yanliudesign/mono-color-skill

★ 3,175⑂ 85

One-ink editorial print image skill — warm paper, halftone photography, active negative space, and restrained typography.

About yanliudesign/mono-color-skill

yanliudesign/mono-color-skill is an open-source project on GitHub, mainly written in Python. One-ink editorial print image skill — warm paper, halftone photography, active negative space, and restrained typography. It currently holds 3,175 stars and 85 forks with 0 open issues, and was last pushed on an unknown date (repository created unknown).

Project Overview

AI Homed tracks it on the AI Prompt Engineering board.

GitHub Repository Details

Repository yanliudesign/mono-color-skill · default branch - · size 0 KB · watchers 0 · source: GitHub REST API and repository README

README

中文 · English

Monocolor Editorial Print

A one-ink and controlled two-ink editorial image skill for posters, zines, portraits, packaging, and visual field notes.

Version Skills Stars Validate skill

Claude Code SKILL.md

Turn a theme, sentence, object, or supplied photograph into an original editorial image. The skill defaults to controlled two-ink with one dominant plate and one narrowly assigned accent plate. Explicit one-ink or monochrome requests remain pure one-ink. White, cool gray, or pale-beige substrates are selected from the image and palette; contemporary editorial is the default, while vintage treatment appears only when requested.

It preserves a visual system rather than copying a reference. Every composition is rebuilt around the subject, intent, words, and image role.

Selected examples

| Summer cycling | Field study | Ordinary spaces | |:---:|:---:|:---:| | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Cobalt and terracotta summer cycling editorial print | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Cobalt and orange zebra field-study editorial print | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Green and oxblood chair zine editorial print |

| Sardine packaging | Headphone packaging | Sunscreen packaging | |:---:|:---:|:---:| | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Violet and orange sardine tin packaging | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Cobalt and black headphone packaging | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Cyan and coral sunscreen packaging |

| A little warmth | Brand merchandise | Slow leaf | |:---:|:---:|:---:| | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Green halftone teapot editorial poster | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Cobalt and orange mono-color merchandise collection | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Green halftone tea packaging |

| Night photography | Sunday radio | Night market | |:---:|:---:|:---:| | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Cobalt urban night photography exhibition poster | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Cobalt and black radio listening poster | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Red and cyan mushroom night market poster |

These original generated examples demonstrate the skill's range; they are outputs, not templates to reproduce.

Visual references

The visual system was informed by a twelve-image research set. See Visual References and Attribution for the complete image index, research notes, attribution status, and correction link. Third-party references remain the property of their respective creators and rights holders.

| 01 | 02 | 03 | |:---:|:---:|:---:| | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Visual reference 01 | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Visual reference 02 | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Visual reference 03 | | 04 | 05 | 06 | | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Visual reference 04 | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Visual reference 05 | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Visual reference 06 | | 07 | 08 | 09 | | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Visual reference 07 | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Visual reference 08 | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Visual reference 09 | | 10 | 11 | 12 | | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Visual reference 10 | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Visual reference 11 | https://github.com/yanliudesign/mono-color-skill/blob/HEAD/Visual reference 12 |

What it does

| System | Direction | |---|---| | Input | A theme, phrase, object, article idea, or supplied photograph | | Palette | Adaptive neutral white, cool gray, or pale beige + controlled two-ink by default; explicit one-ink requests stay one-ink | | Modes | Pure one-ink, chromatic + black, complementary duotone, or overprint duotone | | Image | Halftone, risograph grain, cyanotype exposure, or photocopy breakup | | Space | 25%–55% visible empty paper on an asymmetric editorial grid | | Type | Content-responsive literary serif, cultural grotesk, condensed civic, programmatic, rotated, handwritten, or word-as-object roles | | Output | Generated raster image, exact production prompt, and a short recipe |

Use cases

All of these formats can use a single ink to reduce production costs while retaining a distinctive visual identity.

How it works

1  Read the input       →  identify subject, intent, words, and image role
2  Choose the layout    →  image, specimen, declaration, object field, overprint, journal, or cover
3  Assign the plates    →  default to two assigned roles; switch to one plate when explicitly requested
4  Compose the page     →  preserve 25%–55% silence and add one deliberate disruption
5  Generate and inspect →  check ink count, identity, hierarchy, texture, and originality

Ink system

One-ink palette: Cobalt, Royal Blue, Botanical Green, Mint Green, Terracotta Orange, Signal Red, Aubergine, and Charcoal.

| Swatch | Ink | Hex | |---|---|---| | Cobalt | Cobalt / Ultramarine | #2148B8 | | Royal Blue | Royal Blue | #2058D4 | | Botanical Green | Botanical Green | #008A4B | | Mint Green | Mint Green | #5EB783 | | Terracotta Orange | Terracotta Orange | #C65F38 | | Signal Red | Signal Red | #C83232 | | Aubergine | Aubergine | #63365F | | Charcoal | Charcoal | #30343A |

