ppt-creator

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

PPT Creator

PPT Creator

Goal: Transform a simple topic into a presentation-ready, high-quality slide deck. When key information is missing, use the minimal intake form (references/INTAKE.md) to gather context or apply safe defaults. Then follow the workflow (references/WORKFLOW.md) to produce an outline, slide drafts, charts, and speaker notes. After generation, self-evaluate using the rubric (references/RUBRIC.md); if the score is < 75, automatically refine up to 2 iterations until ≥ 75 before delivery. See Deliverables section for final output structure.
目标:将简单主题转换为可直接用于演示的高质量幻灯片。当关键信息缺失时,使用最小化收集表单(references/INTAKE.md)获取上下文或应用安全默认值。然后遵循工作流(references/WORKFLOW.md)生成大纲、幻灯片草稿、图表和演讲备注。生成完成后,使用评分标准(references/RUBRIC.md)进行自我评估;如果得分<75分,自动优化最多2次迭代,直到得分≥75分再交付。最终输出结构请查看交付物部分。

When to Use This Skill

何时使用此技能

Use this skill when the user requests:
  • "Make a presentation/deck/PPT/slides" on any topic
  • "Improve/optimize a presentation/pitch/demo"
  • Converting scattered materials into a structured, persuasive slide deck
  • Creating presentations with data visualization and speaker notes
  • Building decks for business reviews, product pitches, educational content, or reports
当用户提出以下需求时使用本技能:
  • 为任意主题“制作演示文稿/幻灯片/PPT/演示页”
  • “改进/优化演示文稿/推介内容/演示Demo”
  • 将零散资料转换为结构化、有说服力的幻灯片
  • 创建包含数据可视化和演讲备注的演示文稿
  • 为业务回顾、产品推介、教育内容或报告制作幻灯片

Quick Start

快速开始

  1. Gather Intent: If critical information is missing, ask the 10 Minimal Questions (references/INTAKE.md). If the user doesn't respond after 2 prompts, use the safe default for each item and clearly note assumptions in speaker notes.
  2. Structure the Story: Apply the Pyramid Principle to establish "one conclusion → 3-5 top-level reasons → supporting evidence." Each slide uses assertion-style headings (complete sentences), with body content providing evidence (charts/tables/diagrams/data points). Templates are in references/TEMPLATES.md.
  3. Choose Charts: Use the Chart Selection Dictionary in references/VIS-GUIDE.md to pick the most appropriate visualization for each point. If the user provides data (tables/CSV), optionally call
    scripts/chartkit.py
    to generate PNG charts; otherwise, create placeholder diagrams with a list of required data fields.
  4. Layout & Style: Follow references/STYLE-GUIDE.md for font sizes, line spacing, white space, contrast ratios, color palettes, and accessibility (WCAG AA compliance).
  5. Speaker Notes: Generate 45-60 second speaker notes for each slide, structured as: opening → core assertion → evidence explanation → transition.
  6. Self-Check & Score: Use references/CHECKLIST.md for a pre-flight check, then score with references/RUBRIC.md. If total score < 75, identify the weakest 3 items and refine; repeat scoring (max 2 iterations).
  7. Deliverables (all saved to
    /output/
    ):
    • /output/slides.md
      : Markdown slides (Marp/Reveal.js compatible), with assertion-style headings + bullet points/chart placeholders + notes
    • /output/assets/*.png
      : Generated charts (if applicable)
    • /output/notes.md
      : Full speaker notes and delivery outline
    • /output/refs.md
      : Citations and data sources
    • /output/presentation.pptx
      : If
      python-pptx
      is available, export to PPTX; otherwise, keep Markdown and include instructions for "one-click conversion to PPTX" in the first screen (does not block delivery)
  1. 收集需求意图:如果关键信息缺失,询问10个核心问题(references/INTAKE.md)。如果用户在2次提示后未回复,为每项内容使用安全默认值,并在演讲备注中明确标注假设条件。
  2. 构建叙事结构:应用Pyramid Principle建立“一个结论 → 3-5个顶层理由 → 支撑证据”的结构。每张幻灯片使用assertion式标题(完整句子),正文内容提供证据(图表/表格/示意图/数据点)。模板请参考references/TEMPLATES.md。
  3. 选择图表类型:使用references/VIS-GUIDE.md中的图表选择字典为每个要点挑选最合适的可视化形式。如果用户提供了数据(表格/CSV),可选择性调用
    scripts/chartkit.py
    生成PNG图表;否则创建占位示意图,并列出所需数据字段。
  4. 布局与样式:遵循references/STYLE-GUIDE.md中的字体大小、行间距、留白、对比度、调色板和无障碍设计(WCAG AA合规)规范。
  5. 演讲备注:为每张幻灯片生成45-60秒的演讲备注,结构为:开场 → 核心主张 → 证据解释 → 过渡语。
  6. 自我检查与评分:使用references/CHECKLIST.md进行预检查,再通过references/RUBRIC.md评分。如果总分<75分,找出得分最低的3项进行优化;重复评分(最多2次迭代)。
  7. 交付物(全部保存至
    /output/
    目录):
    • /output/slides.md
      :Markdown格式幻灯片(兼容Marp/Reveal.js),包含assertion式标题 + 项目符号/图表占位符 + 备注
    • /output/assets/*.png
      :生成的图表(如有)
    • /output/notes.md
      :完整演讲备注和演示大纲
    • /output/refs.md
      :引用来源和数据出处
    • /output/presentation.pptx
      :若
      python-pptx
      可用,导出为PPTX格式;否则保留Markdown格式,并在首页添加“一键转换为PPTX”的操作说明(不影响交付)

