image-to-editable-ppt-skill

Why we included this project

Anyone who has tried to edit a slide deck that arrived as screenshots, scanned PDFs, or image-only PPTX files knows the usual options are grim: retype everything or redraw it. This Codex skill automates that conversion. It splits the input into per-page tasks and rebuilds a genuine .pptx, restoring readable text as native text boxes, simple geometry as PowerPoint shapes, and complex visuals as separate image assets with source records. The work runs as a multi-agent pipeline inside OpenAI's Codex, dispatching page workers, running OCR, and looping through rebuild, self-check, and fix until the result is close to the original. That thoroughness costs tokens and time, often ten minutes or more per page, so it is aimed at people who genuinely need an editable deck rather than a quick visual patch. If you only want a one-off fix on a single slide, a lighter image-editing approach will serve you better.