Takes any image and either beautifies it to marketing perfection or shows what it really looks like in daily use, depending on what you feed it. The skill analyzes whether your input is already idealized (renders, ads, promotional shots) or shows real-world wear, then generates the opposite. Works across domains like architecture, food, products, portraits, travel spots. You can also get side-by-side comparisons with that clean split-screen look. It maps five universal dimensions (lighting, materials, color, atmosphere, composition) to domain-specific expressions, so a tired apartment render gets the harsh fluorescent truth while your grainy phone photo gets the developer's promise treatment. Honestly just a systematized way to generate expectation versus reality memes with actual visual intelligence.

skills.sh installs
1.3K
GitHub stars
8
last updated
Jul 14, 2026

Install

$ npx -y skills add lionad-morotar/anti-render-skill --skill anti-render --agent claude-code

Installs into .claude/skills of the current project.

Files

Anti-Render 理想vs现实视觉对比生成器

核心理念

通过并置(juxtaposition)手法,揭示任何领域中"承诺与交付之间巨大落差"的普遍困境。左侧呈现理想化的完美渲染,右侧揭示真实的日常面貌。

执行流程

  1. 接收图像 → 分析内容,识别所属领域为 $domain,计算图片宽高比为 $ratio
  2. 状态判断 → 确定当前状态(破败/普通/理想)
  3. 意图识别 → 根据用户指令确定输出模式
  4. 参数映射 → 将通用五维度映射到领域专属表达
  5. 构建提示词 → 构建基于领域专属表达的提示词
  6. 生成图像 → 生成目标图像

works well with skills: image-to-prompt, prompt-to-image

工作模式

1. 图像状态识别

用户上传图片后,分析其当前状态:

状态特征输出目标
破败质量问题、使用痕迹、维护不良生成理想化渲染图(即 step 2.1)
理想用户上传了营销图片、广告图片等精修后照片生成轻微破败渲染图片(即 step 2.2)
普通/正常无明显破损、日常使用状态生成理想化渲染和轻微破败的对比图(即 step 2.3)

如状态模糊,无法判断意图,主动询问用户期望方向

2. 三种输出模式

2.1 理想化渲染 (Ideal)

  • 目标:输出对应领域的宣传级别完美呈现
  • 特征:高饱和度、完美光影、无瑕疵、精心构图

2.2 真实面貌 (Reality)

  • 目标:输出对应领域日常的真实状态(非破败)
  • 特征:自然光线(“死亡打光”)、真实质感、日常氛围、未经修饰

2.3 对比图 (Comparison)

  • 目标:输出理想化渲染和真实面貌并置的对比图
  • 排列:根据原图宽高比自动选择左右或上下排列

领域识别与适配

领域检测

基于图像内容关键词匹配:

建筑领域: 建筑外观、城市景观、室内空间、建筑效果图、楼盘、住宅、商业空间
人像领域: 人像写真、Cosplay、自拍、证件照、活动拍摄、肖像
产品领域: 电商产品、商品展示、包装设计、电子产品、服饰
食物领域: 美食摄影、菜品展示、烘焙、饮品、餐厅菜单
旅游领域: 风景照、景点打卡、酒店房间、度假胜地
游戏领域: 游戏截图、游戏宣传、UI界面、角色设计
健身领域: 健身照、运动场景、瑜伽、健身房
家居领域: 室内装修、家具展示、样板间、智能家居
科技领域: 产品发布会、概念设计、VR/AR、智能汽车

Sponsors

Details

Repository
github.com/lionad-morotar/anti-render-skill

Recommended

1juliusbrussee avatarcavemanjuliusbrussee/cavemanUltra-compressed communication mode. Cuts output tokens 65% (measured) by speaking like caveman while keeping full technical accuracy. Supports intensity levels618.3Kskills.sh installs
2mattpocock avatargrill-memattpocock/skillsA relentless interview to sharpen a plan or design.1.5Mskills.sh installs
3shadcn avatarimproveshadcn/improveSurvey any codebase as a senior advisor and produce prioritized, self-contained implementation plans for OTHER models/agents to execute. Strictly read-only on s40Kskills.sh installs
4obra avatarsystematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixes326.7Kskills.sh installs
5forrestchang avatarkarpathy-guidelinesforrestchang/andrej-karpathy-skillsBehavioral guidelines to reduce common LLM coding mistakes through explicit assumptions, simplicity, and verifiable success criteria.17.9Kskills.sh installs
6vercel-labs avatarfind-skillsvercel-labs/skillsHelps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express int4.3Mskills.sh installs