ppt-workflow-builder
PPT Workflow Builder
Quick start
Goal: build slide decks through a repeatable workflow instead of jumping straight into page-by-page slide writing.
Reference notes:
skills/ppt-workflow-builder/references/ppt-workflow.mdskills/ppt-workflow-builder/references/ppt-workflow_zh.md
Best fit
- Conference decks, client presentations, internal reports, pitch decks, teaching slides, and proposal decks.
- Cases where the source material is messy and needs to be turned into a clear slide narrative.
- Tasks that benefit from separate stages for structure, research, planning, and final slide visuals.
- Cases where an SVG-first workflow is acceptable before final PPT assembly.
Core workflow
- Frame the deck before writing slides:
- Identify audience, presentation goal, tone, and expected deck length.
- Define the single through-line the deck should prove.
- Build a structured outline first:
- Create a JSON outline that lists sections, slide goals, and key claims.
- Do not start page visuals before the outline is stable.
- Research slide content by section:
- Use
context7for official technical docs when the deck is about libraries, APIs, or standards. - Use
ref-skillfor GitHub examples, issues, implementation references, and real-world code. - Use
grok-searchfor broader research, comparisons, framing, best practices, supporting examples, and non-technical material. State the Grok model when it is known.
- Use
- Turn the outline into a planning draft:
- For each slide, define the core message, supporting evidence, visual type, and speaker intent.
- Kill slides that do not move the argument forward.
- Generate final slide pages in an SVG-first format:
- Produce each slide as a separate SVG page with a consistent 16:9 canvas.
- Keep the layout presentation-grade, editable, and suitable for later PPT import.
- Assemble the PPT:
- Import the SVG pages into presentation software when that workflow is acceptable.
- If native PPT editing is required later, use the SVG pages as the visual source of truth while rebuilding or refining slides inside the deck tool.
Working files
Use a predictable artifact chain:
deck-outline.jsonfor the approved slide structure.deck-research.mdfor gathered facts, references, and supporting notes.deck-plan.mdfor per-slide messages, layouts, and evidence.slides/slide-01.svg,slides/slide-02.svg, ... for final page assets.
Do not skip directly from topic to final slides. The outline and planning files are part of the deliverable workflow.
Prompt pattern
Use a staged prompt pattern rather than one-shot prompting:
- Ask for or infer audience, goal, tone, and length.
- Generate JSON outline.
- Research missing facts and supporting material.
- Produce slide planning draft.
- Generate final SVG pages.
The reference files include reusable prompt scaffolds for each stage.
Guardrails
- Do not start with slide decoration before the deck argument is clear.
- Do not generate final slide pages before the outline is approved.
- Do not let every slide become a text wall; every page needs a single job.
- Do not treat research notes as slide copy. Rewrite for presentation flow.
- Do not use a single one-shot mega prompt when a staged workflow is more reliable.
- If a factual gap blocks a slide, go back to research instead of improvising.
More from ry-run/run-skills
jetbrains-skill
Use the JetBrains IDE MCP Server (IntelliJ IDEA 2025.2+) for IDE-backed indexing, inspections, symbol info, project file operations, and structural refactoring. Use it when you need exact project-aware analysis or safe IDE refactors, rather than broad semantic repository discovery.
78desktop-commander
Use Desktop Commander MCP (typically tools like `mcp__desktop-commander__*`) for local files, precise edits, search, and long-running processes on the machine. Use it when the task requires direct local execution, rather than IDE-only actions or remote platform operations.
62desktop-commander-mcp
使用 Desktop Commander MCP(常见为 `mcp__desktop-commander__*` 工具)进行本地文件/目录读写与搜索、精确文本替换、Excel/PDF 处理、启动与交互长生命周期进程(Python/Node/SSH/DB)、查看与终止进程/会话、读取工具调用历史与用量统计、管理服务器配置。适用于:需要“在机器上执行操作”的任务(读写代码/配置、批量改文件、搜索代码、分析 CSV/Excel、生成/修改 PDF、跑命令并持续读取输出、交互式 REPL 调试/数据分析)时。
2deepwiki
Use DeepWiki MCP for repository-focused documentation from deepwiki.com. Use it when you need a fast overview of an external repository, architecture, or repo-specific concepts, rather than live library docs or broad web research.
1ace-tool
Use Ace Tool MCP (`mcp__ace_tool__search_context`) for semantic repository discovery. Use it when the exact file, symbol, or implementation path is unknown, rather than starting with exact IDE search or direct file edits.
1context7
Use Context7 MCP for up-to-date library and framework documentation tied to an exact library ID. Use it for versioned API usage, setup, migrations, and code examples, rather than generic documentation-site search.
1