Smart Illustrator
Analyzes an article, script, or topic and auto-generates matching illustrations, slide infographics, and cover images via Gemini, Excalidraw, and Mermaid.
Image & VideoIntermediate★ 563⑂ 83AI score 9/10Last updated: Jun 26, 2026
What it does
- Reads a Markdown file, picks 3–5 spots that need visuals, and produces images plus an illustrated copy of the article.
- Routes each visual to the right engine: Gemini for metaphorical/creative art, Excalidraw for concept maps, comparisons, and simple flows (≤ 8 nodes) in a hand-drawn look, and Mermaid only for complex structures (> 8 nodes, layered architecture, multi-actor sequences), exported as high-res PNG.
- Three modes:
article(inline illustrations),slides(batch 16:9 infographics for decks), andcover(platform-sized covers — YouTube 16:9, WeChat 2.35:1, Twitter 1.91:1, Xiaohongshu 3:4). - Forces the model to load a style file (
styles/style-light.md,style-dark.md,style-cover.md,style-bento.md) instead of inventing its own system prompt, keeping output visually consistent. --prompt-onlyskips the API entirely: it emits a JSON prompt, copies it to the clipboard, and saves a backup so you can paste into Gemini Web.- Never edits the source file; it writes
{name}-image.md, a cover PNG, and numbered illustration PNGs.
Who it's for
- Technical bloggers and newsletter writers who keep skipping the visuals.
- Creators who need correctly sized thumbnails for YouTube, WeChat, Twitter, or Xiaohongshu.
- Course/content producers turning scripts into a full infographic slide set.
- Developers who want Mermaid diagrams with a consistent semantic color palette and layout rules.
Examples
/smart-illustrator posts/agent-guide.md— generates a cover plus 3–5 inline images and writesagent-guide-image.md./smart-illustrator script.md --mode slides --prompt-only— builds a batch JSON prompt for the whole deck and copies it to the clipboard for Gemini Web./smart-illustrator --mode cover --platform youtube --topic "Claude 4 deep dive"— produces just a 2K 16:9 YouTube thumbnail./smart-illustrator arch.md --engine mermaid— converts an architecture doc into a high-resolution Mermaid PNG with input/process/decision/output color coding.
· · · Install guide · · ·
Try it now, no install
Paste this into Claude to use the skill without installing anything.
Read the instructions in this file and follow them to help me: https://raw.githubusercontent.com/axtonliu/smart-illustrator/HEAD/SKILL.md What I want: (describe your task here)
If Claude can't open the link, open it yourself and paste the contents instead.
↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.
Install in the Claude app (no terminal)
- Download the ZIP with the button below.
- In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
- Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
Install in Claude Code
Let Claude do it — paste this into Claude Code
Install the skill I found on Claude Skill Mart. Copy the SKILL.m folder from the GitHub repo axtonliu/smart-illustrator into my ~/.claude/skills/smart-illustrator/. When it's done, tell me in one line what this skill can do.
Install with a command instead
git clone https://github.com/axtonliu/smart-illustrator.git ~/.claude/skills/smart-illustrator⚠ This is a third-party skill. Check the source repository before installing.
- Make sure Claude Code is installed and you can use a terminal.
- Create the skills folder:
mkdir -p ~/.claude/skills - Clone the repo into it:
git clone https://github.com/axtonliu/smart-illustrator.git ~/.claude/skills/smart-illustrator - Verify
SKILL.md,styles/,scripts/, andreferences/are present:ls ~/.claude/skills/smart-illustrator - Export your Gemini API key:
echo 'export GEMINI_API_KEY=your_key' >> ~/.zshrc && source ~/.zshrc - Install Node.js so
npxworks — the scripts run throughnpx -y bun ..., so no separate bun install is needed. - Restart Claude Code and try
/smart-illustrator path/to/article.md. - No API key yet? Add
--prompt-onlyto get the prompt and generate manually in Gemini Web (automatic clipboard copy works on macOS only).
View source on GitHub ↗License: MIT