Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Basic-XYZ/baku-skills --skill baku-illustrationsgit clone --depth 1 https://github.com/Basic-XYZ/baku-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/basic-xyz/baku-skills/baku-illustrations)<a href="https://agentmods.dev/skills/basic-xyz/baku-skills/baku-illustrations"><img src="https://agentmods.dev/badge/skills/basic-xyz/baku-skills/baku-illustrations/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/basic-xyz/baku-skills/baku-illustrations"><img src="https://agentmods.dev/badge/skills/basic-xyz/baku-skills/baku-illustrations.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00148 | $0.01426 |
| Opus 5 | $0.00074 | $0.00713 |
| Sonnet 5 | $0.00030 | $0.00285 |
| Haiku 4.5 | $0.00015 | $0.00143 |
Grade A, and why
baku-illustrations scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Baku 中心配图
为中文文章、PPT、卡片、知识库、教程、复盘和产品说明生成“中心配图”。这类图是可嵌入正文或版式里的视觉资产,不是整页海报、PPT 页面或个人 IP 人像。
默认使用中心配图方法:画面主角是概念物件、流程关系、图表、机制模型或材质化结构;人物只是小型操作者、观察者、比例尺或动作提示,不承载固定个人 IP。
先读这些参考
按任务需要读取:
references/style-dna.md:整体视觉系统、三种渲染模式、颜色和禁忌。references/character-system.md:中性辅助角色的比例、动作和边界。references/composition-patterns.md:中心配图布局和适用场景。references/prompt-template.md:生成提示词模板、模式片段和修复提示。references/qa-checklist.md:生成后的检查、失败信号和迭代规则。
工作流
1. 读内容并决定是否需要图
先消化用户给的正文、HTML、Markdown、PPT、截图、数据表或简短想法。提炼:
- 核心概念、流程、对比、机制或图表是什么。
- 哪些段落值得用图解释,哪些只适合文字。
- 图片要放在文章、PPT、卡片还是社交图文里。
- 是否需要保留精确数据、轴、单位、类别顺序或数值。
不要平均配图。优先选择能成为认知锚点的内容:循环、流程、分流、前后变化、系统局部、层级、证据链、指标结构、科学机制或一个清楚的文本场景。
2. 自动选择渲染模式
除非用户明确指定,否则自行判断:
handdrawn:默认模式。纸面、铅笔、马克笔、手绘线条,适合正文配图、知识整理、复盘和轻量解释。handdrawn-3d:手绘线条加轻微体积、纸雕、黏土、木块、泡沫板或手工模型感,适合用户想保留手绘但希望更立体。material-3d:中心模型布局 + Baku Workbench 色系,黑色轮廓、暖纸/Field/Ink 材质、Baku Yellow 主强调,适合 PPT、卡片、汇报和更正式的解释图。
如果用户说“手绘”,优先 handdrawn。如果用户说“3D 但不要太玩具”,优先 handdrawn-3d。如果用户说“材质图”“汇报配图”“更正式”,优先 material-3d,但仍使用 Baku Design System 色系,不默认白色 studio。
3. 规划 shot list
如果用户只是让你分析怎么配图,先输出 shot list。每张图写清楚:
- 放在哪个段落或页面。
- 图的主题和核心意思。
- 构图模式。
- 渲染模式。
- 主体概念物件。
- 角色是否出现、出现在哪里、做什么。
- 3-5 个短中文标签。
默认 3-6 张;短文 1-3 张;长文也不要轻易超过 8 张。
4. 单张生成
用户明确要求“生成、输出、做图、给 demo”时,使用可用的图像生成工具逐张生成,不要把多张图拼在一张里。
每张图只讲一个核心结构。提示词必须包含:
- 画幅:常用
16:9,也可按用途选择wide horizontal 1.9:1或1:1。 - 图的用途:文章正文、PPT、卡片、报告或图文内嵌资产。
- 主体:概念物件、流程、图表、机制模型或文本场景,而不是人物。
- 模式:
handdrawn/handdrawn-3d/material-3d。 - 标签:3-5 个短简体中文标签,横排、高对比、靠近对象、远离边缘。
- 构图:主体完整可见、居中、安全边距充足、无裁切。
- 角色:可选小型辅助角色,通常 5%-12%,最多 15%,不做大头像和中心人物。
- 禁止项:无个人头像、无固定 IP、无 logo、水印、假 UI、标题、长段文字、装饰性 blobs、渐变背景。
5. 检查与迭代
生成后读取 references/qa-checklist.md 并检查。优先修这些问题:
- 人物太大、太像主角或抢走主题。
- 中文标签错字、太长、不可读或贴边。
- 主体裁切、边距不足。
- 画面变成整页 PPT、课程页、商业插画或假 UI。
- 图只好看但没有解释当前内容。
- 数据图没有保留准确数值、轴、单位和类别顺序。
6. 保存交付
如果在 workspace 内生成图片,默认保存到任务目录,而不是写进 skill 目录:
local-tests/<slug>/assets/
local-tests/<slug>/PROMPTS.md
What ships with it
7 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 95 lines · 148 tokens per session scan A 66aa7a601c32
baku-illustrations is a skill published in the GitHub repository Basic-XYZ/baku-skills (2 stars, last pushed 1mo ago), licensed MIT. It adds 148 tokens to every session and 1,426 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
design-system-synthesizer
Use when the user says 'build a design system for my site', 'extract design tokens', 'capture my brand', or 'build my style guide', or after a rebrand. Reads representative pages, theme files, and media to extract logo, colors, typography, spacing, and components, then writes a visible style-guide page.
art-direction
Use on any page build, redesign, or 'make it look better' request on a site that has (or needs) a saved Art Direction. Loads the direction before the first element is written, plans tokens in two passes against the direction's donts and the known AI-default aesthetics, builds with tokens preserved, checks the result…
page-template-library
Use when the user says 'save this as a template', 'create a playbook from this page', 'capture this layout', or has hand-built the same page type three times. Captures canonical page patterns as reusable playbooks that later page generation can spawn from in seconds.
figma-to-elementor
Use when the user says 'figma to elementor', 'build this figma design in elementor', or hands over a Figma frame to rebuild in WordPress. Maps the Figma node tree to Elementor widgets and containers and writes clean Elementor JSON into a draft duplicate, so nothing live is touched.
figma-to-bricks
Use when the user says 'figma to bricks', 'build this figma design in bricks', or hands over a Figma frame to rebuild on a Bricks Builder site. Maps the Figma node tree to native Bricks elements with BEM global classes, imports the frame's palette into the Bricks design system, and writes to a draft, so nothing live…
figma-to-divi
Use when the user says 'figma to divi', 'build this figma design in divi', or hands over a Figma frame to rebuild in WordPress on a Divi site. Maps the Figma node tree to native Divi sections, rows, columns and modules (Divi 5 blocks or Divi 4 shortcodes, auto-detected) and writes to a draft, so nothing live is…