pptx_skill

This skill helps you create, edit, analyze, and automate PowerPoint presentations (.pptx) using Python with design-aware, template-based workflows.
  • Python

310

GitHub Stars

3

Bundled Files

2 months ago

Catalog Refreshed

3 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill tfriedel/claude-office-skills --skill pptx

  • html2pptx.md19.4 KB
  • ooxml.md10.1 KB
  • SKILL.md24.9 KB

Overview

This skill provides presentation creation, editing, and analysis for .pptx files, enabling automated workflows for new slides, template-based builds, and low-level OOXML editing. It supports full-content extraction, visual design decisions, HTML-to-PPTX conversion, thumbnail validation, and safe XML edits for comments, notes, and layouts. Use it to produce production-ready slides, modify existing decks, or inspect design tokens like fonts and colors.

How this skill works

For high-level creation, the skill converts HTML slide files into PowerPoint using an html2pptx pipeline and PptxGenJS for charts and placeholders. For deep edits and analysis, it unpacks the .pptx (OOXML) archive to expose XML files (slides, notes, themes, comments) so you can read or modify raw XML, then validate and repack. Visual validation uses generated thumbnail grids to catch cutoff, overlap, and contrast issues before finalizing.

When to use it

  • Create a new presentation from scratch with precise layout and design control
  • Convert HTML slide content or templates into a .pptx file with accurate positioning
  • Edit slide text, speaker notes, comments, layouts, themes, or embedded media via OOXML
  • Analyze a presentation’s typography, color palette, or template structure before redesign
  • Produce thumbnail grids to QA visual issues (text cutoff, overlap, contrast)

Best practices

  • Always state your content-informed design approach before generating slides
  • Use web-safe fonts only (e.g., Arial, Helvetica, Times New Roman) for compatibility
  • Analyze theme and sample slides first to extract font and color tokens from theme1.xml
  • Prefer two-column or full-slide layouts for charts and tables; do not vertically stack
  • Validate the OOXML after each XML edit and generate thumbnails to confirm visual quality

Example use cases

  • Generate a 10-slide investor pitch from an outline and branded color palette
  • Adapt an existing company template by inventorying template slides and mapping content
  • Extract all speaker notes and comments for a review cycle or compliance archive
  • Replace placeholder charts with updated data and export new thumbnails for QA
  • Fix a corrupted slide by unpacking the .pptx, repairing slideN.xml, and repacking

FAQ

Yes. Speaker notes and modern comments live in specific OOXML files (notesSlideN.xml and modernComment_*.xml), so unpacking is required for direct edits.

How do I ensure text won’t be cut off after conversion?

Generate thumbnail grids, inspect each slide for cutoff and overlap, adjust HTML margins/spacing, then regenerate until all issues are resolved.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational