, if available, saves them to Eagle.
- References with people remain your responsibility: the skill does not check image and portrait rights.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/nuyoah-xiezhen-promptDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- On a direct request the skill generates images through Codex Image and, if available, saves them to Eagle.
- References with people remain your responsibility: the skill does not check image and portrait rights.
This is third-party code. Review the repository files before installing.
What it does
Nuyoah Xiezhen Prompt writes detailed Chinese prompts for portrait photos and editorial shoots. The skill picks a mode from how the request is phrased: a full prompt, a series master template, variants for one series, short gacha keywords for web interfaces, a reference breakdown into features, or a reshoot. Reshoot mode takes your reference photo series, recovers the shooting plan from it (light source, camera position, exposure, background hierarchy, the event with the subject, texture) and writes new frames with a different subject. The skill creates images through Codex Image only when you explicitly ask it to generate them, and checks each frame. On every new round it starts again from the original photos and uses the previous result only to diagnose mistakes. A signature or watermark is added only at your request.
Who it is for. Authors of portrait series and prompt designers who work with Chinese generative services and run their own photo projects.
Good fit when
- The theme of a portrait series is set and a detailed prompt has to be written for it.
- A web interface needs a batch of short keywords to spin the generator like a lottery.
- You have a set of references and want a similar series shot with a different subject.
- A finished photo has to be broken down into light, exposure, framing and textures.
- A ten-frame series needs one master template and variants.
Not a fit when
- You need a face swap on someone else's photo: the skill creates new subjects and does not carry over identity.
- You lack rights to the references or to the likeness of the people in them.
- You need English text: the prompts and documentation are written in Chinese.
Example request
`使用 $nuyoah-xiezhen-prompt`, then: take this series, change the subject to an original one and design five new frames without generating.Limitations
The skill does not swap faces and does not promise pixel-level copying: it recreates the observable shooting plan, and the result depends on the model, reference quality and permission for the subject. Before using references, check image rights, trademarks and platform rules. If the environment has no image generation tool, the skill delivers ready prompts and says generation did not run. The Eagle archive is optional. By default the subject is an adult woman; the skill writes no minors or childlike looks. The last commit was made in August 2026, version 1.1.0.
How to disable. Remove it through the skills manager or delete the `nuyoah-xiezhen-prompt` folder from the agent's skills directory. Images saved to Eagle remain there.
FAQ
How do I install and update the skill?
Install: `npx skills add nuyoah-ai-works/nuyoah-xiezhen-prompt -g -y`. Update: `npx skills update nuyoah-xiezhen-prompt -g -y`. After updating, open a new task so the agent loads the fresh instructions.
Does the skill create images by itself?
No. A regular request returns only a prompt. Images appear in reshoot mode when you explicitly ask for generation, and they are created through Codex Image.
What is a reshoot?
You provide a photo series, and the skill breaks down how it was shot and designs new frames of a different subject in the same manner. The face from the reference is not carried over: an original adult character is created for the new frame.
Is Eagle required?
No. If the app is available the skill also saves the result and names the item ID. If not, the prompts, parameters and check record stay in the chat.
Will the images carry signatures?
Not by default. The skill adds signature text only if you ask for it and supply the words yourself.
Discussion
No questions or reviews yet.
Related
SkillsQwen-Image 2.1 Prompter
qwen-image-2.1-skill
A skill that rewrites prompts for Qwen-Image-2.1 by Alibaba's official rules: text-to-image and multi-image editing
Female Portrait Director
Female Portrait Prompt Director Skill
A skill that turns a short set of parameters into a photography-directed portrait prompt for an adult woman, plus negative constraints
A Codex skill that turns a photograph into a vertical composition with an abstract memory panel and an English title
image-prompt-reverse analyzes an uploaded picture and writes a generator prompt together with a negative-term list.
Portrait skill that makes a snapshot cinematic while keeping face and scene, or builds an original series from scratch.
A skill that turns photos of buildings, streets and local culture into art-directed editorial posters via an image model
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Repository stars | Main difference |
|---|---|---|---|
| Nuyoah Xiezhen Prompt (this card) | no | 786 | |
| Female Portrait Director | no | 1,704 | assembles one adult female portrait prompt from a short parameter set; this one adds series, gacha keywords, reference breakdown and plan-based reshoots |
| Photo Abstract Editorial | no | 6,118 | turns a photograph into a vertical composition with a memory panel; Nuyoah Xiezhen Prompt writes prompts and runs plan-based reshoots |
README in short
The README covers the working modes, signature rules and capability boundaries, and lists the repository structure with reference cases in `evals/` and contract tests. Detailed rules are in `SKILL.md`.
