Skip to content
SkillModel-led1,403 tokens in use

Ian Xiaohei Illustrations

Plans and generates 16:9 hand-drawn illustrations for Chinese articles, built around a black mascot called Xiaohei doing the key action of each idea.

13,304 stars on the repository MIT Licensehelloianneo/ian-xiaohei-illustrations · ian-xiaohei-illustrations
git clone https://github.com/helloianneo/ian-xiaohei-illustrations.git
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing Ian Xiaohei Illustrations costs

168tokens idle
1,403tokens in use
5,548tokens, up to

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Up to adds the files its instructions link. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

Nothing detected: no shell, network, file writes, git or credentials in its files.

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Fair
  • Frontmatter8/15
  • Structure8/25
  • Examples12/15
  • Output0/15
  • Size13/15
  • Resources12/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent

No findings.

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Fair
  • Development activity44
  • Maintenance28
  • Community60
  • Maturity24
  • Momentum40

This skill reads a Chinese article, picks the few paragraphs that carry a key judgment, process or turning point, and turns each into one white-background, hand-drawn 16:9 illustration. The recurring character is Xiaohei, a solid black figure with dot eyes and thin legs, who must take part in the main action of every picture.

It first produces a shot list (where each image goes, its theme, the structure type and suggested labels), and when you ask for images it generates them one at a time with the built-in image tool and saves them in your workspace. The instructions are in Chinese.

What You Get

  • A shot list with placement, theme, structure type, Xiaohei’s action and suggested Chinese labels
  • One 16:9 PNG per chosen anchor point, generated separately
  • Files saved as numbered PNGs under an assets folder named after the article
  • A short delivery note listing each image’s purpose and which ones are safest
  • Image edits such as removing a stray title or a wrongly written character

Common Use Cases

  • Illustrate a Chinese blog post or newsletter with a consistent hand-drawn look
  • Turn a method or workflow document into a few explanatory pictures
  • Plan illustrations first, without generating anything, and review the shot list
  • Remove an unwanted heading from a generated image and keep the rest

What Makes It Unique

The skill is about one mascot with fixed rules and one drawing style, and it tells the agent to reinvent the metaphor for each article and never copy the sample compositions. It includes a QA checklist that rejects images where the mascot is only decoration, the page is too full, or the result looks like a slide.

Under The Hood

SKILL.md is the entry point and points to reference files for style, the mascot, composition patterns, a prompt template and a QA checklist, which the agent reads as needed. The skill has no scripts. The agent reads your article, link or Notion page, calls the built-in image generation tool once per picture, checks each result against the checklist, regenerates when it fails, and copies the final PNGs into an assets folder in your workspace without overwriting existing files.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers