Image to Editable PPT skill
Image to Editable PPT Skill
A Codex skill that rebuilds slides from images, PDFs and image-based PPTX files into an editable PowerPoint deck
- 3,327 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Runs the editppt tool, writes files and spawns subagents
- Sends page images to external recognition and image-generation services
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/image-to-editable-ppt-skillDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Runs the editppt tool, writes files and spawns subagents
- Sends page images to external recognition and image-generation services
This is third-party code. Review the repository files before installing.
What it does
The skill takes slides that exist only as images and reconstructs an editable pptx from them. It decomposes each page: it recognizes text, separates background from foreground, recreates simple shapes and places visual assets by coordinates. Work runs through its own editppt tool, which keeps run state deterministic and resumable. Multi-page inputs are spread across parallel workers by the configured slot count; single-page inputs are rebuilt in place. For text accuracy it can call Baidu PaddleOCR-VL with a token, and image generation and editing go through the agent's built-in tool or a fallback path via Codex.
Who it is for. For anyone who needs to edit a deck again when it arrived only as screenshots, images or an image-based PPTX.
Good fit when
- You have slides only as images, a PDF or an image-based PPTX and need an editable pptx
- You need text, shapes and visual assets separated so you can change them independently
- You need to reconstruct a single slide or a whole deck while keeping speaker notes
Not a fit when
- You need to author a deck from scratch out of text, a report or an idea rather than rebuild existing slides
- Editability does not matter and editing the slide image directly is simpler
Example request
Turn this slide PDF into an editable PowerPoint presentationLimitations
Reconstruction spends a noticeable amount of tokens: the author estimates a single slide can take over ten minutes, and a ten-page deck can use up a full five-hour window, so a GPT Pro plan is recommended. The skill relies on image generation and authorization through Codex or an OpenAI-compatible API, which are not reachable from Russia without a VPN. PaddleOCR recognition needs a free Baidu token; without it text quality is lower. Multi-page input requires an agent that can dispatch subagents. Stable results are stated from models at the gpt-5.5 level and up.
How to disable. Delete the image-to-editable-ppt folder from your agent's skills directory, for example ~/.codex/skills/, and restart the agent. Credentials and the OCR token are stored separately in ~/.editppt/config.yaml.
FAQ
Does the skill create new decks?
No. It only rebuilds existing slides from images into an editable form. For authoring a deck from scratch the author offers a separate skill, codex-ppt-skill.
Is an OCR token required?
No, but it helps. Without a Baidu PaddleOCR-VL token the skill uses a built-in offline detector that measures geometry only and does not read text, so text reconstruction quality is lower.
Discussion
No questions or reviews yet.
Related
SkillsA Codex skill that rebuilds a slide from an image or PDF into PowerPoint where text, tables and arrows stay editable objects.
The image-to-ppt skill turns images, PDFs and picture-only slides into a PowerPoint deck whose text and elements can be edited.
Skill turns the agent into a deck producer: brief, storyboard, slide posters, QA and a PDF or editable PPTX.
A skill that turns an article, report or notes into a deck of full-slide images and assembles them into a pptx
SlideMuse turns a PDF, paper or notes into an image-based deck in 15 styles and can rebuild it as an editable PPTX
A skill that builds a deck through dialog and previews: structure and style first, then page images generated and packed into PPTX
README in short
The README describes a skill that makes editable the slides that exist only as images: one image, several images, a multi-page PDF or an image-based PPTX. Work runs through the editppt tool in stages of decomposition, reconstruction, validation and assembly, with run state kept in a separate output directory. The skill automatically runs recognition, image generation and editing, file writes and subagents, so Full Access mode in Codex is recommended. It installs with a one-sentence request to the agent or manually from a zip into the skills directory, while API keys and the OCR token live outside the skill folder in ~/.editppt/config.yaml. Documentation is available in English, Chinese and Korean. MIT licensed.
README badge
Original SKILL.md
--- name: image-to-editable-ppt description: Rebuild slide images, scanned or image-based PPT/PPTX files, and PDF decks into object-level editable PowerPoint (.pptx), preserving speaker notes when supplied. Use for making visual slides editable or reconstructing slides from screenshots; not for authoring new presentations from scratch. --- # Image to Editable PPT ## Overview Use the `editppt` runtime to decompose, reconstruct, validate, and assemble visual slides as editable `.pptx`. Inputs may be single or multiple images, PDF, or image-based PPT/PPTX. ## References Each rule in this skill has exactly one authoritative home; the other files point to it instead of restating it. - `prompts/page-worker.md`: execution template for page workers — ownership boundary, execution order, required outputs, and return format. The parent agent uses it when generating page-worker prompts. - `scripts/build-page-worker-prompt.py`: skill-local prompt builder. It reads `prompts/page-worker.md`, fills run/page paths, writes `worker-prompt.md`, and prints the dispatch command template. - `references/cli-helper.md`: CLI install check (Pre-Run Check), command tree, and command syntax examples. Read it when deciding which `editppt` command to call. - `references/manifest-schema.md`: the single home for JSON field contracts of deck/page/image artifacts. - `references/page-decision-tree.md`: the single source of truth for page object decisions.