Orchestration Mode (End-to-End Automation)

编排模式(端到端自动化)

When the user requests a "complete" or "presentation-ready" deliverable, ppt-creator automatically orchestrates the full pipeline: content creation → data synthesis → chart generation → dual-path PPTX creation (Marp + document-skills:pptx) → chart insertion. This typically delivers TWO complete PPTX files with different styling for user comparison.
Activation: Phrases like "complete PPTX", "final deliverable", "ready for presentation" Duration: 4-6 minutes (parallel execution) Output: presentation_marp_with_charts.pptx + presentation_pptx_with_charts.pptx
For orchestration details, see
references/ORCHESTRATION_OVERVIEW.md
(start here), then navigate to specialized guides as needed.
当用户要求“完整”或“可直接演示”的交付物时,ppt-creator会自动编排全流程:内容创作 → 数据合成 → 图表生成 → 双路径PPTX创建(Marp + document-skills:pptx)→ 图表插入。通常会交付两份风格不同的完整PPTX文件供用户选择。
触发关键词:“完整PPTX”、“最终交付物”、“可直接演示” 耗时:4-6分钟(并行执行) 输出:presentation_marp_with_charts.pptx + presentation_pptx_with_charts.pptx
编排细节请参考
references/ORCHESTRATION_OVERVIEW.md
(入门首选),再根据需要查看专项指南。

Core Principles (Must Follow)

核心原则(必须遵守)

  • Information Organization: Conclusion first, then evidence (Pyramid Principle). Each slide conveys only 1 core idea. Headings must be testable assertion sentences, not topic labels.
  • Evidence-First: Use charts/tables/evidence blocks instead of long paragraphs; limit to 3-5 bullet points per slide.
  • Data Visualization: Chart selection and labeling (axes/units/sources) must comply with references/VIS-GUIDE.md. If data is insufficient, provide "placeholder chart + list of missing fields."
  • Accessibility: Color and text contrast must meet AA standards (see STYLE-GUIDE). Provide alt/readable descriptions for charts and images.
  • Reusability: Use consistent naming, stable paths, reproducible output. Do not hard-code random numbers in code.
  • Safety & Dependencies: Do not scrape the web without permission. Only run scripts when user provides data. If
    matplotlib/pandas
    are unavailable, fall back to text + placeholder diagram instructions.
  • 信息组织:结论先行,后附证据(Pyramid Principle)。每张幻灯片仅传递1个核心观点。标题必须是可验证的assertion句子,而非主题标签。
  • 证据优先:使用图表/表格/证据块替代长段落;每张幻灯片最多3-5个项目符号。
  • 数据可视化:图表选择和标注(坐标轴/单位/来源)必须符合references/VIS-GUIDE.md规范。若数据不足,提供“占位图表 + 缺失字段列表”。
  • 无障碍设计:颜色和文本对比度必须符合AA标准(见STYLE-GUIDE)。为图表和图片提供简短的替代描述。
  • 可复用性:使用统一命名、稳定路径、可复现的输出结果。不得在代码中硬编码随机数值。
  • 安全与依赖:未经许可不得爬取网页。仅在用户提供数据时运行脚本。若
    matplotlib/pandas
    不可用, fallback为文本 + 占位示意图说明。

