Ian Xiaohei Illustrations

Turn judgments, flows, states, and metaphors in Chinese articles into white-background, hand-drawn, grotesque yet clean in-text illustrations.

16:9 horizontal | Xiaohei IP | white hand-drawn | minimal red-orange-blue Chinese annotations

Use Cases

Suitable for:

• People writing Chinese articles who need in-text illustrations
• Those creating knowledge-based content, methodology content, or AI workflow content
• Anyone wanting to visualize abstract judgments as concrete metaphors
• Those seeking a lighter, weirder, more distinctive illustration style than PPT infographics
• People using AI agents for content production who want a consistent visual language

Not suitable for:

• Commercial illustrations, brand key visuals, or polished flat illustrations
• Traditional PPT infographics, complex architecture diagrams, or flowcharts
• Children's cartoons, cute IP, or sticker-style art
• Packing large amounts of text, long explanations, or full course pages into a single image
• Strictly editable vector source files

Preview

The following images are style calibration samples, not composition templates. When using the skill, invent new metaphors for each article rather than copying the compositions.

Core Capabilities

• Identify cognitive anchors (judgments, flows, structures, states, metaphors) in Chinese articles and generate 16:9 horizontal in-text illustrations
• Output a shot list: theme, core meaning, structure type, Xiaohei's action, and suggested Chinese annotations for each image
• Each image expresses only one core action or structure, avoiding information overload
• Default visual IP "Xiaohei" must participate in the core action, not just decorate
• Visual style: white background, black hand-drawn fine lines, generous whitespace, minimal red-orange-blue Chinese hand-written annotations
• Support planning-only mode (output shot list without generating images)
• Support direct generation of multiple images
• Support generating a single image for one concept
• Support editing generated images (e.g., remove a title or incorrect text)

How It Works

1. You provide a Chinese article, Markdown, Notion content, a screenshot, or a topic description
2. The skill extracts key points, cognitive transitions, structural elements, and visualizable paragraphs
3. First, it outputs a shot list: each image selects one cognitive anchor
4. For each image, it chooses a structure type: Workflow, System Partial, Before/After, Character State, Conceptual Metaphor, Method Layers, Map Route, or Mini Comic Panel
5. It invents a low-tech, grotesque but plausible physical metaphor
6. Xiaohei takes on the core action
7. Each image is generated independently by an image model
8. A QA checklist is applied: white background, whitespace, Xiaohei's action, Chinese annotations, non-PPT feel, no copy of old examples
9. Final PNGs are saved, and the usage and paths are reported

What You Need to Provide

• A Chinese article (paste text, Markdown, or Notion content)
• Or a screenshot of the article content
• Or a direct description of a topic/concept
• Optional edit requests (e.g., "remove the title in the top-left corner")

What You Get

• 16:9 horizontal in-text illustrations (PNG format)
• A shot list for each image: theme, core meaning, structure type, Xiaohei's action, suggested Chinese annotation
• Final images saved to the workspace directory assets/<article-slug>-illustrations/

Usage Examples

Planning Only (No Image Generation)

Use $ian-xiaohei-illustrations first don't generate images.
Please analyze the following article to identify where illustrations are needed, and output a shot list of about 5 images.
For each image, clearly state: which paragraph it follows, theme, core meaning, structure type, what Xiaohei is doing, and suggested Chinese annotation words.

<paste article>

Direct Generation of Full Illustrations

Use $ian-xiaohei-illustrations to generate 4 Xiaohei grotesque in-text illustrations for the following article.
Requirements: 16:9 horizontal, white background, black hand-drawn line art, minimal red-orange-blue Chinese hand-written annotations.

<paste article>

Generate a Single Image for One Concept

Use $ian-xiaohei-illustrations to generate an in-text illustration for the concept: "Trust is not shouted, but laid brick by brick with evidence."
The image should be grotesque but clean, and Xiaohei must take on the core action.

Edit an Existing Image

Use $ian-xiaohei-illustrations to help me edit this image: remove the title "Flowchart" in the top-left corner, keep everything else unchanged.

Notes

• The shorter the Chinese text in the image, the more stable the generation
• Each image should convey only one core structure; don't turn the article into a user manual
• Xiaohei must perform the core action; if the image still works without Xiaohei, then Xiaohei is too decorative
• Example images are only for calibrating line density, whitespace, color restraint, and Xiaohei's participation; do not copy their compositions
• AI image models may produce typos, hallucinated labels, style drift, or extra titles; review after generation
• If Chinese characters are seriously garbled, reduce the annotation words and regenerate