This is a specialized illustration pipeline for a specific Chinese book about AI productivity. It analyzes manuscript chapters, identifies where diagrams would help, then uses subagent fan-out to generate image prompts in parallel before batch rendering them with a strict three-color palette. The workflow is opinionated: it enforces a 10-layout system (binary comparison, funnel, iceberg, etc.), includes prompt injection blocks to prevent AI clichés, and won't proceed without user confirmation at key gates. It's built for one book's visual system, not general purpose illustration work. If you're writing technical documentation with repeated visual patterns and need consistent batch illustration with human checkpoints, the architecture here is worth studying even if the content domain is narrow.

GitHub stars
375
last updated
Jun 3, 2026

Install

$ npx -y skills add jimliu/illustrated-agent-skills --skill book-illustrator --agent claude-code

Installs into .claude/skills of the current project.

Files

本书智能配图技能

一次性分析《图解 Skill —— AI 提效实战指南》的章节/附录书稿,批量生成所有配图提示词(subagent 并行),批量出图,自动回写。不在每张图之间停下询问。

使用方式

# 方式 1:指定本书章节/附录 Markdown 路径
/book-illustrator path/to/chapter/final.md

# 方式 2:指定本书书稿 URL(自动下载到本地)
/book-illustrator https://example.com/book/chapter-03.md

开工前必读

先读 references/design-system.md(本 SKILL.md 所在目录下),里面有:

  • 出版三色调色板(#D87757 + 黑 + 白)
  • 对比规则(防止暗底+暗字,最常见翻车点)
  • 反 AI 俗套禁用清单
  • 可读性自查清单(出图后必走)
  • Prompt Injection Block(subagent brief 里必须内联)

工作流程

步骤一:获取本书内容

  1. 本地路径:直接读取文件
  2. URL:下载本书章节/附录 Markdown 内容到临时位置

步骤二:分析本书配图需求

一次性逐段分析整篇章节/附录书稿,识别所有需要配图的位置。

配图的三种作用:

  1. 信息补充:帮助理解抽象概念(如"小步迭代"用滑板→自行车→汽车演进图)
  2. 概念具象化:将抽象观点转化为具体画面,让读者一眼就懂
  3. 引导想象:营造氛围、激发联想,增强阅读体验

适合配图的本书内容:

  • 抽象概念需要视觉化
  • 流程/步骤需要图示
  • 对比关系需要可视化
  • 核心观点需要强化
  • 场景描述需要引导想象

不需要配图的本书内容:

  • 已经很直观的描述(如代码示例、具体数字)
  • 简单的列表枚举
  • 引用的原话

配图数量(按章节/附录长度):

  • 每个主要小节至少考虑 1 张,优先选择核心观点和抽象概念
  • 开头/结尾可酌情增加 1 张(如需要氛围烘托)
  • 原则:宁多勿少,视觉内容能显著提升阅读体验

布局选择:根据内容的信息模式匹配布局,详见 references/layouts/ 目录下的 10 种布局定义:

Sponsors

Details

Repository
github.com/jimliu/illustrated-agent-skills
Category
AI & Agent Building
First seen
Jun 11, 2026

More from jimliu/illustrated-agent-skills

All 7
1microsoft avatarmicrosoft-foundrymicrosoft/azure-skillsDeploy, evaluate, fine-tune, and manage Foundry agents end-to-end with azd: hosted agent scaffold/run/deploy, prompt agent create, batch eval, continuous eval, 671.4Kskills.sh installs
2microsoft avatarazure-aimicrosoft/azure-skillsUse for Azure AI: Search, Speech, OpenAI, Document Intelligence. Helps with search, vector/hybrid search, speech-to-text, text-to-speech, transcription, OCR. WH660.5Kskills.sh installs
3microsoft avatarazure-aigatewaymicrosoft/azure-skillsConfigure Azure API Management as an AI Gateway for AI models, MCP tools, and agents. WHEN: semantic caching, token limit, content safety, load balancing, AI mo657.6Kskills.sh installs
4google avatargoogle-agents-cli-evalgoogle/agents-cliThis skill should be used when the user wants to "run an evaluation", "evaluate my ADK agent", "write an eval dataset", "analyze eval failures", "compare eval r558.9Kskills.sh installs
5google avatargoogle-agents-cli-observabilitygoogle/agents-cliThis skill should be used when the user wants to "set up tracing", "monitor my ADK agent", "configure logging", "add observability", "debug production traffic",558.7Kskills.sh installs
6juliusbrussee avatarcavecrewjuliusbrussee/cavemanDecision guide for delegating to caveman-style subagents. Tells the main thread WHEN to spawn `cavecrew-investigator` (locate code), `cavecrew-builder` (1-2 fil374.2Kskills.sh installs