菜单
End-to-end presentation workflow usable in Codex and Claude Code. Routes the user request to the HTML-mode or image-native-mode sub-skill, then runs the shared plan, design, and export stages. PDF and per-slide PNG are preferred; PPTX/Figma export is experimental / unstable.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/NomaDamas/slides-grab --skill slides-grab命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Stage 2 design skill usable in Codex and Claude Code. Generate and iterate slide-XX.html files in the selected slides workspace.
HTML-mode presentation pipeline usable in Codex and Claude Code. Build semantic slide-XX.html files, generate bespoke hero/accent imagery with slides-grab image, run the design gate, and export. Use when the deck needs editable, searchable, accessible text or when no existing template must be matched pixel-for-pixel.
Image-native presentation pipeline usable in Codex and Claude Code. Generate whole-slide raster images one slide at a time with slides-grab image, passing the reference template page as --reference so the model copies the layout and style and only swaps the content. Use when visual fidelity to an existing template matters more than editable text.
基于 SOC 职业分类
正在显示 SKILL.md
| name | slides-grab |
| description | End-to-end presentation workflow usable in Codex and Claude Code. Routes the user request to the HTML-mode or image-native-mode sub-skill, then runs the shared plan, design, and export stages. PDF and per-slide PNG are preferred; PPTX/Figma export is experimental / unstable. |
| metadata | {"short-description":"Routes a presentation request to the HTML or image-native pipeline"} |
Decides how the deck should be built and routes to the right pipeline. Read the user's intent, pick a delivery mode, then hand off to the matching sub-skill for the detailed stage instructions. Both pipelines share the same Stage 1 (slides-grab-plan) and Stage 3 (slides-grab-export) skills; the difference is how Stage 2 produces slides.
Ask the user (or infer from the request) which pipeline to use:
| Signal | Route to |
|---|---|
| "make slides about X", "write a deck", editable text, searchable content, accessibility, charts, diagrams, Card-news, most greenfield decks | slides-grab-html — semantic HTML slides |
| "make slides in this template/form", "match this existing deck", visual fidelity to a corporate template matters more than editable text, user handed a filled PPTX/PDF and wants the same look | slides-grab-image — image-native raster wrapper slides |
When in doubt, default to slides-grab-html. Only choose image-native when the user explicitly prioritizes visual matching to an existing template over editability.
After picking the mode, read the matching sub-skill in full and follow its pipeline. Do not mix instructions from the two sub-skills — the image generation steps are different and must not be combined.
Use the installed slides-grab-plan skill. The plan stage records the chosen mode: html or mode: image-native in slide-outline.md meta so Stage 2 knows which pipeline to run. Do not proceed to Stage 2 without approval of both style and outline.
Use the installed slides-grab-export skill. Export is gated on a Stage 2 design-gate Proceed verdict. PPTX and Figma are experimental / unstable.
slides-grab png --slides-dir <path> --slide-mode card-news --resolution 2160p (see slides-grab-card-news).slides-grab pdf --slides-dir <path> --output <name>.pdf.slides-grab png --slides-dir <path> --output-dir <path>/out-png --resolution 2160p.slides-grab convert --slides-dir <path> --output <name>.pptx --engine raster for visual fidelity (default), or --engine text for best-effort editable text.slides-grab figma --slides-dir <path> --output <name>-figma.pptx.Proceed verdict (zero unresolved Critical findings) on the latest rendered state. slides-grab pdf, slides-grab convert, and slides-grab figma block if the receipt is missing, stale, or fails the CLI-enforced report contract.decks/<deck-name>/ as the slides workspace for multi-deck projects.tldraw-generated asset unless the user explicitly asks for a different approach.slides-grab image (reuses local Codex ChatGPT login — no API key required; run codex login once) and keep the saved asset local under <slides-dir>/assets/. Image-native decks generate each whole-slide raster with slides-grab image --image-native --name slide-XX so the wrapper and editor metadata are created together — see the image sub-skill.slides-grab validate plus slides-grab build-viewer review so blank canvases are caught as empty-canvas before export. Local videos live under <slides-dir>/assets/ with poster="./assets/<file>" thumbnails; use slides-grab fetch-video or yt-dlp to pull supported web videos first.slides-grab list-styles and slides-grab preview-styles for bundled style discovery during planning.references/presentation-workflow-reference.md — archived end-to-end workflow guidance from the legacy skill set../slides-grab-html/SKILL.md — HTML-mode pipeline (semantic slides)../slides-grab-image/SKILL.md — image-native pipeline (raster wrapper slides)