Two-ink recipes: Powder Blue + Signal Red, Cobalt + Terracotta, Botanical Green + Oxblood, Charcoal + Signal Red, Electric Blue + Carbon, Mint Green + Charcoal, Ultramarine + Safety Orange, Cyan + Brick Red, and Tangerine + Slate Blue.

| Swatches | Two-ink recipe | Hex | |---|---|---| | Powder Blue and Signal Red | Powder Blue + Signal Red | #9EB8D3 + #C83232 | | Cobalt and Terracotta | Cobalt + Terracotta | #2148B8 + #C65F38 | | Botanical Green and Oxblood | Botanical Green + Oxblood | #008A4B + #8F3434 | | Charcoal and Signal Red | Charcoal + Signal Red | #30343A + #C83232 | | Electric Blue and Carbon | Electric Blue + Carbon | #173AE3 + #242321 | | Mint Green and Charcoal | Mint Green + Charcoal | #5EB783 + #302D2E | | Ultramarine and Safety Orange | Ultramarine + Safety Orange | #263E99 + #E55D2B | | Cyan and Brick Red | Cyan + Brick Red | #159DDA + #B64032 | | Tangerine and Slate Blue | Tangerine + Slate Blue | #E46C2D + #4773A5 |

In two-ink work, the dominant plate normally carries 70%–85% of the printed area. The accent plate carries 15%–30% and must have a specific job such as dates, annotations, selected objects, or overprint intersections. The substrate is not a third color, and the darker color created where two plates overlap is not a third ink.

Substrate is adaptive, not nostalgic by default: Neutral White #FAFAF7 supports crisp cultural, social, event, and image-led work; Cool Gray #E9E9E5 supports architecture, technology, charcoal-led systems, and restrained branding; Pale Beige #F5F1E8 supports tactile, travel, food, intimate, archival, or explicitly nostalgic subjects. Limited inks and halftone do not automatically imply vintage styling.

Visual rules

1. No more than two printing inks. Controlled two-ink is the default: 70%–85% dominant plate and 15%–30% accent plate, each with a distinct role. Explicit one-ink requests use one plate. 2. Paper stays visible. The result is a printed page, not a digitally tinted monochrome wash. 3. Mechanical reproduction leads. Photographs become dots, grain, clipped highlights, ink pooling, and mild registration drift between plates. 4. Silence has structure. Empty paper occupies 25%–55% of the canvas and controls pacing. 5. Type has tension and range. Choose one content-responsive display skeleton and one utility voice; use handwriting only as a short optional interjection. A series may change display categories instead of repeating one house treatment. 6. Identity stays intact. Supplied people, objects, and scenes remain recognizable. 7. References are grammar, not templates. At least four structural variables change from every supplied reference.

Not this

Install

Clone the repository into your Claude Code skills directory:

git clone https://github.com/yanliudesign/mono-color-skill.git \
	~/.claude/skills/mono-color

Restart Claude Code after installation. Other agent environments can load SKILL.md as the skill entry point.

Try it

Use mono-color to make a vertical poster about a midnight convenience store.
The exact headline is “still open”.
Turn this portrait into a cobalt one-ink editorial zine cover.
Keep the person's identity and expression.
Create a botanical-green risograph field note about ferns.
The title is “field note 07”.
Create an Ultramarine + Safety Orange overprint poster about city cycling.
Let the bicycle and oversized title cross in selected areas.
Turn this product into a Cyan + Brick Red repeated-object cover.
Keep one open zone for the title and factual microcopy.
Make a Signal Red one-ink party poster for a rooftop gathering.
Use “after sunset” as the headline and keep the date small.
Design a Charcoal + Signal Red exhibition poster about concrete architecture.
Use the red plate only for the date, location, and one geometric interruption.
Create a cobalt city-walk poster from this street photograph.
Turn the buildings into coarse halftones and title it “north by foot”.
Make a Terracotta Orange Xiaohongshu cover about a weekend flea market.
Use one oversized Chinese headline and leave at least one third of the paper empty.
Turn this host portrait into an Aubergine one-ink podcast cover.
Keep the face recognizable and use “the quiet hour” as the episode title.
Create a Botanical Green + Oxblood invitation for an independent bookstore opening.
Let green carry the paper texture and image; reserve oxblood for event details.
Transform these travel photographs into a Cobalt + Terracotta postcard series.
Keep the same grid across the set, but vary each crop and handwritten annotation.
Design a Mint Green + Charcoal poetry title page for the line “we kept the window open”.
Use no photograph, generous paper space, and one small archival-style note.

