This is a full pipeline for turning web articles, blog posts, or existing PowerPoint files into polished slide decks. It uses an orchestrator-workers pattern with a content.json file as the single source of truth, keeping extraction, translation, generation, and review separate. It handles COM Automation for editing open PowerPoint files directly, includes validation scripts for overflow and consistency checks, and comes with detailed implementation patterns for things like shape-based architecture diagrams and hyperlink auditing. The workflow is structured around phases with human-in-the-loop checkpoints, and it enforces rules like verifying technical content with MCP and keeping operational notes out of slides. If you need to automate presentation creation from source material or manage existing decks programmatically, this gives you the scaffolding and guardrails to do it consistently.
npx -y skills add aktsmm/agent-skills --skill powerpoint-automation --agent claude-codeInstalls into .claude/skills of the current project.
AI-powered PPTX generation using Orchestrator-Workers pattern.
From Web Article
Create a 15-slide presentation from: https://zenn.dev/example/article
From Existing PPTX
Translate this presentation to Japanese: input/presentation.pptx
Edit Open PPTX with COM
Edit the currently open PowerPoint deck with COM Automation and verify RefURL, notes, overflow, and hyperlinks.
TRIAGE → PLAN → PREPARE_TEMPLATE → EXTRACT → TRANSLATE → BUILD → REVIEW → DONE
| Phase | Main Actor | Purpose |
|---|---|---|
| EXTRACT | extract_images.py | Source -> content.json |
| BUILD | create_from_template.py | content.json -> PPTX |
| REVIEW | PPTX Reviewer | Overflow / consistency / quality |
Script selection, arguments, and validation entry points are defined in references/SCRIPTS.md.
content.json はこの skill の SSOT。抽出、翻訳、生成、レビューの間は常にこれを基準にする。
{
"slides": [
{ "type": "title", "title": "Title", "subtitle": "Sub" },
{ "type": "content", "title": "Topic", "items": ["Point 1"] }
]
}
スキーマ詳細は references/schemas/content.schema.json を参照。
標準テンプレートは assets/template.pptx。レイアウトや用途の詳細は template 側で管理し、main SKILL には最小限だけ残す。
python scripts/create_from_template.py assets/template.pptx content.json output.pptx --config assets/template_layouts.json
Role definitions and handoffs are in references/AGENTS.md and references/agents/.
SlideMaster.CustomLayouts に置き、編集する文字は placeholder / text shape として残すscripts/clean_template.py で docProps/custom.xml(MIP ラベル、tenant、SharePoint フィールド、同僚メール)と docProps/core.xml(cp:lastModifiedBy / cp:revision)を sanitize する。PowerPoint は OneDrive / SharePoint から開くたびに MIP を再付与するので、check-in 直前に毎回実行する。詳細は references/instructions/template.instructions.md。スライド本文検索だけで安全判定しないBIZ UDPゴシック に揃え、生成スライド側の後処理で毎回フォントを直す設計にしないUse Implementation Patterns for shape diagrams, template XML, SlideMaster edits, visual QA, hyperlinks, RefURL, locks, centering, corruption recovery, and media embedding.
sickn33/antigravity-awesome-skills
moizibnyousaf/ai-agent-skills
github/awesome-copilot