@ianneo_ai: Wow! My Skill is ranked #2 on GitHub this week! I just saw Master Zang's project hit #1, and when I clicked in, I laughed—mine is also on the list at #2. It's this Chinese 'Xiaohei' grotesque illustration generation Skill: http://github.com/helloianneo/ian-xiaohei-illustrations…
Summary
The author published a Codex Skill named Ian Xiaohei Illustrations, used to instruct AI agents to generate 16:9 white-background hand-drawn grotesque illustrations for Chinese articles. This repository ranked second on GitHub's weekly trending list.
View Cached Full Text
Cached at: 05/31/26, 09:04 AM
Oh wow! My Skill ranked #2 on GitHub this week! I just saw that Cang’s project hit #1, clicked in, and laughed — mine was on the list too, at #2. It’s this Chinese Xiaohei bizarre text illustration Skill: http://github.com/helloianneo/ian-xiaohei-illustrations… Originally just made it for my own Notion docs: 16:9 white background hand-drawn, with a few red/orange/blue annotations. Never expected it to blow up like this.
helloianneo/ian-xiaohei-illustrations
Source: https://github.com/helloianneo/ian-xiaohei-illustrations
Ian Xiaohei Illustrations
Turn the judgments, flows, statuses, and metaphors in Chinese articles into white-background, hand-drawn, bizarre yet clean text illustrations.
16:9 horizontal | Xiaohei IP | plain white hand-drawn | minimal red/orange/blue Chinese annotations | Codex Skill
What is this repo
Ian Xiaohei Illustrations is a Codex Skill that guides an AI Agent to generate text illustrations for Chinese articles, posts, blogs, Notion docs, and methodology content.
It’s not a general illustration prompt, nor a PPT infographic template. The core goal is: first understand the cognitive anchors in the article, then turn one of its judgments, flows, structures, states, or metaphors into a memorable 16:9 hand-drawn explanatory image.
The default visual IP is “Xiaohei”: a small black solid character with white dot eyes, thin legs, and a blank expression. Xiaohei is not a mascot, not a sticker, not a decorative element standing in the corner — it’s an absurd worker seriously participating in the system’s operation.
In one sentence: Make AI not just “match an image,” but draw a key cognitive action from the article.
Who is it for
Perfect for:
- People writing Chinese articles who need text illustrations and article images
- People creating knowledge-based content, methodology content, AI workflow content
- People who want to turn abstract judgments into concrete metaphors
- People who want a lighter, weirder, more recognizable illustration style than PPT infographics
- People using Codex for content production and want a consistent visual language
Not suitable for:
- People who want commercial illustrations, brand KV, or refined flat illustrations
- People who want traditional PPT infographics, complex architecture diagrams, or flowcharts
- People who want children’s cartoons, cute IPs, or sticker pack styles
- People who want to cram large amounts of text, long explanations, or full course pages into a single image
- People who need strictly editable vector source files
What does it produce
Default output:
- 16:9 horizontal text illustrations
- 4-8 shot lists per article
- Theme, core meaning, structure type, Xiaohei’s action, and suggested Chinese annotations for each image
- Final PNG images, saved to workspace
assets/-illustrations/
Default no output:
- PPTX / PDF / Keynote
- SVG / HTML / Canvas editable graphics
- Commercial posters or cover KV
- Large text-based infographics
Visual Style
This skill uses Ian’s “Xiaohei Bizarre Text Illustration” style by default:
- Pure white background, no paper texture, beige, shadows, or gradients
- Black hand-drawn line art, thin lines, slight jitter
- Lots of white space, subject occupying about 40%-60% of the frame
- Minimal red, orange, blue Chinese handwritten annotations
- Each image expresses only one core action, structure, state, or metaphor
- Xiaohei must participate in the core action, not just be decorative
- Bizarre, creative, clean, but not childish or cute
Example Effects
Two Breakpoints
Two Breakpoints
Sort by Purpose
Sort by Purpose
One Fish, Multiple Eats
One Fish, Multiple Eats
Receiving Path
Receiving Path
Information Well
Information Well
Idea Press
Idea Press
Content Fermentation
Content Fermentation
Trust Bridge
Trust Bridge
These images are style calibration examples, not composition templates. When using, you should reinvent metaphors from the current article, not copy objects and compositions from past examples.
Installation
Clone the repo:
git clone https://github.com/helloianneo/ian-xiaohei-illustrations.git
cd ian-xiaohei-illustrations
Copy the skill to Codex skills directory:
mkdir -p "${CODEX_HOME:-$HOME/.codex}/skills"
cp -R ./ian-xiaohei-illustrations "${CODEX_HOME:-$HOME/.codex}/skills/"
After installation, use in Codex:
Use $ian-xiaohei-illustrations to design and generate 5 Xiaohei bizarre text illustrations for this Chinese article.
How to Use
Only make illustration plan
Use $ian-xiaohei-illustrations Don't generate images yet.
Analyze where the following article is worth illustrating, output a shot list of about 5 images.
For each image, write clearly: after which paragraph, theme, core meaning, structure type, what Xiaohei is doing, suggested Chinese annotation words.
<paste article>
Directly generate text illustrations
Use $ian-xiaohei-illustrations Generate 4 Xiaohei bizarre text illustrations for the following article.
Requirements: 16:9 horizontal, pure white background, black hand-drawn line art, minimal red/orange/blue Chinese handwritten annotations.
<paste article>
Generate a single image for one concept
Use $ian-xiaohei-illustrations Generate a text illustration for "Trust is not shouted, but laid piece by piece with evidence."
The image should be bizarre but clean, Xiaohei must perform the core action.
