#128128th of 389 in Skills, plugins and promptsFor data and documents, documents and ocr
image-to-editable-ppt-skill
Skill that rebuilds slide images and PDFs as editable PowerPoint files
- Stars
- 3.3k
- License
- MIT
- Last commit
- Sep 2026
- Last release
- Sep 2026
- Language
- Python
12 on 19 May 20263,320 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
An agent skill, mainly for Codex, that converts images, PDFs and image-based .pptx files into editable .pptx. It uses OCR measurements (Baidu PaddleOCR-VL) to restore text boxes and font sizes, rebuilds simple shapes and native tables, and keeps complex visuals as separate image assets. Multi-page inputs are split across page workers; a CLI called editppt records state and validates output.
Who it is for: Codex users who need editable PPTX from slide screenshots, PDFs or image decks
Strengths
- Accepts single images, multiple images, multi-page PDFs and image-based .pptx, preserving page order
- Restores text as native boxes with OCR-measured sizes and consistent font-size groups
- Rebuilds plain tables as native PowerPoint tables and copies .pptx page notes unchanged
- Per-page manifests, validation and preview files allow local repair of failed pages
Weaknesses
- High token cost; a 10-page deck may exhaust a ChatGPT Plus 5-hour quota
- A single page can take over 10 minutes to reconstruct
- Photos, illustrations and textures stay as images; not guaranteed to match the original exactly
- Needs an agent with skill loading and subagents; results below gpt-5.5 not guaranteed
What it needs
- no GPU
- Needs Codex or a compatible agent runtime, Baidu AI Studio PaddleOCR-VL token (recommended), Image generation: Codex image_gen, Codex OAuth or OpenAI-compatible API, editppt CLI
- Models: gpt-5.5 or newer recommended, OpenAI-compatible image APIs
Quick answers
Is image-to-editable-ppt-skill open source?
Yes. Its code is on GitHub under the MIT license.
Also in Skills, plugins and prompts
See all 389| Rank | Project | Score |
|---|---|---|
| 1 | claude-memPersistent cross-session memory plugin for Claude Code and other coding agents | 97 out of 100 |
| 2 | impeccableDesign skill with 24 slash commands for AI coding agents' frontend output | 97 out of 100 |
| 3 | gstackMarkdown slash-command skills that give Claude Code planning, review, QA and ship roles | 96 out of 100 |
| #4 | rufloAgent harness adding swarms, memory and hooks to Claude Code and Codex | 96 out of 100 |
| #5 | cavemanShortens coding-agent replies and compresses tool output before the model reads it | 95 out of 100 |