pptx

Design each slide as an HTML file with fixed body dimensions (720pt x 405pt for 16:9), then use the html2pptx.js library with PptxGenJS to convert them into slides. Charts and tables are added to placeholder areas via the PptxGenJS API before saving with pptx.writeFile().

Unpack the .pptx with the unpack.py script to expose its OOXML structure, edit the slide XML files directly, validate changes with validate.py against the original, then repack with pack.py. This is required for comments, speaker notes, layouts, and animations.

Run python -m markitdown on the .pptx file to convert its text content to Markdown. For structured extraction with positions, fonts, and formatting, use the inventory.py script which outputs a JSON inventory of all text shapes.

No, CSS gradients do not convert to PowerPoint and cause validation errors. Rasterize gradients as PNG images using Sharp first, then reference them as background images in the HTML.

Common causes include using # prefixes on hex colors in PptxGenJS, broken relationship references after slide duplication, or missing Content_Types.xml declarations. Always run validate.py after OOXML edits and check _rels files for orphaned references.

Extract the template's text and generate thumbnail grids to inventory its layouts, then use rearrange.py to duplicate and reorder template slides, and replace.py with a JSON file of formatted paragraphs to populate placeholders while preserving styling.