Skip to content
Back to search

gc-minimal-zine-poster

Codex skill for generating quiet minimal zine-style editorial poster prompts and images - 转存自 LiamGvchi/gc-minimal-zine-poster

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/yub369302-cyber--gc-minimal-zine-poster/badge.svg)](/listings/yub369302-cyber--gc-minimal-zine-poster)

HTML

<a href="/listings/yub369302-cyber--gc-minimal-zine-poster"><img src="/api/listings/yub369302-cyber--gc-minimal-zine-poster/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Clone the public repository directly into the Codex skills directory:

git clone https://github.com/LiamGvchi/gc-minimal-zine-poster.git \
  ~/.codex/skills/gc-minimal-zine-poster-v0-1

Restart Codex if the skill does not appear immediately.

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 · 244 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: gc-minimal-zine-poster-v0-1
  3. description: Generate Minimal Zine Poster v0.1 poetic paper-poster prompts and the matching generated image. Use when the user gives a theme, sentence, object, mood, article idea, photo, or content brief and wants a quiet Japanese/Korean zine-like editorial poster with large negative space, aged paper texture, experimental typography, restrained color accents, and a generated bitmap image.
  4. ---
  5. # Minimal Zine Poster v0.1
  6. Turn the user's content into both:
  7. 1. a final image-generation prompt, and
  8. 2. a generated raster image made from that prompt.
  9. ## Mode Policy
  10. Use **Standard Mode** for all generation. Use the Standard Mode Prompt Compiler in this `SKILL.md` to convert the user's content into a compact, imageable, high-fidelity prompt. If the user asks for higher quality, strengthen the prompt using the rules below.
  11. ## Standard Mode Prompt Compiler
  12. Default generation should compile only the parts that become pixels in the final image prompt.
  13. ### Visual Rules Used by the Prompt Compiler
  14. Use these rule groups as prompt material:
  15. - **风格总述:** use only the visual identity and anti-identity: poetic minimal paper poster, huge negative space, old paper, tiny anchor, sparse type, one clear high-chroma anchor, zine/editorial mood.
  16. - **核心视觉规则:** use the concrete renderable rules for canvas, composition, background, image anchor, typography, color, texture, lighting, and mood.
  17. - **稳定共性:** use as non-negotiable must-haves: vertical 3:5 paper canvas, small cluster, scanned-paper view, old print defects, serif/typewriter text, and a saturated color anchor visible at thumbnail size.
  18. - **可替换变量:** use as slot choices: object, photo/cutout/silhouette/block type, accent color, text line, date/weather, position, paper tone.
  19. - **反向约束:** use as negative prompt material.
  20. - **Prompt 结构模板:** use its field order, not its sample wording.
  21. Do not use these as default prompt material:
  22. - source path, sample count, README/metadata notes, or analysis scope
  23. - long explanatory prose about why the style works
  24. - sample-specific signatures, dates, captions, objects, or text
  25. - example prompts as text to imitate line by line
  26. - checklist phrasing unless it becomes a concrete visual constraint
  27. ### First-Principles Prompt Fields
  28. Every Standard Mode prompt must answer these rendering questions in this order:
  29. 1. **Canvas:** What is the output frame and base surface?
  30. - tall vertical 3:5 phone-poster; full-frame aged paper; no border, no mockup.
  31. 2. **Attention Geometry:** Where does the eye go and how much is empty?
  32. - 70%-90% plain paper; one visual cluster occupying about 8%-25%; placed center, upper-middle, lower-middle, lower-left, or upper-right; no edge-hugging.
  33. 3. **Image Anchor:** What is the one imageable subject?
  34. - convert the user's theme into one object, fragment, photo crop, specimen, cutout, silhouette, old printed illustration, texture window, or small conceptual relation.
  35. 4. **Anchor Treatment:** What material process makes the anchor belong to paper?
  36. - grayscale photos and paper fragments may use low contrast, photocopy softness, torn edge, softened edge, halftone, scanline, risograph grain, xerox wear, ink bleed, or slight misregistration. Do not apply low saturation or low contrast to the chosen color anchor.
  37. 5. **Typography System:** How does text behave visually?
  38. - small serif/typewriter/monospaced type; one short readable phrase; optional tiny date/location/weather and signature; semi-legible microtext or fragmented letters; text can drift, press against the image edge, blur, or misregister.
  39. 6. **Color Logic:** What is the restrained accent strategy?
