⏳ This skill is pending AI review.
Scores will appear once the review pipeline completes.
surreal-pop-collage
把照片变成超现实波普拼贴画(surreal pop collage)——照片去色保留为"现实锚",背景替换为巨大平涂色形,全图只有一个"不可能的巨物"。当用户想要超现实拼贴、现实超现实化、小红书/抖音超现实海报、黑白照片+平涂色块的梦境感图片,或要把风景/人物/城市照做成超现实艺术时使用。
Choose how to use this skill
You do not need every option. Choose the path your AI client supports. The stable page stays the same; versioned files are immutable.
1. Native installer
This listing has no registered native installer command. Use the complete package or source fallback below, depending on what your client supports.
Do not guess an installer command or replace an existing version without reviewing the diff.
2. Complete package recommended
Download the ZIP when available. It includes SKILL.md plus the references, security notes and version metadata.
No complete ProSkills package is published for this listing yet.3. Prompt-only
Copy the prompt above when the agent can read the stable page or when you want to adopt the workflow without installing a skill.
Need only the instruction file?
Download SKILL.md only if your client requires a single file. The complete ZIP is safer for a full installation because it preserves the references and release context.
No path installs or executes anything by itself. Your agent still needs access to the project files. Before updating, compare the installed version and review the diff.
// RATINGS
Not yet listed on ClawHub or SkillsMP
// README
超现实波普拼贴 · Surreal Pop Collage
一个给 AI 编程/agent 工具用的图像生成 skill:把任意照片变成超现实波普拼贴画——照片去色保留为"现实锚",背景换成巨大平涂色形,全图只有一个"不可能的巨物"。
黑白现实、平涂梦境、一物不对劲、元素源于图、色从图中来。
效果
(示例图见 examples/——风景、城市、人物三类示例。)
安装
Kimi Code / Claude Code / Codex 等 agent 工具:把本仓库的 SKILL.md 与 agents/ 复制到你的 skills 目录(如 ~/.kimi-code/skills/surreal-pop-collage/),agent 会在你说"把这张照片做成超现实拼贴"时自动调用。
任何生图工具(ChatGPT / Midjourney / 即梦等):不用安装,直接按下面的配方手写 prompt。
用法(30 秒版)
- 选一张照片,写下三行:主体是什么 / 主色是哪两个 / 它在哪
- 定平涂色形:从主色出发——提纯、互补、或情绪反转,三选一,2–3 个大色形替换背景
- 定巨物:只能一个,从场景里长出来——图里的小东西放大 / 语义最远的东西 / 尺度颠倒;文化地标就用它的典故
- 按四段式拼 prompt:
surreal pop collage, vertical 3:4,
keep the [主体] clearly recognizable but desaturated to black and white,
the background replaced by huge flat matte color shapes: [色形与颜色],
one impossible giant element: [巨物],
[小元素群] in graduated sizes following an arc,
a few white hand-drawn graffiti strokes,
flat matte colors, no gradients, no text, no watermark
完整规则(决策优先级、色形推导、巨物选择、纠偏表、质量门)见 SKILL.md。
核心规则速览
- 一物不对劲:巨物唯一。第二个不可能元素出现就删——超现实的力量全在"只有一处不对劲"
- 元素源于图:巨物和色形都必须能从原图指出出处;禁止默认的红日/蓝天/海豚/鲸鱼
- 平涂无渐变:色形不许有渐变和立体阴影,这是和照片真实感对撞的关键
- 典故巨物:文化地标照片,巨物从典故里选,但先翻译成可见的物件或动作(例:白居易开凿山塘河→灯笼升空成月)
- 中文后期排:生图模型写中文必乱码,prompt 里
no text,标题后期用代码排
示例图
见 examples/:
- 城市街景:黄色天空来自出租车的黄,巨物是吊在楼谷间的红绿灯
- 更多示例陆续补充(风景/人像/文化地标系列)
License
MIT
// HOW IT'S BUILT
KEY FILES