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 huangrichao2020/pretty-skills --skill ai-image-to-pptx-ai-pptgit clone --depth 1 https://github.com/huangrichao2020/pretty-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/huangrichao2020/pretty-skills/ai-image-to-pptx-ai-ppt)<a href="https://agentmods.dev/skills/huangrichao2020/pretty-skills/ai-image-to-pptx-ai-ppt"><img src="https://agentmods.dev/badge/skills/huangrichao2020/pretty-skills/ai-image-to-pptx-ai-ppt/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/huangrichao2020/pretty-skills/ai-image-to-pptx-ai-ppt"><img src="https://agentmods.dev/badge/skills/huangrichao2020/pretty-skills/ai-image-to-pptx-ai-ppt.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00093 | $0.03264 |
| Opus 5 | $0.00046 | $0.01632 |
| Sonnet 5 | $0.00019 | $0.00653 |
| Haiku 4.5 | $0.00009 | $0.00326 |
Grade A, and why
ai-image-to-pptx 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 — 292 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI 生图 → 真实 PPTX Skill
⚠️ 出图前必走 · ppt 出图法 V2 7 步
重要：本 skill 只负责"把已有图嵌入 .pptx"。生图前必须走ppt 出图法 V2 7 步：
1. fetch 完整仓库内容（README + 子目录 + examples 并行） ⚠️ 不能"读 80 行就开干"
2. reset todo（明确节奏） ⚠️ 不能"流式想到啥做啥"
3. 整理 "N 张图主题+核心一句话" 清单（每张 1 行） ⚠️ 不能把所有画面细节列死
4. 单张 prompt ≤ 60 行（不堆字体铁律） ⚠️ 不能堆 100+ 行
5. 出 P1 hook → 渲染核对清单（10+ 项）→ 走批量 ⚠️ 不能凭感觉看图就过
6. 接受 AI 主动加的细节和布局调整 ⚠️ 不能否定 AI 偏差
7. 沉淀到 case 文档 ⚠️ 不能不做 case 沉淀
ppt 出图法 V2 完整方法论见 ~/.mavis/knowledge/knowhub/domains/visual-creation/methodology/ppt-out-image-method.md。
反例教训（2026-07-02 how-to-agent-ppt）：没走 V2 7 步 → PPT 图字体幼稚、布局不丰富 → 用户反馈"前天的好多了"。
正例（2026-06-30 ai-training-ppt）：走了 V2 → 8 张图 + 真实 .pptx（20MB）一气呵成。
跳过 V2 7 步直接调本 skill = 极大概率翻车。
一句话定位
"AI 生图很强，但用户要能打开的 PPT 文件"
↓
AI 直生图（matrix MCP）→ python-pptx 嵌入图片 → .pptx 文件
↓
ai-training-methodology-handdrawn.pptx · 任何"知识图集 PPT"
完整流程（5 步）
1. 确定章节结构（问"做几张图？每张讲什么？"）
↓
2. 选风格 + 写 5 段式 prompt 模板（统一风格段）
↓
3. 批量调 matrix_generate_image 生成图（2 张/批防 timeout）
↓
4. 复制图到 PPT 项目目录，命名 01-overview.png / 02-xxx.png ...
↓
5. python-pptx 脚本嵌入图片 → 输出 .pptx
↓
6. （可选）部署或导出 PDF
Step 1：5 段式 Prompt 模板
[1 场景] 是什么类型的图（信息图/海报/分镜）
[2 主体] 核心内容是什么（一句话）
[3 结构] 画面分几块、每块放什么（编号 1./2./3.）
[4 风格] 风格关键词 + 配色 + 字体 + 构图 ← 关键，跨图必须共享
[5 约束] 无错字、中文/英文、分辨率、不许出现XX
核心原则：结构 > 文采。把"画面分几块、每块放什么"编号说清楚，比写漂亮话有用 10 倍。
共享 [4 STYLE] 段（3 套速选）
手绘科教风（传播首选）：
Chinese hand-drawn doodle educational infographic.
Soft cream-pink paper background (#FFF5F5).
All outlines are sketchy hand-drawn black ink lines with slight wobble,
like drawn with a marker.
Hand-written style mixed Chinese text, key words in coral pink (#FF6B9D)
with hand-drawn wavy underlines.
Cards have soft pastel borders alternating:
mint green #B5EAD7, baby blue #A2D2FF, peach pink #FFC8DD,
butter yellow #FFF1A8, lavender #CDB4DB.
Macaron pastel color scheme, NO realistic textures, NO photo,
hand-drawn cute sticker style, kawaii aesthetic,
slightly imperfect charming lines, watercolor wash fills,
educational comic infographic, very clean and readable,
soft and friendly.
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 · 292 lines · 93 tokens per session scan A db332b407d04
ai-image-to-pptx is a skill published in the GitHub repository huangrichao2020/pretty-skills (54 stars, last pushed yesterday), licensed MIT. It adds 93 tokens to every session and 3,264 once invoked, about $0.0005 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 2026-08-30.
Other skills, from other repositories
qmd
Skill "qmd" from rodaddy/mcp2cli, covering qmd, quick reference, usage and notes.
cn-check
Install and run the Continue CLI (cn) to execute AI agent checks on local code changes. Use when asked to "run checks", "lint with AI", "review my changes with cn", or set up Continue CI locally.
morph-ppt
Use this skill when the user wants a .pptx with smooth cross-slide animation — PowerPoint Morph transitions, Keynote-style continuous motion, shapes that grow / move / rotate as the slide advances. Trigger on: 'morph', 'morph transition', 'smooth transition', 'continuous animation across slides', 'Keynote-style…
officecli-word-form
Use this skill to create fillable Word forms (.docx) with real Content Controls (SDT) + legacy FormField checkboxes + MERGEFIELD mail-merge placeholders + document protection. Trigger on: 'fillable form', 'form fields', 'content controls', 'SDT', 'word form', 'fill in', 'only editable fields', 'protect document'…
officecli-academic-paper
Use this skill to build academic-style .docx output: journal / conference / thesis chapters carrying formal citation style (APA, Chicago, IEEE, MLA), numbered equations, figure & table cross-references, footnotes/endnotes, bibliography, or multi-column journal layout. Trigger on: 'research paper', 'journal paper'…
officecli-data-dashboard
Use this skill to build a multi-element Excel dashboard — Dashboard sheet on open, multiple formula-driven KPI cards, multiple charts, sparklines, and conditional formatting — from CSV or tabular input. Trigger on: 'dashboard', 'KPI dashboard', 'analytics dashboard', 'executive dashboard', 'metrics dashboard', 'CSV to…