Workflow Overview

工作流概述

Stage 0 - Archive Input: Record user's original request, defaults used, and assumptions made.
Stage 1 - Structure Goals: Rewrite the goal into "who takes what action when" (clear CTA).
Stage 2 - Storyline: Use Pyramid Principle to define "one-sentence conclusion → 3-5 first-level reasons → evidence."
Stage 3 - Outline & Slide Titles: Create a 12-15 slide chapter skeleton. Each slide has one assertion-style heading.
Stage 4 - Evidence & Charts: Use the Chart Selection Dictionary from VIS-GUIDE. If data is provided, call chartkit.py to generate PNGs; otherwise, create placeholder + required field list.
Stage 5 - Layout & Accessibility: Apply STYLE-GUIDE for font sizes, spacing, contrast ratios, color palettes; unify units and decimal places.
Stage 6 - Speaker Notes: Generate 45-60 second notes per slide: opening → assertion → evidence explanation → transition.
Stage 7 - Self-Check & Scoring: Run CHECKLIST; score with RUBRIC. If score < 75, focus on weakest 3 items, refine, re-score (max 2 iterations).
Stage 8 - Package Deliverables: Generate
/output/
directory with slides.md / notes.md / refs.md / assets/*.png. If
python-pptx
is available, export PPTX.
Stage 9 - Reuse Instructions: Append a "5-step guide to replace data/colors with your own" at the end of notes.md.
阶段0 - 输入归档:记录用户原始请求、使用的默认值和做出的假设。
阶段1 - 目标结构化:将目标重写为“谁在何时采取什么行动”(明确的行动号召CTA)。
阶段2 - 叙事线:使用Pyramid Principle定义“一句话结论 → 3-5个一级理由 → 支撑证据”。
阶段3 - 大纲与幻灯片标题:创建12-15页的幻灯片框架。每张幻灯片配备一个assertion式标题。
阶段4 - 证据与图表:使用VIS-GUIDE中的图表选择字典。若提供数据,调用chartkit.py生成PNG图;否则创建占位符 + 所需字段列表。
阶段5 - 布局与无障碍设计:应用STYLE-GUIDE中的字体大小、间距、对比度、调色板规范;统一单位和小数位数。
阶段6 - 演讲备注:为每张幻灯片生成45-60秒的备注:开场 → 主张 → 证据解释 → 过渡语。
阶段7 - 自我检查与评分:执行CHECKLIST;通过RUBRIC评分。若得分<75分,聚焦得分最低的3项进行优化,重新评分(最多2次迭代)。
阶段8 - 交付物打包:生成
/output/
目录,包含slides.md / notes.md / refs.md / assets/*.png。若
python-pptx
可用,导出PPTX格式。
阶段9 - 复用指南:在notes.md末尾添加“5步替换数据/颜色”的操作指南。

Resources

资源

references/INTAKE.md

references/INTAKE.md

Minimal 10-Item Questionnaire (use defaults if missing):
  1. Who is the audience? (Default: general public)
  2. Core objective? (Default: "understand and accept" a proposition)
  3. Desired action/decision? (Default: agree to move to next step after the meeting)
  4. Duration & slide count limit? (Default: 15-20 min, 12-15 slides)
  5. Tone & style? (Default: professional, clear, friendly)
  6. Topic scope & boundaries? (Default: given topic + 1 layer related)
  7. Must-include points/taboos? (Default: none)
  8. Available data/tables? (Default: none; can generate structure placeholder + list required fields)
  9. Brand & visual constraints? (Default: built-in neutral theme)
  10. Deliverable format preference? (Default: slides.md + optional PNG charts; export PPTX if available)
最小化10项问卷(缺失时使用默认值):
  1. 受众是谁?(默认:普通大众)
  2. 核心目标?(默认:让受众理解并接受某个主张)
  3. 期望的行动/决策?(默认:会议后同意进入下一阶段)
  4. 时长与幻灯片数量限制?(默认:15-20分钟,12-15页)
  5. 语气与风格?(默认:专业、清晰、友好)
  6. 主题范围与边界?(默认:给定主题 + 1层相关延伸)
  7. 必须包含/禁忌内容?(默认:无)
  8. 可用数据/表格?(默认:无;可生成结构占位符 + 所需字段列表)
  9. 品牌与视觉约束?(默认:内置中性主题)
  10. 交付物格式偏好?(默认:slides.md + 可选PNG图表;若可用则导出PPTX)

