Back to search
photo-to-zine-postcard
SkillGitHub
把你拍摄的照片转换成一套极简、留白充足、带手绘二创元素的 Zine 风格明信片。
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
Markdown
[](/listings/whiplashzeb--photo-to-zine-postcard)HTML
<a href="/listings/whiplashzeb--photo-to-zine-postcard"><img src="/api/listings/whiplashzeb--photo-to-zine-postcard/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 · 466 lines
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
- ---
- name: photo-to-zine-postcard
- description: Turn a user-provided photograph into a minimal two-sided 2:3 zine-style postcard. Preserve the original photo above, then create a sparse hand-drawn source-specific motif, minimal metadata, three sampled color swatches, and a matching functional postcard back.
- ---
- # SKILL.md — photo-zine-postcard-v3.3
- ## 1. Name
- **photo-zine-postcard-v3.3**
- A compact, self-contained skill for generating a minimal postcard front and back from one source photo.
- Use only this document.
- Do not retrieve or reference external skills.
- ---
- ## 2. Goal
- Generate two coordinated images:
- 1. **Front** — original photo embedded above, minimal editorial composition below
- 2. **Back** — unified postcard back
- The design must remain:
- - minimal
- - airy
- - source-specific
- - visually attractive
- - lightly hand-crafted
- - print-friendly
- The source photo remains primary.
- The lower area contains one strong visual motif, not a collection of samples.
- ---
- ## 3. Default Format
- - card ratio: portrait `2:3`
- - reference print size: `100 × 150 mm`
- - background: warm off-white / ivory paper
- - texture: very subtle paper grain
- - original photo: embedded directly
- - lower fragment style: `hand_drawn_first`
- - palette: enabled by default
- ---
- ## 4. Fixed Front Structure
- Do not redesign the layout.
- Use this structure:
- 1. original photo in the upper area
- 2. generous blank transition space
- 3. metadata block on the lower left
- 4. one large main visual motif on the lower right
- 5. optional one much smaller supporting motif
- 6. exactly three small color swatches near the lower area
- Do not add any other visual modules.
- ---
- ## 5. Main Photo
- - Use the actual source photo.
- - Preserve its exact original aspect ratio.
- - Do not stretch, repaint, replace, or crop it by default.
- - Place it in the upper half, centered horizontally.
- - Add a very thin frame.
- - Leave a small even paper gap between photo and frame.
- - At most one tiny archival tape detail is allowed.
- ---
- ## 6. Main Motif Selection
- The model must first select the **most visually attractive and source-defining motif**, not merely the easiest object to isolate.
- ### Selection priority
- Choose the motif with the strongest combination of:
- 1. distinctive source identity
- 2. visually appealing color
- 3. clear silhouette
- 4. strong contrast
- 5. elegant shape
- 6. relevance to the overall photograph
- ### Prefer
- - vivid turquoise or blue water forms
- - distinctive shoreline contours
- - strong mountain silhouettes
- - expressive plant clusters
- - recognizable window-and-vine structures
- - elegant architectural fragments
- - visually dominant clouds or shadows
- ### Avoid as main motif
- - dull beige fragments when a stronger colored subject exists
- - visually insignificant debris
- - the easiest isolated object if it is not the most attractive
- - low-contrast secondary details
- - fragments that do not represent the photograph’s main visual character
- ### Mandatory rule
- When the source contains a clearly dominant color feature, the main motif should normally come from that feature.
- For example:
- - turquoise lake photo → select turquoise water / shoreline motif
- - mountain sunset → select illuminated ridge or sky-lit peak
- - ivy window → select window + vine cluster
- - flower close-up → select the flower, not the pot or background
- ---
- ## 7. Hand-Drawn First Rule
- The main motif should be **hand-drawn by default** when the subject is suitable.
- This is not optional styling language.
- It is the preferred rendering mode.
- ### Required hand-drawn treatment
- Render the main motif as a restrained editorial illustration using one or a combination of:
- - watercolor
- - gouache
- - ink
- - pencil
- - cut-paper illustration
- - painted contour
- - soft hand-rendered texture
- The result must:
- - preserve the original silhouette
- - preserve the major internal structure
- - preserve the recognizable identity
- - keep the source color character
- - feel more lively than a raw crop
- - remain clean and controlled
- ### Do not use a raw crop as the main motif when the motif is suitable for drawing.