  40. - paper tones plus gray/black support one unmistakably high-chroma anchor. Prefer cobalt or ultramarine; rotate through cyan, violet, magenta-pink, lemon yellow, pear green, orange, or tomato red. The color may be the subject, a flat silhouette, an irregular cutout, a substantial block, a partial-color photo region, or bold fragmented type. It must not be reduced automatically to a tiny dot or hairline.
  41. 7. **Reproduction Texture:** What print/scanning process defines the whole image?
  42. - flat orthographic scanned-paper appearance; matte absorbent paper; diffuse light; low-to-medium contrast; no hard shadow; no 3D depth.
  43. 8. **Emotional Temperature:** What should the viewer feel before identifying the object?
  44. - quiet, poetic, nostalgic, sparse, diary-like, archival, distant, memory-like, Japanese/Korean indie zine or minimal editorial.
  45. 9. **Hard Avoids:** What must not appear?
  46. - full-bleed scene, commercial headline, product ad, logo/CTA, glossy mockup, clean UI white, cinematic lighting, 3D, neon, cute cartoon, fashion editorial drama, dense scrapbook, too many colors, long clean text.
  47. ### Standard Color Engine
  48. This section defines the color strategy for Standard Mode.
  49. - Default to one visibly saturated, opaque chromatic ink anchor. Use wording such as `fully saturated cobalt-blue risograph ink`, `opaque ultramarine cutout`, `vivid pear-green flat silhouette`, or `clean tomato-red printed block`.
  50. - Keep the paper, grayscale photo, microtext, and secondary marks subdued. Preserve saturation in the color anchor even when adding grain, halftone, ink bleed, or misregistration.
  51. - The high-chroma area should occupy roughly 0.8%-2.5% of the whole canvas or 15%-35% of the small visual cluster. It must remain visible when the image is viewed as a thumbnail.
  52. - Color can carry the subject itself. Prefer a colored tree, fruit, shell, flower, geometric cutout, window, poster fragment, or image panel over a gray object with one colored registration tick.
  53. - For a single image, use a substantial color anchor by default. For batches, at least 60% of images must use a colored subject, cutout, or block; the remaining images may use dots, hairlines, or colored type for rhythm.
  54. - Do not use `near-monochrome`, `no strong accent`, `pale accent`, `muted accent`, `faded accent`, or `pastel accent` unless the user explicitly requests monochrome, muted, or pastel output.
  55. - Do not describe the entire image as low saturation. Apply `low contrast` and `muted grayscale` only to paper, photos, and secondary ink.
  56. - Use only one main high-chroma hue per image. A tiny secondary hue is allowed only when it supports the subject and does not make the poster commercially colorful.
  57. ### Standard Prompt Shape
  58. Write the final Standard Mode prompt as four compact paragraphs:
  59. 1. canvas + paper + negative space + cluster size/location
  60. 2. subject metaphor + anchor type + anchor treatment
  61. 3. typography + accent strategy + print defects
  62. 4. flat scan mood + avoid-list
  63. In paragraph 3, state the exact high-chroma hue, its material form, and its approximate visual share. This structure is more important than reciting every rule. Prefer a concrete, imageable prompt over a long style essay.
  64. ## Variation Engine
  65. Before writing the prompt, choose one option from each axis. Randomness must change visual grammar, not only position. If recent outputs used the same layout or anchor, choose a different one.
  66. ### Layout Family
  67. - **center-fragment:** tiny central image or object with surrounding air
  68. - **lower-left-float:** small anchor in the lower-left quadrant, lots of empty top space
  69. - **upper-right-block:** small color/photo block in the upper-right with loose text drift
  70. - **dual-panel:** two small overlapping or adjacent panels with a narrow gap
  71. - **irregular-cutout:** torn or organic paper shape carrying image or type
  72. - **type-led:** typography is the main visual anchor, image secondary or absent
  73. - **dot-orbit:** dots, letters, or hairline create an orbit around a small subject
  74. - **single-specimen:** one isolated object or mark with almost no support graphics
  75. ### Image Anchor
  76. - tiny faded photo
  77. - torn-paper clipping
  78. - flat silhouette
  79. - solid color block
  80. - old printed illustration
  81. - object specimen
  82. - translucent geometric overlay
  83. - abstract texture window
  84. ### Typography Mode
  85. - fragmented floating letters
  86. - short phrase pressed against image edge
  87. - archive microtext with date/weather
  88. - diagonal scattered words
  89. - low-contrast gray ghost text
  90. - headline-as-object with rough letterpress
  91. - text inside a color block or cutout
  92. - almost textless, only a tiny caption
  93. ### Texture Mode
  94. - xerox softness
  95. - risograph grain
  96. - letterpress ink bleed
  97. - halftone degradation
  98. - film grain photo
  99. - scan noise and paper fibers
