Skip to content
Back to search

photo-revival

Turn ordinary photos or everyday snapshots into poetic hand-drawn illustrations with large white-paper negative space, vivid localized color, subtle collage accents, and tiny handwritten captions. Use when a user wants a real photo reimagined as an illustration rather than filtered, beautified, or photorealistically edited.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/dacnay816y62-hub-photo-revival/badge.svg)](/listings/dacnay816y62-hub-photo-revival)

HTML

<a href="/listings/dacnay816y62-hub-photo-revival"><img src="/api/listings/dacnay816y62-hub-photo-revival/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

SKILL.md · Manifest · 59 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. ---
  2. name: photo-revival
  3. description: Turn ordinary photos or everyday snapshots into poetic hand-drawn illustrations with large white-paper negative space, vivid localized color, subtle collage accents, and tiny handwritten captions. Use when a user wants a real photo reimagined as an illustration rather than filtered, beautified, or photorealistically edited.
  4. ---
  5. # Photo Revival
  6. ## Core idea
  7. Keep the subject and scene recognizable, but redraw the image as a fresh illustration on white paper.
  8. Default to:
  9. - 3:4 vertical composition
  10. - large white-paper negative space
  11. - a tiny subject or focal cluster occupying about 10-16% of the page
  12. - delicate hand-drawn texture
  13. - vivid local color only where it matters
  14. - tiny handwritten text, dates, or short captions
  15. ## Workflow
  16. 1. Identify the photo's main subject, setting, and one or two memorable details.
  17. 2. Preserve those details in simplified form; do not copy the photo literally.
  18. 3. Rebuild the scene as an illustration with paper texture, brush marks, pencil edges, or watercolor softness.
  19. 4. Leave 80-88% of the page as blank white paper.
  20. 5. Add text only as a small note, label, date, or poetic fragment.
  21. ## Style rules
  22. - Use white or very light paper as the base.
  23. - Keep the illustration area small and concentrated: default 10-16% of the page, absolute maximum 18%.
  24. - Keep the strongest color inside that small illustrated area; do not let color spread into the blank field.
  25. - Favor vivid but controlled color in a limited area.
  26. - Let the drawing feel handmade, not filtered.
  27. - Allow subtle collage, stamp, or clipping hints only when they stay tiny and integrated.
  28. ## Avoid
  29. - Full-bleed image fills
  30. - Photo-filter looks
  31. - Heavy poster typography
  32. - Dense layouts
  33. - Subject areas larger than 18% of the page
  34. - Flat vector treatment
  35. - Large collage blocks
  36. - Overly realistic retouching
  37. ## Output prompt shape
  38. When directing image generation, describe:
  39. - what must be preserved from the photo
  40. - what should become illustrated
  41. - the paper base and empty space
  42. - the tiny text treatment
  43. - the desired color accents
  44. Keep the instruction short and focused.