Delivery format

One run returns:

1. a generated raster image when image-generation tools are available; 2. the exact production-ready prompt used for generation; 3. a recipe naming the print mode, exact ink palette, layout family, type pairing, print process, and originality changes.

If image generation is unavailable, the skill returns the production-ready prompt and states the limitation.

Stability and validation

Before prompt compilation, the skill resolves every request into a fixed recipe manifest. Unspecified requests use a 3:4 ratio, Neutral White substrate, contemporary editorial direction, 35% empty substrate, and deterministic palette and layout rules. The substrate may switch to Cool Gray or Pale Beige when the image and ink contrast call for it. Supplied photos default to faithful reproduction; requests for abstract, artistic, loose, experimental, or less realistic treatment switch to deterministic symbol extraction that preserves 2-4 identity anchors. Explicit user choices still take precedence within the two-ink and originality limits.

The design-system/ catalogs make the visual grammar reusable and inspectable. They separate color tokens, typography roles, composition geometry, carrier-specific signals, visual rhythm, and controlled print imperfections from the prose workflow. Catalog IDs are the shared contract between reference boards, recipes, and validation.

design-system/rhythm.json defines relaxation as uneven energy rather than uniformly reduced intensity. Each page selects one audacious focal event—oversized type, an extreme crop, one giant detail, a concentrated overprint, or an abnormal scale relationship—then releases the rest through paper, pale screening, and sparse functional type. Without a supplied photograph, people become 2-4 identifying anchors and partial crops instead of complete stock figures or a safe headline-left/photo-right split. Empty paper and unresolved edges now respond to the focal event instead of fixed quotas.

Controlled chance stays in the reproduction layer: contemporary work selects 0-2 restrained effects, while tactile, vintage, or archival-aging work selects 2-3 bounded effects such as uneven ink density, dry-edge breakup, halftone drift, registration drift, or one broken gesture. The stable recipe seed preserves the same marks across retries without moving the composition or reducing text readability.

Mono-color visual system reference board

The four focused reference-analysis boards document the typography, color, layout, and style evidence behind the system:

Typography

Typography system

Color

Color system

Layout

Layout system

Style

Style system

Regenerate the catalog-driven complete board after editing a catalog:

python3 scripts/build_design_system_board.py

The evaluation contract covers defaults, supplied portraits, botanical work, overprint, event information, long-form text, prompt-only output, conflicting color requests, repeated objects, and reference-copying requests. Run it locally with:

python3 scripts/validate_evals.py
python3 scripts/validate_design_system.py

GitHub Actions runs the same contract on every pull request and push to main.

Repository layout

mono-color-skill/
├── .github/workflows/ # Continuous validation
├── design-system/    # Machine-readable color, composition, rhythm, and print patterns
├── examples/         # Original generated examples shown in the READMEs
├── scripts/          # Evaluation and design-system validators
├── swatches/         # One-ink and two-ink palette previews
├── SKILL.md          # Trigger rules, visual system, workflow, and quality gate
├── README.md         # English documentation
├── README.zh.md      # 中文说明
├── CHANGELOG.md      # Release history
└── evals/
		├── evals.json    # Representative prompts and deterministic assertions
		└── schema.json   # Evaluation contract schema

Originality

This skill extracts system-level qualities such as palette, print process, spacing, hierarchy, and communication tone. It does not copy a reference's composition, wording, labels, logos, border system, or distinctive arrangement.

When a user supplies a photograph, the subject is preserved as content while the crop, screening, grid, type placement, and metadata treatment are newly constructed.

License

The source code, skill instructions, and scripts are available under the MIT License.

Original examples by Yan Liu in examples/ are © 2026 Yan Liu and are not covered by the MIT License. The twelve third-party research references remain the property of their respective creators and rights holders. See the Visual Asset License and Visual References and Attribution.

---

Created by Dreameryanyan · LinkedIn · X

GitHub Stars & Activity

3,175Stars
85Forks
0Open issues
PythonLanguage

GitHub Popularity

GitHub stars3,175
Forks85
Open issues0
Primary languagePython
License-
Stars gained today0
Created-
Last pushed-

Trending History

Trending statusnot on today's boards

Related AI Projects

1

headroomlabs-ai / headroom

Python★ 73,186⑂ 5,631
2

blader / humanizer

Python★ 50,467⑂ 4,061
3

Imbad0202 / academic-research-skills

Python★ 48,823⑂ 3,789
4

mlflow / mlflow

Python★ 28,051⑂ 6,328
5

alirezarezvani / claude-skills

Python★ 26,168⑂ 3,686
6

comet-ml / opik

Python★ 22,157⑂ 1,809
7
8

tradecatlabs / vibe-coding-cn

Python★ 16,314⑂ 1,653

More AI Rankings