  100. - aged paper mottling
  101. - soft motion blur on selected text
  102. ### Mood Mode
  103. - quiet
  104. - summer
  105. - solitude
  106. - childhood
  107. - seaside
  108. - afternoon
  109. - night
  110. - memory
  111. - slight surrealism
  112. ## Workflow
  113. 1. Determine mode.
  114. - Use Standard Mode.
  115. 2. Parse the user's content.
  116. - Identify the core subject, mood, exact text if supplied, possible visual metaphor, and any reference image role.
  117. - For an article or complex idea, extract one central imageable idea rather than summarizing the whole argument.
  118. - If no image text is supplied, invent one short poetic English or Chinese phrase.
  119. 3. Select a variation recipe.
  120. - In Standard Mode, pick layout, image anchor, typography, texture, and mood from the Variation Engine, then choose color through the Standard Color Engine. Do not select `near-monochrome` unless the user explicitly asks for it.
  121. - Do not default to "tiny photo + blue dots + microtext" unless it truly fits.
  122. - If the recipe becomes too dense, simplify typography or color treatment first.
  123. 4. Write the final image prompt.
  124. - In Standard Mode, use the Standard Mode Prompt Compiler to compile the user's content into the four-paragraph prompt shape: canvas, anchor, typography/accent/print, flat-scan mood and avoid-list.
  125. - Specify exact in-image text only when useful. Keep it short because image models distort long text.
  126. - Make the prompt decisive: say where the anchor sits, how large it is, how text behaves, what accent appears, and how the print/scan texture looks.
  127. 5. Generate the image.
  128. - Use the built-in image generation capability by default.
  129. - Do not stop after prompt-only unless the user explicitly asks for prompt-only.
  130. - If the result obviously violates the selected mode or recipe, tighten the prompt and regenerate once.
  131. - In Standard Mode, inspect the result at thumbnail scale. If the high-chroma anchor is absent, washed out, or reduced to an imperceptible mark, regenerate once with stronger color wording and a larger colored area.
  132. 6. Return the image and prompt.
  133. ## Negative Constraints
  134. Always avoid:
  135. - full-bleed subject or scene
  136. - commercial poster headline hierarchy
  137. - product ad layout, logo lockup, CTA, or brand campaign feeling
  138. - clean digital UI background
  139. - glossy paper mockup or heavy paper shadow
  140. - 3D rendering, cinematic lighting, hard shadows, depth of field, neon, cyberpunk
  141. - cute cartoon, kawaii illustration, anime poster, fashion editorial drama
  142. - too many objects, stickers, colors, captions, or decorative textures
  143. - high-resolution stock-photo realism
  144. - long, clean, perfectly readable text blocks
  145. ## Output Format
  146. ````markdown
  147. **生成图**
  148. ![Minimal Zine Poster v0.1 style poster](absolute-image-path-or-rendered-image)
  149. **最终 Prompt**
  150. ```text
  151. [final prompt used for image generation]
  152. ```
  153. **说明**
  154. - Mode: Standard
  155. - Recipe: [layout / anchor / typography / accent / texture / mood]
  156. - [one short note about the content interpretation]
  157. ````
  158. If generated images render directly without a file path, show the image normally and still include the final prompt.
  159. ## Quality Gate
  160. Before finalizing, check:
  161. - Did the run use the Standard Mode Prompt Compiler?
  162. - Did the run choose a variation recipe across layout, anchor, typography, accent, texture, and mood?
  163. - Is the structure materially different from recent visible outputs?
  164. - Does the image remain a sparse vertical paper poster?
  165. - Does 70%-90% of the poster read as paper?
  166. - Is the subject cluster roughly 8%-25% of the canvas?
  167. - Is there one clear visual metaphor rather than a whole illustrated scene?
  168. - Does the anchor have old-photo, clipping, print, scan, or paper-specimen treatment?
  169. - Are typography and microtext part of the composition?
  170. - Is there only one restrained accent strategy?
  171. - In Standard Mode, is the high-chroma anchor clearly visible at thumbnail size?
  172. - In Standard Mode, does saturated color occupy about 0.8%-2.5% of the canvas or 15%-35% of the visual cluster?
  173. - In Standard Mode, did the prompt avoid weakening the color anchor with `pale`, `muted`, `faded`, `pastel`, `low saturation`, or `near-monochrome` wording?
  174. - Did the prompt avoid full-bleed, commercial, 3D, neon, cinematic, cartoon, cute, brand, and generic template aesthetics?
  175. - Did you actually generate the image?
  176. ## Example Requests
  177. - "用 $minimal-zine-poster-v0.1 做一张关于雨天的图"
  178. - "用 $minimal-zine-poster-v0.1 标准模式,做一张关于旧书的图"
  179. - "用这张照片做一张同风格 poster"
README.md · README · 80 lines