references/WORKFLOW.md

references/WORKFLOW.md

Detailed step-by-step process from "topic" to "presentation-ready output."
从“主题”到“可演示输出”的详细分步流程。

references/TEMPLATES.md

references/TEMPLATES.md

Slide Template Library (assertion-evidence style):
  • Cover, Table of Contents, Problem Statement, Opportunity/Goal, Solution Overview, Evidence 1-3, Risk & Mitigation, Case Study/Comparison, Roadmap/Timeline, Conclusion & Actions, Backup Slides
  • Micro-templates: Comparison (A vs B), Pyramid Summary, Process 4-Step, KPI Dashboard, Geographic Distribution, Funnel, Pareto, Sensitivity, Cost Structure (Waterfall), Contribution (Stacked)
幻灯片模板库(assertion-evidence风格):
  • 封面、目录、问题陈述、机会/目标、解决方案概述、证据1-3、风险与缓解、案例研究/对比、路线图/时间线、结论与行动、备用幻灯片
  • 微模板:对比(A vs B)、金字塔总结、四步流程、KPI仪表盘、地理分布、漏斗图、帕累托图、敏感性分析、成本结构(瀑布图)、贡献占比(堆叠图)

references/VIS-GUIDE.md

references/VIS-GUIDE.md

Data Visualization Selection & Labeling Standards:
  • Chart Selection Dictionary (common questions → chart types)
  • Labeling & units (axes, units, data scope, time range; source in footer)
  • Accessibility & contrast (WCAG 2.1 AA: text vs background ≥ 4.5:1; UI elements ≥ 3:1)
  • Assertion-Evidence writing tips
数据可视化选择与标注标准
  • 图表选择字典(常见问题 → 对应图表类型)
  • 标注与单位(坐标轴、单位、数据范围、时间范围;来源标注在页脚)
  • 无障碍与对比度(WCAG 2.1 AA:文本与背景≥4.5:1;UI元素≥3:1)
  • assertion-evidence写作技巧

references/STYLE-GUIDE.md

references/STYLE-GUIDE.md

Layout & Style (neutral theme, supports brand replacement):
  • Canvas: 16:9; safe margins ≥ 48px; grid column spacing 24px
  • Fonts: Chinese (Source Han Sans/PingFang/Hiragino Sans), English (Inter/Calibri)
  • Font sizes: Heading 34-40, Subheading 24-28, Body 18-22, Footer 14-16
  • Line spacing: Heading 1.1, Body 1.3; bullet spacing ≥ 8px
  • Color palette (AA compliant): Dark ink #1F2937 / Background #FFFFFF / Accent #2563EB / Emphasis #DC2626
  • Components: unified 6-8px border radius; charts and images with 8px padding
  • Images: add brief alt descriptions for screen readers
  • Page density: ≤ 70 words per slide (excluding captions)
布局与样式(中性主题,支持品牌替换):
  • 画布:16:9;安全边距≥48px;网格列间距24px
  • 字体:中文(思源黑体/苹方/冬青黑体)、英文(Inter/Calibri)
  • 字体大小:标题34-40号、副标题24-28号、正文18-22号、页脚14-16号
  • 行间距:标题1.1倍、正文1.3倍;项目符号间距≥8px
  • 调色板(AA合规):深墨色#1F2937 / 背景色#FFFFFF / 强调色#2563EB / 警示色#DC2626
  • 组件:统一6-8px圆角;图表和图片带8px内边距
  • 图片:为屏幕阅读器添加简短的替代描述
  • 页面密度:每张幻灯片≤70词(不含说明文字)

