Turn Images into Editable Slides with Image2editable

Where this fits

Imagine you've just received a deck of design mockups as high-resolution images, or a PDF with slide layouts that you need to turn into a presentation that colleagues can actually edit. Rebuilding every text box and shape from scratch is exactly the kind of repetitive, detail-heavy work that wastes hours. That's where Image2editable steps in: it converts images, PDFs, and image-based PPTX files into editable PowerPoint slides, so your agent can do the heavy lifting and you're left with a working .pptx file instead of a static render.

Why agents benefit

• Automatic content extraction: Recognizes text and visual elements, placing them as editable PowerPoint components rather than flattened images.
• Layout preservation: Maintains the original page structure, so slide dimensions, object positions, and sizes survive the conversion—critical for design fidelity.
• Mixed-file handling: When the input is a PPTX with some original editable content, that content stays intact alongside the newly recognized elements.
• Batch capability: Processes multiple images or PDFs in one pass, letting an agent convert an entire folder of slide decks without manual intervention.
• CLI/skill integration: Works as both a standalone CLI and a skill for Codex and Claude Code, enabling agents to invoke it directly in a workflow.

Practical scenarios

Course slide screenshots → editable lecture notes

A student or instructor captures slides from a video lecture as screenshots. Instead of retyping the text and rebuilding charts, the agent converts them all into an editable PPTX, making it easy to annotate, correct, or share with peers.

Design mockup PDF → stakeholder-editable presentation

A product designer exports a PDF of UI mockups. The agent turns that PDF into a PowerPoint where text can be changed, elements moved, and notes added—so non-designers can iterate without touching a design tool.

Scanned report pages → data-updatable slides

A business analyst receives a scanned report as images. The agent converts each page into a slide with recognized text and numbers, then the analyst can update figures and regenerate the deck instead of re-creating every slide.

Add it to your agent workflow

Install the skill with:

npx skills add DSY-Xueai/image2editable

Once added, your agent can call the CLI directly. For example, to convert a single image to a 16:9 PPTX:

image2editable input.png --aspect 16:9 --output editable.pptx

Or from a PDF: image2editable report.pdf --output report_editable.pptx. Inside Claude Code or Codex, you can simply ask: "Convert design.pdf into an editable PPTX, preserving the layout." The agent will handle the invocation and output the file.

Compare before adopting

Look at the conversion quality of text recognition (OCR accuracy) and how well visual elements are decomposed into editable shapes. Check the repository's recent commit activity and issue resolution—10 stars suggests early adoption, so verify it's actively maintained before relying on it. Also weigh alternatives: python-pptx for programmatic creation, or OCR libraries like Tesseract combined with manual layout reconstruction. Finally, confirm the CLI and skill work in your agent's environment (Python 3.10–3.12, Windows/Linux/macOS).

Why it is worth tracking

Image2editable solves a real gap: turning static visual content into editable artifacts without rebuilding from scratch. For agents that need to produce deliverables like reports, slide decks, or client-ready documents, this skill reduces both token usage and manual effort. The tool is early—only 10 stars—but its focused feature set and dual skill/CLI design make it worth evaluating now. Watch for updates in OCR accuracy, handling of complex graphics, and community adoption. You can find it on OpenAgentSkill to stay informed as it evolves.