#34 · Primary category: Self-Hosted File Processing & Conversion

codex-visio-paper-figure-skill

academic-figures codex editable-diagrams image-to-visio scientific-figures skills visio

Skill for scientific and paper figure drawing, converting images to editable Visio format for easy detail adjustments.

Project last updated:08/16/26

GitHub Stars

786

Forks

37

Contributors

3

License

MIT

Why we included this project

Anyone who has tried to edit an AI-generated figure knows the frustration: the image looks fine until you need to change a label, and then you are stuck redrawing it. This skill fixes that by having Codex rebuild a reference image, screenshot, or generated diagram as a native Visio .vsdx file made of real shapes, text, connectors, and groups instead of a flattened bitmap, then exports the same source to PNG, SVG, PDF, and PPTX so every format stays consistent. It is aimed at architecture, model, and flow diagrams, and includes PowerShell scaffolding plus panel-calibration helpers that keep complex multi-panel figures from drifting or overlapping. One honest caveat: full reconstruction depends on Visio COM automation, so you need Windows with Visio installed; without it you can still inspect the .vsdx package structure but not do a complete rebuild.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category