- Good hand-drawn candidates:
- - water shapes
- - shorelines
- - islands
- - mountains
- - plants
- - branches
- - windows
- - clouds
- - rocks
- - simple buildings
- - facades
- - natural contours
- ### Crop fallback
- Use a source crop only when exact fidelity is essential, such as:
- - faces
- - hands
- - text or signage
- - precise machinery
- - complex perspective
- - dense repeated detail
- - objects that lose identity when simplified
- When using crop fallback, keep it lightly softened and integrated with the paper.
- ---
- ## 8. Main Motif Size and Placement
- The main motif should be clearly visible and larger than in V3 Lite.
- Recommended size:
- - width: `28%–38%` of card width
- - height: `14%–22%` of card height
- Placement:
- - lower right
- - lower center-right
- The motif must remain subordinate to the main photo, but should feel intentional and substantial.
- ---
- ## 9. Supporting Motif
- Optional only.
- Rules:
- - one supporting motif maximum
- - size: `20%–35%` of the main motif
- - placed close to the main motif
- - must support the same visual story
- - may be hand-drawn or lightly cropped
- - never equal in size to the main motif
- ---
- ## 10. Color Swatches
- Color swatches are enabled by default.
- Use exactly **three small swatches** sampled from the source photo:
- 1. dominant color
- 2. dark structural color
- 3. pale neutral or accent color
- Rules:
- - small and secondary
- - simple circles or rectangles
- - no texture sample cards
- - no large palette strip
- - no more than three
- - placed near the lower edge or near the main motif
- - never replace the main motif
- For the turquoise lake example, the swatches should normally include:
- - bright turquoise
- - deep teal
- - pale mineral sand
- ---
- ## 11. Metadata
- Allowed text only:
- - title
- - optional short subtitle
- - `LOCATION`
- - `DATE`
- - small index number
- Preferred placement:
- - lower left
- - restrained and compact
- Do not add:
- - keyword lists
- - descriptive paragraphs
- - multiple labels
- - phrases such as `SHORE / ISLAND / MINERAL / LAGOON / SALT`
- - invented metadata such as `Unknown` or `Undated` unless explicitly requested
- Leave metadata values blank when not provided.
- ---
- ## 12. Forbidden Front Elements
- Do not add:
- - rows of multiple cutouts
- - multiple main motifs
- - texture sample boxes
- - material sample cards
- - image sample grids
- - large color blocks
- - generic circles
- - decorative dots
- - wave doodles
- - badges
- - seals
- - logos
- - keyword lists
- - long captions
- - full-width lower compositions
- - rounded card corners unless explicitly requested
- If uncertain, simplify.
- ---
- ## 13. Back
- Generate a unified postcard back.
- Include:
- - thin outer border
- - one vertical divider slightly right of center
- - stamp box in the upper-right
- - 3 or 4 address lines on the right
- - large blank message area on the left
- Optional:
- - small `POST CARD` text near upper-left
- - tiny metadata or index near lower-left
- - one extremely faint source-derived watermark
- Do not add:
- - large collage
- - large palette
- - decorative sample blocks
- - anything that reduces writing space
- ---
- ## 14. Quality Requirements
- Target:
- - highest detail quality
- - sharp and clear rendering
- - crisp edges
- - refined hand-drawn texture
- - clean paper texture
- - low noise
- - no blur
- - no muddy watercolor
- - no smudged edges
- - suitable for later 4× super-resolution upscaling
- Quality requirements must improve rendering only.
- They must not increase the number of elements.
- ---
- ## 15. Priority Order
- Always follow this order:
- 1. original photo embedded correctly
- 2. V3 Lite layout preserved
- 3. large white space preserved
- 4. choose the most visually attractive source-defining motif
- 5. use hand-drawn rendering when suitable
- 6. make the main motif large enough to read clearly
- 7. optional one small supporting motif
- 8. include exactly three small source-derived color swatches
- 9. minimal metadata
- 10. omit everything else
- If any later rule conflicts with an earlier rule, keep the earlier rule.
- ---
- ## 16. Front Prompt Template
- ```text
- Create a minimal portrait postcard front using the fixed V3 Lite structure.
- Use the actual source photo in the upper area. Preserve its exact original aspect ratio. Do not repaint, replace, stretch, or crop it unless cropping was explicitly requested. Center it horizontally, keep generous margins, add a very thin frame, and leave a small even paper gap between the photo and frame.
- Keep a large blank transition area below the photo.
- In the lower area, use:
- - one compact metadata block on the lower left
- - one large source-specific main motif on the lower right
- - optionally one much smaller supporting motif near it
- - exactly three small color swatches sampled from the source photo
- Select the main motif based on visual attractiveness and source identity, not ease of extraction. Prefer the most distinctive color-rich and elegant feature in the image. When the source contains a dominant color feature, choose that feature as the main motif. Do not choose a dull neutral fragment when a stronger colored motif exists.