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

  1. # GC Minimal Zine Poster
  2. **English** · [简体中文](README.zh-CN.md) · [日本語](README.ja.md)
  3. A Codex skill for turning a theme, sentence, object, mood, article idea, photo, or content brief into a quiet minimal zine-style editorial poster prompt and a generated raster image.
  4. The callable skill name is `gc-minimal-zine-poster-v0-1`.
  5. ## Visual Direction
  6. The skill compiles each request into a sparse vertical paper poster with:
  7. - a 3:5 aged-paper canvas
  8. - 70%-90% negative space
  9. - one small imageable subject or visual cluster
  10. - serif, typewriter, or monospaced typography
  11. - one clearly visible high-chroma color anchor
  12. - xerox, risograph, halftone, letterpress, or scanned-paper defects
  13. - a quiet Japanese/Korean indie-zine or minimal editorial mood
  14. It avoids commercial advertising layouts, glossy mockups, cinematic lighting, 3D rendering, neon, dense scrapbooks, and long clean text blocks.
  15. ## Examples
  16. | Night Door | Yellow Step |
  17. | --- | --- |
  18. | ![Night Door](examples/night-door.jpeg) | ![Yellow Step](examples/yellow-step.jpeg) |
  19. | Shore Pause | Pause Map |
  20. | --- | --- |
  21. | ![Shore Pause](examples/shore-pause.jpeg) | ![Pause Map](examples/pause-map.jpeg) |
  22. | Typhoon Memory | Moon Tide |
  23. | --- | --- |
  24. | ![Typhoon Memory](examples/typhoon-memory.jpeg) | ![Moon Tide](examples/moon-tide.jpeg) |
  25. ## Installation
  26. Clone the public repository directly into the Codex skills directory:
  27. ```bash
  28. git clone https://github.com/LiamGvchi/gc-minimal-zine-poster.git \
  29. ~/.codex/skills/gc-minimal-zine-poster-v0-1
  30. ```
  31. Restart Codex if the skill does not appear immediately.
  32. ## Usage
  33. Invoke the skill by name and provide a theme or brief:
  34. ```text
  35. 用 $gc-minimal-zine-poster-v0-1 做一张关于雨天旧书店的海报
  36. ```
  37. You can also provide a sentence, article idea, object, mood, or reference image.
  38. ## Output
  39. For every generation request, the skill returns:
  40. 1. the generated raster poster image
  41. 2. the final image-generation prompt
  42. 3. the selected variation recipe and a short interpretation note
  43. The workflow uses Standard Mode and generates the image by default. It only stops at prompt-only output when the user explicitly asks for that.
  44. ## Repository Structure
  45. - `SKILL.md`: the complete Codex skill instructions
  46. - `README.md`: public overview and installation instructions
  47. - `LICENSE`: MIT license
  48. - `examples/`: selected generated posters
  49. This repository publishes one standalone skill. A separate private vault may aggregate backups of multiple local skills, but private backup automation and unrelated skills are intentionally excluded here.
  50. ## License
  51. MIT. See `LICENSE`.

Ratings and comments

Sign in to rate this skill.

Sign in

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