Remove title or wrong text from image
Use $ian-xiaohei-illustrations Help me edit this image, remove the "Flowchart" title in the top left. Keep everything else unchanged.
More examples at examples/prompts.md.
Workflow
This skill’s workflow is:
- Read the article, Markdown, Notion content, screenshot, or user-provided topic
- Extract core insights, cognitive turning points, process structures, and paragraphs suitable for visualization
- First output a shot list: each image selects only one cognitive anchor
- Choose a structure type for each image: Workflow, system partial, before/after comparison, role state, conceptual metaphor, method layering, map route, or mini comic panel
- Reinvent a low-tech, bizarre but plausible physical metaphor
- Let Xiaohei perform the core action
- Generate each image individually with an image model
- Check against QA checklist: white background, white space, Xiaohei action, Chinese annotations, non-PPT feel, not a copy of old examples
- Save final PNG, report usage and path
Directory Structure
.
├── README.md
├── LICENSE
├── NOTICE.md
├── assets/
│ └── ian-wechat-qr.jpg
├── examples/
│ ├── images/
│ │ ├── 01-two-breakpoints.png
│ │ ├── 02-sort-by-purpose.png
│ │ └── ...
│ └── prompts.md
└── ian-xiaohei-illustrations/
├── SKILL.md
├── agents/
│ └── openai.yaml
├── assets/
│ └── examples/
└── references/
├── style-dna.md
├── xiaohei-ip.md
├── composition-patterns.md
├── prompt-template.md
└── qa-checklist.md
The subdirectory to actually install into Codex is:
ian-xiaohei-illustrations/
The root-level README, LICENSE, NOTICE, and examples are GitHub sharing documents.
Notes
- Chinese text in images: shorter is more stable.
- Each image tells only one core structure; don’t turn the article into a manual.
- Xiaohei must perform the core action; if removing Xiaohei still makes the image fully valid, it means Xiaohei is too decorative.
- Example images only calibrate line density, white space, color restraint, and Xiaohei’s participation; don’t copy compositions.
- AI image models may have typos, hallucinated labels, style drift, or extra titles; check after generation.
- If Chinese typos are severe, prioritize reducing annotation words and regenerate.
Related Projects
- Ian Handdrawn PPT (https://github.com/helloianneo/ian-handdrawn-ppt) — Chinese hand-drawn tech PPT-style page image generation Skill
- Awesome Claude Code Skills (https://github.com/helloianneo/awesome-claude-code-skills) — Curated collection of Claude Code Skills / Agents / Plugins
- Obsidian + Claude AI Second Brain (https://github.com/helloianneo/obsidian-ai-second-brain) — Guide to building personal knowledge base with Obsidian + Claude AI
About the Author
Ian (Yi’en) — Product Designer / Solo Practitioner / AI Builder
Building a one-person company with an AI team.
- GitHub: helloianneo (https://github.com/helloianneo)
- X/Twitter: @ianneo_ai (https://x.com/ianneo_ai)
- Website: ianneo.xyz (https://ianneo.xyz)
- WeChat:
ianneoxyz - Email: [email protected]
Keep Exploring
This Xiaohei illustration Skill is just a small tool in my personal production system built with AI.
If you’re also using AI for content, knowledge bases, workflows, or productization, check out my website: ianneo.xyz (https://www.ianneo.xyz/).
If you just want to observe, follow me on X/Twitter (https://x.com/ianneo_ai).
Interested in Indie Builders Club? Add WeChat: ianneoxyz, note “OPC”.
If scanning is inconvenient, search WeChat: ianneoxyz.
License
MIT License. See LICENSE.
Similar Articles
@op7418: Holy cow, that's amazing! My Skill reached #1 in GitHub's new project Star ranking this week! It's this social media card Skill (guizang-social-card-skill). A group member just discovered it when his Agent was looking for projects…
Developer @op7418's social media card project guizang-social-card-skill took first place in GitHub's new project star ranking this week.
@axichuhai: https://x.com/axichuhai/status/2062146611472400461
Shares 8 curated AI skills, covering basic configuration, product development, and content creation, to boost AI productivity for agents such as Claude Code and CodeX.
@axichuhai: Whoa, a programming guru distilled his engineering experience into an open-source project that topped GitHub trending, with stars soaring past 90k+. The author is former Vercel engineer Mat, known for making complex tech easy to understand and involved in early Next.js development. He distilled his daily collaboration with Claud…
Introducing hello-agents, an open-source project that topped GitHub trending. It systematically organizes courses on AI and Agents from theory to practice, covering skills like Agentic RL, SFT, GRPO. Created by former Vercel engineer Mat, it distills engineering experience into 16 skills, such as Grim prompting techniques and red-green cycle testing methods.
@GitTrend0x: GitHub Agent & AI Tools Dominate the Trending List Again: Deep Dive into the Top 5 Projects with Explosive Star Growth, Professional Breakdown + Practical Scenarios, All in One Article! 1. anthropics/financial-services: Anthropic’s official Financial Services Agent Framework! Supports complex…
The article reviews the top five AI Agent projects on GitHub with the fastest star growth recently, highlighting Anthropic's Financial Services Agent Framework, ByteDance's UI-TARS Desktop, and various coding Agent tools.
@GitTrend0x: 46 AI agents 杀手级开源神器 https://github.com/rohitg00/skillkit… 这就是 SkillKit,1k star 爆款 AI coding agent 技能包管理器!
SkillKit is a package manager for AI coding agent skills, supporting 46 agents and 400K+ skills from 31 sources, allowing one skill to be used across multiple agents.