Anti Render
Turn uploaded photos into ideal-vs-reality juxtaposition visuals for memes, honest marketing, or before/after storytelling without hand-writing image prompts.
Overview
Anti-render is an agent skill most often used in Launch (also Validate landing tests and Grow content) that analyzes an uploaded image and generates ideal-vs-reality visual comparisons across many domains.
Install
npx skills add https://github.com/lionad-morotar/anti-render-skill --skill anti-renderWhat is this skill?
- Detects image domain (architecture, portrait, product, food, travel, and nine other domains) and maps five universal con
- Three output modes: ideal render only, everyday reality only, or full comparison with auto left-right or top-bottom layo
- Classifies upload state as dilapidated, ideal/marketing, or normal to route ideal upgrade, slight wear-down, or dual-pan
- Six-step pipeline: receive image, state judgment, intent recognition, parameter mapping, prompt build, then generate
- Pairs with image-to-prompt and prompt-to-image skills for end-to-end prompt and render workflow
- Five core contrast dimensions mapped per domain
- Three output modes: ideal, reality, and comparison
- Ten-plus domain detection buckets including architecture, product, food, and travel
Adoption & trust: 1.2k installs on skills.sh; 7 GitHub stars; 3/3 security scanners passed (skills.sh audits).
What problem does it solve?
You have a real or marketing photo but need a sharp visual joke or honest contrast that shows the gap between sold perfection and everyday reality.
Who is it for?
Indie founders, creators, and marketers making juxtaposition creatives for social, ads, or landing pages when the concept is promise versus delivery.
Skip if: Builders who need factual photo correction, legal-grade product representation, or workflows where model auto-invocation is required—this skill sets disable-model-invocation and expects explicit orchestration.
When should I use this skill?
User says anti-render, 理想vs现实, 对比图, 渲染vs真实, or asks for ideal-vs-reality visual comparison across any domain.
What do I get? / Deliverables
You get domain-tuned prompts and generated ideal, reality, or split comparison images ready to pair with image-to-prompt and prompt-to-image skills.
- Domain-mapped image prompts
- Ideal-only, reality-only, or comparison generated image
- Layout choice matched to source aspect ratio
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Visual contrast content ships when you are preparing launch and growth assets that need scroll-stopping, shareable imagery. Distribution is where polished render fantasies meet real-user expectations—this skill automates that narrative in one image.
Where it fits
A/B a landing hero that shows condo render fantasy beside a realistic unit photo before you commit to ad spend.
Ship a launch-thread meme contrasting product ad lighting with everyday phone-photo lighting.
Publish recurring expectation-vs-reality posts for food, travel, or fitness niches to drive shares.
Illustrate README or pitch decks with honest render-vs-reality panels for physical or UI products.
How it compares
Use instead of generic “make two versions” chat prompts when you want structured domain detection, state routing, and comparison layout rules.
Common Questions / FAQ
Who is anti-render for?
Solo builders, marketers, and creators who want scroll-stopping ideal-vs-reality images for products, places, food, fitness, and other domains without writing comparison prompts from scratch.
When should I use anti-render?
Use it during Validate when testing landing hero concepts, at Launch when shipping campaign or meme creatives, and in Grow when publishing honest-before-after or expectation-gap content—trigger with anti-render, ideal vs reality, or comparison image phrases.
Is anti-render safe to install?
Review the Security Audits panel on this Prism page for install risk and file hash; image-generation skills typically need network or API access, so confirm permissions match your agent policy before enabling.
Workflow Chain
Then invoke: image to prompt, prompt to image
SKILL.md
READMESKILL.md - Anti Render
# 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 ``` ### 核心对比维度（通用框架） 所有领域共享以下五个核心对比维度： #### 1. 光影 (Lighting) | 理想侧 | 现实侧 | |--------|--------| | 精心计算的完美光照 | 自然/现场实际光线 | | 黄金时刻或柔和补光 | 硬光、顶光或平淡漫射光 | | 明暗层次丰富、无死黑/过曝 | 曝光妥协、阴影浓重 | | 方向性明确、立体感强 | 低对比度、缺乏层次 | #### 2. 材质 (Material) | 理想侧 | 现实侧 | |--------|--------| | 完美无瑕的表面 | 真实使用痕迹 | | 色彩饱和、质感强化 | 褪色、污渍、磨损 | | 无灰尘、无水痕、无瑕疵 | 自然老化、环境痕迹 | | CG般的精确反射/折射 | 混浊、不完美的反射 | #### 3. 色彩 (Color) | 理想侧 | 现实侧 | |--------|--------| | 高饱和度、鲜艳夺目 | 低饱和度、略显平淡 | | 色温精准、统一协调 | 色温偏移、白平衡未校正 | | 后期精修的色彩增强 | 相机原生色彩还原 | | 广告级别的视觉吸引力 | 日常感、朴素感 | #### 4. 氛围 (Atmosphere) | 理想侧 | 现实侧 | |--------|--------| | 充满活力、生机勃勃 | 冷清、平凡或略显尴尬 | | 精心布置的场景元素 | 杂乱的现场环境 | | 梦幻、理想化的背景 | 真实、暴露现场的环境 | | 情绪饱满、引人入胜 | 纪实感、冷峻客观 | #### 5. 构图/细节 (Composition) | 理想侧 | 现实侧 | |--------|--------| | 完美的透视与比例 | 自然的镜头畸变 | | 瑕疵移除、穿帮修复 | 保留所有现场细节 | | 精心安排的元素布局 | 随机、不规则的真实分布 | | 后期添加的特效/光效 | 无后期加持的原始状态 | ## 对比图技术规范 ### 排列规则 如生成对比图，需根据原图宽高比 `$ratio` 判断新的排列规则： - **原图横向**（宽 > 高）：对比图上下排列，水平分割线 - **原图纵向**（高 > 宽）：对比图左右排列，垂直分割线 ### 分割线规范 - 位置：画面正中 - 宽度：2-5像素 - 颜色：纯白或极浅灰 - 边缘：锐利清晰，无羽化 ### 内容规则 - 对比图左或上（根据排列规则）：理想化渲染 - 对比图右或下（根据排列规则）：普通现实主义（默认）或破败状态（用户明确要求强烈对比效果） - **默认不要添加标题文字**，保持画面纯净，让视觉对比本身说话