README badge
Original SKILL.md
--- name: nuyoah-xiezhen-prompt description: 当需要生成、改写、扩展或反向拆解 AI 写真、人像摄影、editorial photography 中文提示词时使用；也用于用户提供一组写真参考图，希望换一位人物、按同一摄影企划重新拍摄新分镜，并在明确要求生图时完成 Codex Image 生成、检查与可选归档。普通提示词请求默认不主动生图。 license: MIT metadata: author: "南鸢 nuyoah" version: "1.1.0" homepage: "https://knowledge.nuyoahonline.com/skills/nuyoah-xiezhen-prompt" source: "https://github.com/nuyoah-ai-works/nuyoah-xiezhen-prompt" --- # nuyoah-xiezhen-prompt ## 目标 把写真意图编译成可直接使用的中文 Prompt；当用户明确要“换个人拍同一套写真”并要求生成时，把参考组还原成可执行的摄影方案，在 Codex Image 中生成指定数量的新照片并完成质量检查与可追溯交付。Eagle 只在当前环境可用时作为附加归档，不是完成任务的前提。 ## 入口判断 先选择一种模式： - 用户提供一组写真参考并表达“照这套拍、换个人拍一套、像去同一家店拍、复刻这套写真企划”等意图时，进入**摄影方案复拍**。读取 [摄影方案复拍工作流](references/production-reshoot.md)；只有用户明确要求生图才执行完整生成闭环。 - 用户说“系列母版、一整套系列、系列规划、固定格式”时，输出系列母版。 - 用户只说“变体、同系列、创作变体、仿拍扩展”，但没有表达完整摄影企划复拍意图时，输出同系列变体关键词链。 - 用户说“完整、详细、精准控制、复刻 Prompt”时，输出详细提示词。 - 用户说“抽卡、网页端、大量生成、短关键词、开盲盒”时，输出 6–8 行抽卡关键词。 - 用户给图片并要求拆解时，输出图片关键词；组图先提取共同视觉 DNA，再列单张差异。用户明确“只拆解、不生图”时不得进入生成闭环。 - 用户给出 `主题：`、人物类型、`关键词：` 三段骨架时，直接按 [提示词逻辑与模板](references/prompt-logic.md) 的三段式规则编译，不依赖包外模板。 - 用户说“文化视觉溯源、文化元素、往外扩展灵感”时，只根据当前可见证据区分事实、强关联和灵感联想；需要外部史实或来源时先真实检索，不依赖私有知识库方法。 普通提示词、变体、拆解与模板规则读取 [提示词逻辑与模板](references/prompt-logic.md)。维护或回归验收本 Skill 时，同时读取 `evals/skin_reflection_cases.json` 与 `evals/production_reshoot_cases.json`。 ## 参考图职责 1. 人物图承担身份还是摄影参考，必须先分清。用户另外提供人物形象图但没有说明用途时，只问一次：它是要保持同一位人物，还是只借人物气质？ 2. 人物身份参考只回答“她是谁”，保持主要五官辨识度；目标写真接管妆容、发型、服装、场景、道具、光线和情绪。 3. 写真方案参考只回答“这套写真怎样被拍出来”，不承担人物身份。Prompt 必须正向写入：`本张重新生成一位「【原创女性类型】」类型的成年原创女性，不要继承写真参考图中人物的脸。` 4. 没有人物身份图时，根据写真类型自动替换 `【原创女性类型】`，以人物好看、动作自然和画面成立为优先；不检查组内是否为同一张脸，也不增加生成脸与写真参考人物逐一比较的验收层。 ## 共享原则 1. 先判断照片为什么成立，再写细节。光线必须写清来源、方向、在人身与环境中的落点和成像结果；更换机位时，光源、遮挡物与环境反射固定在世界空间中，先重建新视角下的受光关系，再编译 Prompt。 2. 风格写成可观察的成像机制，不用“高级、氛围感、电影感”代替白平衡、曝光、反差、高光、锐度、噪点、压缩和景深。 3. 穿搭只写服装，饰品放入道具；场景与道具分开。 4. 默认人物为成年年轻女性，强调成年后的青春、灵动和生命力；不写未成年或儿童化表达。