Oh My Cover Design skill
A skill reads your article, asks about style and outputs a ready prompt for a Xiaohongshu or WeChat cover in 10 compositions
- 274 ★ on the repository
- Last commit
Minimal access
Why this level
- Instructions only: the skill runs nothing and makes no network calls
- Asks for your photo as a reference, but the picture stays in your conversation with the agent and goes to a generation model at your decision
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/oh-my-cover-designDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewMinimal accessThe tool runs nothing and makes no network calls.
- Instructions only: the skill runs nothing and makes no network calls
- Asks for your photo as a reference, but the picture stays in your conversation with the agent and goes to a generation model at your decision
This is third-party code. Review the repository files before installing.
What it does
The cover-design-open skill builds a prompt for an article or post cover: you send the agent your text, it reads it and asks questions one at a time. First you pick one of ten compositions, from a dark gradient and flat colour to a collage and direct eye contact. The agent then asks whether you have your own photo for the first reference image (in the prompt it anchors facial features) or a description of appearance, and lets you pick a facial expression from a list, extra materials such as product screenshots, background colour tone, font and text colour. It proposes the headline itself, one to three options of 4 to 8 characters. The output is a text prompt built from the template of the chosen composition, describing pose, main element, spatial layers and the safe zone. The skill does not render the image: you pass the prompt to your own image generation model. The skill's description and questions are written in Chinese.
Who it is for. For authors running Xiaohongshu and WeChat accounts who want cover prompts featuring their own face without picking a composition by hand.
Good fit when
- You have a finished article or post and need a cover in one of ten recognisable styles
- You want your own face on the cover by uploading a reference photo as image one
- A product or interface screenshot should go into the cover as the second reference image
- You need a series of covers in different compositions for one image generation model
Not a fit when
- You need a finished picture: the skill returns only prompt text
- You do not read Chinese and do not want Chinese questions and templates
- You want a cover with no people: all ten compositions are built around a person in frame
- For platforms that do not use Chinese headlines you will have to edit the prompts by hand
Example request
Here is my article about a new AI tool. Make a Xiaohongshu cover prompt in the dark gradient styleLimitations
The skill does not create images itself, so a separate generation model is required. Questions, templates and headlines are in Chinese, so the covers suit Chinese-language platforms. The repository has no separate licence file, and the MIT licence is stated in the README. To preserve facial features you need your own photo as the first reference image. All ten compositions are built around a person in frame. The choice of facial expression, background, font and text colour happens in a dialogue, and the agent asks these questions one at a time.
How to disable. Delete the ~/.claude/skills/cover-design-open folder.
FAQ
Which cover compositions does Oh My Cover Design offer?
Ten: dark gradient, flat colour, product hero, comparison cards, minimal whitespace, collage, figure on the side, back view, partial frame and direct eye contact.
How do I install the cover skill in Claude Code?
Clone the repository into ~/.claude/skills/cover-design-open or download only the SKILL.md file into that folder with curl.
How do I keep my face on the cover?
Upload your photo as the first reference image. The prompt asks the model to preserve the facial features from it. Other images are materials such as screenshots, product photos or brand assets.
Does the skill create a finished image?
No. It outputs prompt text that you run in your own image generation model.
Does the skill work in Codex as well as Claude Code?
Yes. The author lists support for Claude Code, Codex and any agent that can load custom skills.
Discussion
No questions or reviews yet.
Related
SkillsA visual-director skill for Xiaohongshu carousels: style choice, page structure, image prompts and the post caption
A WeChat article layout skill: turns Markdown into paste-ready HTML with inlined styles that survives pasting into the official account editor
Agent skill that builds social card carousels and paired covers from an article, screenshots, notes or video in two visual systems
xiaohu-ip-studio: article illustrations with a fixed character
xiaohu-ip-studio
A skill that illustrates Chinese articles: it decides which sections need a picture and draws them with a recurring character
Content skills: article illustrations, covers, infographics, comics, translation, URL to Markdown and social posting
E-commerce Visual Copywriting
ecommerce-visual-copywriting
A skill that turns product data into a plan for main images, the product page and A+ content, with on-image copy, generation prompts and a claims check
Comparison with similar tools
Entries that share the main search query, and how they differ.
| Entry | Official | Repository stars | Main difference |
|---|---|---|---|
| Oh My Cover Design (this entry) | no | 274 | |
| XHS Visual Director | no | 1,447 | drives whole Xiaohongshu cards: style, sheet structure, prompts; here it is only a cover in one of ten compositions |
| Guizang Social Card Skill | no | 7,841 | builds carousels and paired covers as HTML, while this skill outputs prompt text for an image model |
README in short
The README is in Chinese: it states the idea briefly, shows eight example covers, lists ten compositions with their use cases and explains the role of reference images. Two install commands are given.
README badge
Original SKILL.md
--- name: cover-design-open description: "公众号/小红书封面设计。10种构图风格可选，真人出镜，生成可直接用于图片生成模型的提示词。支持多参考图：图1固定为人脸，图2起为UI截图、产品图等素材。关键词：封面设计、提示词生成、公众号封面、小红书封面。" --- # 封面设计 Skill 读取文章内容 → 逐步问用户 → 生成提示词。 **重要：一次只问一个问题，等用户回答后再问下一个。** --- ## 问题流程（逐一问，不要一次全问） **第1问：选风格** 列出以下10种风格，让用户选一种： 1. **深色渐变风** — 人物居中，大字覆盖后方，深色渐变背景，高对比强冲击 2. **纯色扁平风** — 人物抠图感，纯色背景，干净清爽 3. **产品主视觉风** — UI截图/产品占主体，人物做引导手势，适合有素材图 4. **对比卡片风** — 人物手持两张对比卡片近大远小推向镜头，适合前后对比内容 5. **极简留白风** — 大面积留白，文字是主视觉锤，人物轻量出镜 6. **海报拼贴风** — 多张参考图叠加构图，分层有纵深，适合素材丰富的内容 7. **人物侧置留白风** — 人物偏一侧占小比例，大面积留给标题，版面大气 8. **背影构图风** — 人物背对镜头，制造代入感和想象空间 9. **局部出镜风** — 只露手/半脸/侧脸，产品或文字是绝对主角 10. **正面对视风** — 人物直视镜头，眼神接触，文字环绕脸部四周 --- **第2问：图1** 问用户是否有人脸参考图（图1）： - 有 → 让用户上传，提示词里写"参考图1的五官特征" - 没有 → 让用户描述人物（性别、大致外貌），提示词里直接描述 --- **第3问：人物表情** （背影构图风跳过此问） 给用户以下选项： 1. **捂嘴惊讶** — 双手捂嘴，眼睛睁大，震惊感 2. **张嘴震惊** — 嘴巴微张，眼神放大，强烈惊讶 3. **开心大笑** — 嘴角上扬，眼睛弯起，真实喜悦 4. **兴奋雀跃** — 眉飞色舞，身体前倾，藏不住的激动 5. **自信得意** — 嘴角微扬，眼神笃定，"就这么简单"的感觉 6. **托腮思考** — 单手托腮，眼神若有所思 7. **推荐种草感** — 微笑点头，眼神看向镜头，"你一定要试试" 8. **交给模型决定** --- **第4问：额外素材** 问用户有没有图2、图3等额外素材（UI截图、产品图、其他参考图）：