README.md · README · 221 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # 废片焕新 · Photo Revival
  2. **FANTASY / 梵想美学 · 摄影与照片转译**
  3. 把生活照片转译成白纸上的小幅手绘,保留记忆点,以大量留白与少量诗性文字重新组织画面。
  4. **[快速开始](#start)** · **[下载与安装](#install)** · **[完整规则](SKILL.md)** · **[全部视觉 Skills](https://github.com/dacnay816y62-hub?tab=repositories)**
  5. | 视觉示例 01 | 视觉示例 02 |
  6. | :---: | :---: |
  7. | ![废片焕新 · Photo Revival · 示例 1](examples/01_moon_gate.png) | ![废片焕新 · Photo Revival · 示例 2](examples/06_cat_bed_paper.png) |
  8. <a id="start"></a>
  9. ## 一分钟开始
  10. | 你提供 | 这套 Skill 组织的交付 |
  11. | --- | --- |
  12. | 生活随拍、物件、街角或旅行照片 | 3:4 白纸手绘插画页,主体小、留白多 |
  13. ```text
  14. 用 $photo-revival 把这张照片重新画成白纸上的小幅水彩手绘。保留主体与空间关系,主体约占整页 10–16%,文字只留一句很小的中文批注。
  15. ```
  16. **生成说明:** Skill 组织设计判断、提示词与执行流程;图片由当前环境中可用的图像工具生成或编辑。示例用于理解视觉方向,具体来源以本仓库记录为准,不能据此保证每次得到相同效果。
  17. <a id="install"></a>
  18. ## 下载与安装
  19. **[下载当前分支 ZIP](https://github.com/dacnay816y62-hub/photo-revival/archive/refs/heads/main.zip)** · **[阅读 Skill 规则](SKILL.md)**
  20. 1. 下载并解压仓库。
  21. 2. 将仓库根目录(包含 `SKILL.md`)放入当前助手支持的技能目录。
  22. 3. 安装文件夹命名为 **`photo-revival`**,确保入口是 `photo-revival/SKILL.md`。
  23. 4. 在支持技能调用的会话中使用 **`$photo-revival`**。如果列表未刷新,新开一个任务。
  24. Codex CLI / IDE 的用户级目录是 `~/.agents/skills/`,项目级目录是 `.agents/skills/`;Windows 用户目录可写为 `%USERPROFILE%\.agents\skills\`。以 [OpenAI 官方安装说明](https://learn.chatgpt.com/docs/build-skills#where-codex-loads-local-skills) 为准。ChatGPT 与其他宿主请按各自的技能加载方式使用。
  25. 仓库名与调用名可能不同,以上以 `SKILL.md` 中的名称为准。安装不包含图像服务、账户或生成额度;实际出图取决于你使用的环境。
  26. ---
  27. ## 核心效果
  28. - 3:4 竖构图
  29. - 白色或近白纸张底图
  30. - 80-88% 大面积留白
  31. - 主体插画只占画面 10-16%
  32. - 主体最大不超过整页 18%
  33. - 色彩只集中在这一小块插画区域
  34. - 手绘笔触:铅笔、水彩、干刷、蜡笔、轻微 risograph 颗粒
  35. - 少量中英文手写小字、日期、field note
  36. 一句话:**把照片重新画成一页诗,而不是把照片修成插画。**
  37. ## 适合做什么
  38. - 随手拍、废片、日常照片
  39. - 小猫、食物、杯子、窗户、楼道、旧店、街角
  40. - 旅行碎片、建筑、树、桥、庭院、器物
  41. - 传统文化、非遗、民俗、手作、地方记忆
  42. - 想保留真实主体,但希望画面变得更文艺、更轻、更有纸感
  43. ## 不适合做什么
  44. - 写真人像精修
  45. - 商业产品大海报
  46. - 大标题视觉海报
  47. - 满版拼贴
  48. - 密集信息图
  49. - 需要严格还原原照片像素的修复任务
  50. ## 中文操作指南
  51. 下面是从安装到出图的一套完整中文流程。
  52. ## 基础用法
  53. 在 Codex 里上传一张照片,然后像平常聊天一样说就行:
  54. ```text
  55. 用 $photo-revival,把这张照片画成那种白纸留白的小手绘插画。
  56. ```
  57. 稍微具体一点可以这样说:
  58. ```text
  59. 用 $photo-revival 处理这张照片,保留主体和氛围,但重新画成大留白、白纸质感、带一点诗意小字的手绘插画。
  60. ```
  61. 如果要批量测试,可以说:
  62. ```text
  63. 用 $photo-revival 随机测试这几张照片,每张都保持白纸大留白,小小一块手绘插画,文字少一点、有诗气一点。
  64. ```
  65. 默认情况下,这个 skill 会自己控制比例:主体很小、留白很多、颜色集中在局部。只有结果跑偏时,才需要使用下面的更严格提示词。
  66. ## 推荐工作流
  67. 1. 上传照片。
  68. 2. 先识别照片里最重要的 1-3 个记忆点。
  69. 3. 保留这些记忆点:主体、姿态、空间关系、关键物件、情绪。
  70. 4. 用图像生成重新绘制,不要做本地滤镜或简单风格迁移。
  71. 5. 输出 3:4 竖图。
  72. 6. 检查主体是否过大。如果超过 18%,重新生成。
  73. 7. 检查文字是否过大、重复、乱码。如果有问题,重新生成或减少文字。
  74. ## 通用提示词模板
  75. ```text
  76. Transform the reference photo into a 3:4 vertical poetic hand-drawn illustration page on clean white textured paper.
  77. Preserve:
  78. <照片中必须保留的主体、姿态、空间、物件、氛围>
  79. Redraw:
  80. Turn the photo into a fresh hand-drawn illustration, not a photorealistic edit and not a filter.
  81. Composition:
  82. The illustrated subject occupies only 10-16% of the full page, absolute maximum 18%.
