O conteúdo está em inglês; a navegação e os filtros mantêm o idioma selecionado.

Curated List of Six Aesthetic AI Photo Skills Released

A Chinese tech blogger announced a new series curating useful AI skills, starting with six photo-processing skills covering abstract editorial edits, zine posters, Heytea-style posters, personal-IP illustrations, poetic hand-drawn effects, and CRT retro interfaces.

Publicado em 13/08/2026, 08:40:00Última atualização 13/08/2026, 08:40:00Publicação após revisão automática

A Chinese tech blogger, oran_ge, shared the first batch of a planned series that curates high-quality AI skills. This initial collection focuses on photo-processing skills with a strong aesthetic sense, each accompanied by a direct link.

Among the six skills, photo-abstract-editorial keeps the original photo intact and only extracts spatial relationships, compositional rhythm, and color relations to create an abstract editing panel. gc-minimal-zine-poster turns a theme, sentence, emotion, or photo into a minimal zine-style poster with paper texture. The list also includes Heytea-style poster generation, cute personal-IP illustrations, poetic hand-drawn illustration for ordinary photos, and retro CRT-computer interface illustrations.

The photo-abstract-editorial project is open source on GitHub with around 2.3k stars, matching the share's description. This trend reflects the growing ecosystem of reusable AI skills for platforms such as Claude Skills and Codex Skills, helping users streamline creative workflows.

The blogger said the series will be updated periodically, and interested users can try the skills via the links in the original post. The curation highlights a growing community focus on combining aesthetics with practical AI tooling.

Evidências das fontes 6

A verificação de fontes não garante exatidão permanente. Considere a fonte original, a data e as correções posteriores.

随手拍也能变成杂志大片：最近值得玩的 9 个照片处理 Skill | Claude APIclaudeapi.com · Complementar

## 01 GC Minimal Zine Poster：把普通照片变成安静的独立杂志封面项目地址：LiamGvchi/gc-minimal-zine-poster 这是最近最容易让人一眼心动的图片 Skill 之一。截至 2026 年 8 月 11 日，GitHub 页面显示它已有约 1.3k Star。它的思路很克制：不追求把画面塞满，而是把一张照片、一个物件、一句话或一种情绪，压缩成一张 3:5 的竖版 zine 海报。它的视觉规则很明确：这些规则看起来简单，实际很重要。很多 AI 海报失败，不是模型不会画，而是画面里什么都有：大标题、强光、3D 字、霓虹背景、各种装饰元素一起抢注意力。这个 Skill 反过来做减法，所以成片更像独立出版物，而不是促销模板。它适合：调用时，不必只说“帮我变高级”。可以把希望保留的元素讲清楚：用 GC Minimal Zine Poster 处理这张海边照片。保留远处的小船和人物之间的距离感，整体安静，不要商业广告感。主色使用旧纸白和深蓝，只保留一句很短的中文标题。 Skill 会返回生成图、最终提示词，以及本次选择的视觉方案。也就是说，第一次结果不满意时，你还能继续沿着同一个方向微调，而不是从头抽卡。 ## 02 Photo Abstract Editorial：照片不重画，只在旁边长出一块“抽象记忆” 项目地址：ZzzLc0405/photo-abstract-editorial 这个项目更适合不想让 AI 改动原照片的人。它的成品通常由三个部分组成：原始摄影区域、从原图提炼出的抽象面板，以及一个简短的英文标题。关键点是：原图仍然是原图。 [...] 关键点是：原图仍然是原图。这个项目的 Skill 名称是 `photo-abstract-editorial`。它会把照片中最有辨识度的视觉事实映射成极简抽象符号，再把原始摄影与抽象图形组合到同一张版面中。 `photo-abstract-editorial` Skill 不会把建筑重画一遍，也不会擅自给人物换脸。它只从照片里读取空间关系、构图节奏和色彩，然后在照片下方或旁边生成一块象牙白底的抽象区域。例如：截至 2026 年 8 月 11 日，GitHub 页面显示该项目约有 2.3k Star。它受欢迎的原因并不