references/RUBRIC.md

references/RUBRIC.md

PPT Quality Scoring Rubric (100 points; ≥ 75 to deliver): Each item scored 0-10:
  1. Goal Clarity: Audience, objective, CTA well-defined
  2. Story Structure: Pyramid structure complete, hierarchy clear
  3. Slide Assertions: Headings are "assertion sentences" supported by evidence
  4. Evidence Quality: Data/cases/citations sufficient, credible, consistent calibration
  5. Chart Fit: Correct selection, complete labeling, readable
  6. Visual & Accessibility: Contrast, font size, white space, color compliance
  7. Coherence & Transitions: Natural chapter and page transitions
  8. Speakability: 45-60 sec per slide, natural language
  9. Deliverables Complete: slides.md / notes.md / refs.md / (optional) assets/*.png
  10. Robustness: Gaps explicitly marked, fallback plan & next steps provided
Self-evaluation process: Run CHECKLIST first, then score each item and write top 3 low-scoring items + improvement actions. If total < 75, apply actions and re-score (max 2 iterations).
PPT质量评分标准(满分100分;≥75分方可交付): 每项0-10分:
  1. 目标清晰度:受众、目标、CTA定义明确
  2. 叙事结构:Pyramid结构完整,层级清晰
  3. 幻灯片主张:标题为“assertion句子”且有证据支撑
  4. 证据质量:数据/案例/引用充分、可信、校准一致
  5. 图表适配性:选择正确、标注完整、可读性强
  6. 视觉与无障碍:对比度、字体大小、留白、颜色符合规范
  7. 连贯性与过渡:章节和页面过渡自然
  8. 可演讲性:每张幻灯片对应45-60秒内容,语言自然
  9. 交付物完整性:包含slides.md / notes.md / refs.md /(可选)assets/*.png
  10. 鲁棒性:明确标记缺口,提供 fallback方案与下一步行动
自我评估流程:先执行CHECKLIST,再为每项评分并列出得分最低的3项+改进措施。若总分<75分,执行改进措施后重新评分(最多2次迭代)。

references/CHECKLIST.md

references/CHECKLIST.md

Pre-flight checklist for final quality assurance before delivery.
交付前的最终质量保证预检查清单。

references/EXAMPLES.md

references/EXAMPLES.md

Two Usage Examples:
  • Example A: Ultra-simple topic ("coffee") → trigger minimal questionnaire, generate 12-page framework with placeholder charts
  • Example B: Small business monthly review with attached CSV → parse data, select charts per VIS-GUIDE, call chartkit.py, refine 1-2 iterations if score < 75
两个使用示例
  • 示例A:超简单主题(“咖啡”)→ 触发最小化问卷,生成12页带占位图表的框架
  • 示例B:附带CSV的小企业月度回顾→ 解析数据,按VIS-GUIDE选择图表,调用chartkit.py,若得分<75分则优化1-2次迭代

scripts/chartkit.py

scripts/chartkit.py

Minimal chart renderer for ppt-creator.
Usage:
bash
python scripts/chartkit.py \
  --data path/to/data.csv \
  --type line \
  --x date \
  --y sales profit \
  --out output/assets \
  --filename kpi_trend.png \
  --title "Monthly KPIs"
Notes:
  • Requires:
    pandas
    ,
    matplotlib
  • Fallback: If packages unavailable, print instruction message and exit(0)
  • Uses matplotlib defaults for readability (no hard-coded brand colors)
为ppt-creator打造的轻量图表渲染器
使用方式
bash
python scripts/chartkit.py \
  --data path/to/data.csv \
  --type line \
  --x date \
  --y sales profit \
  --out output/assets \
  --filename kpi_trend.png \
  --title "Monthly KPIs"
注意事项
  • 依赖:
    pandas
    ,
    matplotlib
  • Fallback方案:若依赖包不可用,打印操作说明后退出(0)
  • 使用matplotlib默认样式保证可读性(无硬编码品牌颜色)

Advanced Tips

进阶技巧

  • This skill complements (not conflicts with) Anthropic's built-in PowerPoint generation capabilities. Use this skill to produce "high-quality structure & content," then optionally invoke system capabilities to export the final PPTX file.
  • For complex data analysis needs, combine with other skills (e.g., data analysis, charting) before invoking ppt-creator.
  • The skill is designed to be forgiving: missing information triggers safe defaults rather than blocking progress.
  • 本技能补充而非替代Anthropic内置的PowerPoint生成能力。使用本技能生成“高质量结构与内容”,再可选调用系统能力导出最终PPTX文件。
  • 如需复杂数据分析,可先结合其他技能(如数据分析、图表制作),再调用ppt-creator。
  • 本技能设计容错性强:缺失信息会触发安全默认值,而非阻塞流程。