- Render the main motif as a restrained hand-drawn editorial illustration whenever suitable. Preserve its original silhouette, major internal structure, identity, and source color character. Use controlled watercolor, gouache, ink, pencil, or cut-paper illustration treatment. Keep it clean, sharp, and clearly derived from the source photo.
- Use a source crop only when exact fidelity is essential.
- The main motif should occupy roughly 28% to 38% of the card width and remain subordinate to the photo.
- Use exactly three tiny source-derived color swatches: dominant color, dark structural color, and pale neutral or accent color.
- Add only:
- - title
- - optional short subtitle
- - LOCATION
- - DATE
- - small index number
- Leave location and date values blank if the user did not provide them.
- Do not add keyword lists, sample boxes, rows of cutouts, image cards, generic circles, large color blocks, wave doodles, badges, seals, logos, or long text.
- Use warm ivory paper with subtle grain.
- Quality: highest detail quality, sharp and clear, crisp edges, refined hand-drawn texture, low noise, no blur, suitable for later 4× super-resolution upscaling.
- Overall mood: minimal, airy, refined, source-specific, lightly hand-crafted, and collectible.
- ```
- ---
- ## 17. Back Prompt Template
- ```text
- Create a matching unified postcard back using the same portrait 2:3 ratio, paper tone, thin-line style, and restrained visual language as the front.
- Keep it functional and mostly blank.
- Include:
- - thin outer border
- - one vertical divider slightly right of center
- - stamp box in the upper-right
- - 3 or 4 address lines on the right
- - large blank message area on the left
- Optionally add small POST CARD text near the upper-left and tiny metadata or index near the lower-left.
- Do not add a large collage, palette, sample blocks, or decoration that reduces writing space.
- Quality: clean sharp lines, refined paper texture, low noise, no blur, suitable for later 4× super-resolution upscaling.
- ```
- ---
- ## 18. Quick Checklist
- ### Front
- - [ ] original photo embedded
- - [ ] original aspect ratio preserved
- - [ ] thin frame and even paper gap
- - [ ] V3 Lite layout preserved
- - [ ] large whitespace remains
- - [ ] main motif selected for visual attractiveness
- - [ ] dominant color feature preferred
- - [ ] hand-drawn style used when suitable
- - [ ] main motif large enough
- - [ ] at most one small supporting motif
- - [ ] exactly three small color swatches
- - [ ] no keyword list
- - [ ] no sample boxes
- - [ ] no decorative overload
- - [ ] output sharp and high-detail
- ### Back
- - [ ] unified functional layout
- - [ ] writing space preserved
- - [ ] no large collage
- - [ ] output clean and sharp
- ---
- ## 19. One-Line Definition
- **Preserve the V3 Lite layout, embed the original photo unchanged above, choose the most visually attractive source-defining motif below, render it as a large restrained hand-drawn editorial illustration when suitable, retain one optional supporting motif, and always include exactly three small source-derived color swatches.**
README.md · README · 134 lines
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
- # Photo to Zine Postcard
- **中文** · [English](README_EN.md)
- 把你拍摄的照片转换成一套极简、留白充足、带手绘二创元素的 Zine 风格明信片。
- 这个仓库提供一份可直接复用的 [`SKILL.md`](SKILL.md),用于指导 ChatGPT 将一张照片生成一套统一的明信片正反面:
- - **正面** — 上方完整嵌入原始照片并保持原图比例;下方保留大量留白,生成一个来源明确的手绘主元素、极简元数据和 3 个取自原图的色块。
- - **背面** — 使用统一、可书写的明信片布局,包括邮票区、分割线、地址线和留言区域。
- <table>
- <tr>
- <td><a href="assets/forest-homestead.png"><img src="assets/forest-homestead.png" alt="Forest Homestead" /></a></td>
- <td><a href="assets/alpine-glow.png"><img src="assets/alpine-glow.png" alt="Alpine Glow" /></a></td>
- <td><a href="assets/turquoise-lake.png"><img src="assets/turquoise-lake.png" alt="Turquoise Lake" /></a></td>
- </tr>
- <tr>
- <td><a href="assets/green-door.png"><img src="assets/green-door.png" alt="Green Door" /></a></td>
- <td><a href="assets/blue-arc-lake.png"><img src="assets/blue-arc-lake.png" alt="Blue Arc Lake" /></a></td>
- <td><a href="assets/turquoise-basin.png"><img src="assets/turquoise-basin.png" alt="Turquoise Basin" /></a></td>
- </tr>
- <tr>
- <td><a href="assets/lake-at-dusk.png"><img src="assets/lake-at-dusk.png" alt="Lake at Dusk" /></a></td>
- <td><a href="assets/college-garden.png"><img src="assets/college-garden.png" alt="College Garden" /></a></td>
- <td><a href="assets/valley-current.png"><img src="assets/valley-current.png" alt="Valley Current" /></a></td>
- </tr>
- </table>
- 以上为当前的 9 个官方案例,覆盖森林、雪山、湖泊、航拍、建筑、花园和黄昏等不同场景。点击图片可查看原图。
- ## 快速开始
- 1. 把这个仓库提供给 ChatGPT,或者直接上传 [`SKILL.md`](SKILL.md)。
- 2. 上传一张你拍摄的照片。
- 3. 输入:
- ```text
- 请读取这个仓库里的 SKILL.md,并把它作为唯一的设计规范。
- 把我上传的照片生成一套 Photo to Zine Postcard 明信片正反面。
- ```
- 如果 ChatGPT 无法直接访问仓库,也可以下载 `SKILL.md`,与照片一起上传。
- ## 设计原则
- Photo to Zine Postcard 目前遵循 5 个核心原则:
- 1. **原始照片优先**:原图始终是明信片的视觉主体。
- 2. **保持原始比例**:不拉伸、不随意裁切、不重新绘制上方照片。
- 3. **大量留白**:不把画面填满,保持安静、克制的编辑设计感。
- 4. **只选择一个强主元素**:优先提取最有视觉吸引力、最能代表原图的元素。
- 5. **优先手绘二创**:适合时,将主元素转换成克制的水彩 / 水粉 / 墨线 / 拼贴式手绘效果;不适合时再使用原图裁切。
- 目标不是生成一张普通的 AI 海报,而是建立一套适合个人摄影作品的、可打印的明信片系统。
- ## 默认输出
- - 纵向 `2:3`
- - 参考成品尺寸:`100 × 150 mm / 4 × 6 in`
- - 暖白 / 象牙白纸张背景
- - 上方直接嵌入原始照片
- - 下方一个主要手绘元素
- - 可选一个小型辅助元素
- - 固定 3 个取自原图的色块
- - 一张统一、可书写的明信片背面
- - 最高细节质量,适合后续 4× 超分辨率放大
- ## 如何定制自己的版本
- 这个仓库本身就是为了方便 Fork 和二次创作而设计的。
- 你可以修改:
- - 手绘 / 插画风格
- - 纸张纹理
- - 字体和排版语言
- - 明信片比例
- - 标题、地点、日期等元数据布局
- - 正反面结构
- - 主元素选择策略
- - 色块规则
- 建议先阅读 [`docs/CUSTOMIZATION.md`](docs/CUSTOMIZATION.md)。如果改动较大,更推荐创建独立 variant,而不是不断给默认 Skill 增加规则。
- ## 官方案例
- 当前案例包括:
- - [Forest Homestead](assets/forest-homestead.png)
- - [Alpine Glow](assets/alpine-glow.png)
- - [Turquoise Lake](assets/turquoise-lake.png)
- - [Green Door](assets/green-door.png)
- - [Blue Arc Lake](assets/blue-arc-lake.png)
- - [Turquoise Basin](assets/turquoise-basin.png)
- - [Lake at Dusk](assets/lake-at-dusk.png)
- - [College Garden](assets/college-garden.png)
- - [Valley Current](assets/valley-current.png)
- 案例索引见 [`examples/README.md`](examples/README.md)。
- ## 仓库结构
- ```text
- photo-to-zine-postcard/
- ├── SKILL.md
- ├── README.md
- ├── README_EN.md
- ├── CHANGELOG.md
- ├── CONTRIBUTING.md
- ├── LICENSE
- ├── docs/
- │ └── CUSTOMIZATION.md
- ├── examples/
- │ └── README.md
- └── assets/
- ├── forest-homestead.png
- ├── ... # 其余语义化命名的官方案例
- └── README.md
- ```
- ## 当前版本
- **v1.0.0** — Photo to Zine Postcard 首个公开版本。
- 当前版本主要针对 ChatGPT 图片生成进行优化,同时强调清晰边缘、精细纹理和低噪点,方便后续进行 4× 超分辨率放大。
- ## 参与贡献
- 欢迎提交 PR、改进默认 Skill,或者贡献新的衍生风格。详见 [`CONTRIBUTING.md`](CONTRIBUTING.md)。
- ## License
- MIT License,详见 [`LICENSE`](LICENSE)。
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.