CommunityCoding & Developmentgithub.com

Chengyunlai/dry-media-silhouette

A Chinese Codex Skill that transforms photos into 3:4 minimalist, bold black silhouette illustrations influenced by dry media texture.

What is dry-media-silhouette?

This CDI (Chengyunlai/dry-media-silhouette) Skill is a specialized tool for Codex users (compatible with platform-agnostic AI agents) that converts ordinary photographs into stylized 3:4 aspect ratio silhouette illustrations with a distinctive dry media aesthetic. It applies bold black outlines, minimalist shaping, and a rough, textured 'dry' finish reminiscent of charcoal or chalk on paper. The skill processes user-uploaded images, preserving key subjects while abstracting backgrounds and details into stark, high-contrast silhouettes. It is designed primarily for Chinese-speaking users, with all prompts and outputs localized in Chinese, but the visual results are universally understandable. The skill's workflow typically involves image input, AI analysis, and regeneration following strict 3:4 proportions and a predefined visual style. It is useful for artists, designers, content creators, and anyone seeking to quickly produce consistent, artistic silhouette art from personal or stock photos without manual editing.

Works with~Claude Code✓Codex CLI~Cursor
npx skills add Chengyunlai/dry-media-silhouette

Installed? Explore more Coding & Development skills: steipete/bluebubbles, steipete/eightctl, steipete/blucli · View all 6 →

Ask in your favorite AI

Open a new chat with this agent skill pre-loaded.

Documentation

What does dry-media-silhouette do?

将照片重构为 3:4 极简粗黑干媒介剪影插画的中文 Codex Skill

概述

本技能用于将用户提供的照片转换为 3:4 比例的极简粗黑干媒介剪影插画。它利用 AI 模型理解图像内容,提取主体轮廓,并通过粗黑的线条和干媒介纹理渲染出风格一致的剪影作品。

用法

用户只需上传一张照片,技能将自动处理并返回一张符合 3:4 比例的粗黑干媒介风格剪影插画。

输入要求

  • 图片格式:JPEG, PNG
  • 推荐分辨率:至少 800x600 像素
  • 图片内容:人物、动物或物体均可,但主体应清晰

输出

  • 图片格式:PNG (透明背景可选)
  • 比例:3:4(宽:高)
  • 风格:极简粗黑干媒介剪影

风格指南

  • 颜色:仅使用纯黑 (#000000) 和纯白 (#FFFFFF)
  • 线条:粗犷、不连续,模仿干画笔或炭笔效果
  • 细节:保留主体主要特征,省略次要细节
  • 背景:纯白或透明

兼容性

本技能为 Codex 设计,但理论上可适配支持图像输入和输出的任何 AI 代理平台。

示例

(此处可添加示例图片)

  • 输入:一张人像照片
  • 输出:人像的 3:4 比例粗黑剪影插画

注意事项

  • 复杂场景可能无法完美提取单一主体
  • 建议用于单一主体照片以获得最佳效果
  • 技能提示词为中文,但图片理解不依赖语言

Related Skills