  83. Keep 80-88% of the page as untouched blank white paper.
  84. Place the small illustration slightly below center or gently off-center.
  85. Texture:
  86. Use pencil line, watercolor wash, dry brush, wax pastel edges, light print grain, and subtle paper texture.
  87. Color:
  88. Use vivid color only inside the small illustrated area.
  89. Do not spread color into the blank white-paper field.
  90. Text:
  91. Add tiny handwritten caption: "<一句短中文诗句>"
  92. Add tiny English note: "<FIELD NOTE / DATE>"
  93. The text must be very small, handwritten, imperfect, and part of the image.
  94. Avoid:
  95. full-bleed photo, photo-filter look, large portrait crop, subject larger than 18%, dense collage, old yellow paper, big typography, duplicated text.
  96. ```
  97. ## 示例文案
  98. 可以把中文小字写得短一点,像页边批注:
  99. - 圆镜里藏着今天
  100. - 猫把夜晚抱住
  101. - 风从旧店门口经过
  102. - 蓝里藏着风
  103. - 一剪春天醒了
  104. - 竹声很轻
  105. - 影子也会唱戏
  106. 英文小字适合做成档案感:
  107. - `FIELD NOTE / 08.02`
  108. - `SOFT MEMORY`
  109. - `PAPER STUDY`
  110. - `QUIET ROOM`
  111. - `HAND MEMORY`
  112. - `DAILY ARCHIVE`
  113. ## 样张画廊
  114. | | | |
  115. |---|---|---|
  116. | ![](examples/01_car_page.png)<br/>汽车页 | ![](examples/01_flower_tree_paper.png)<br/>花树 | ![](examples/01_moon_gate.png)<br/>月洞门 |
  117. | ![](examples/02_camera_page.png)<br/>相机笔记 | ![](examples/03_bubbles_page.png)<br/>泡泡 | ![](examples/03_pagoda.png)<br/>塔 |
  118. | ![](examples/03_red_bike_dot.png)<br/>红色自行车 | ![](examples/03_runner_paper.png)<br/>跑步者 | ![](examples/04_skyline_page.png)<br/>城市天际线 |
  119. | ![](examples/05_dumpling_paper.png)<br/>一碗食物 | ![](examples/05_japan_road_page.png)<br/>道路记忆 | ![](examples/05_old_shop_torn_label.png)<br/>旧店 |
  120. | ![](examples/06_cat_bed_paper.png)<br/>床上的猫 | ![](examples/06_mask_dance.png)<br/>面具舞 | ![](examples/06_train_window_page.png)<br/>车窗 |
  121. | ![](examples/07_mountain_yaks.png)<br/>山与牦牛 | ![](examples/10_turkeys.png)<br/>田野笔记 | |
  122. ## 常见问题
  123. ### 主体太大怎么办?
  124. 重新生成,并把约束写得更硬:
  125. ```text
  126. The illustrated subject must occupy only 10-16% of the full page.
  127. Absolute maximum subject area is 18%.
  128. Keep 80-88% of the page blank white paper.
  129. ```
  130. ### 文字重复或乱码怎么办?
  131. 减少文字,只保留一句中文和一个英文日期。文字必须“小、少、像手写批注”,不要当成海报标题。
  132. ### 生成结果像照片滤镜怎么办?
  133. 强调:
  134. ```text
  135. This must be newly redrawn as an illustration, not a filtered version of the photo.
  136. ```
  137. ### 颜色铺太满怎么办?
  138. 强调:
  139. ```text
  140. Color must stay localized inside the tiny illustrated area.
  141. The blank paper field remains white.
  142. ```
  143. ## 文件说明
  144. - `SKILL.md`:Codex skill 的核心指令。
  145. - `agents/openai.yaml`:Codex 展示用元数据。
  146. - `examples/`:公开样张。
  147. ## License
  148. MIT
  149. ## FANTASY / 梵想美学
  150. **让想象先被看见。** 将视觉判断与创作流程整理成可以继续使用的方法。
  151. **[浏览全部视觉 Skills](https://github.com/dacnay816y62-hub?tab=repositories)** · [Street Photo Illustration](https://github.com/dacnay816y62-hub/street-photo-illustration-skill) · [FANTASY Minimal Magazine](https://github.com/dacnay816y62-hub/FANTASY-Minimal-Magazine)

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.