Skip to main content
AI Inspiration World
AI AppsAI SkillsAI GamesAI Short Films

⌕
EN|中文
中
AI AppsAI SkillsAI GamesAI Short Films
Home/AI Skills/Visio Image Rebuilder
AI Skills · ResearchVisio Image Rebuilder

Added Oct 14, 2026

Visio Image Rebuilder

Rebuilds paper model figures and flowcharts from reference images as native Visio drawings where every shape and label stays editable.

Ecosystem
Agent Skills
Skill category
Research
Creator / organization
pengjunchi0
View work and source ↗

What it is

Visio Image Rebuilder is a Codex Skill by GitHub user pengjunchi0 for research paper figures. Give it a reference image, a screenshot or an AI-generated model figure, and Codex writes PowerShell scripts that drive Microsoft Visio through its automation interface (COM), redrawing the figure from native rectangles, connectors, text and groups. The result is saved as a .vsdx file, and PNG, SVG, PDF and PPTX are all exported from that same file. You can also restyle an existing .vsdx to match a reference's colors, fonts or module layout, or check whether a .vsdx just has a whole image pasted in. For dense multi-panel figures it calibrates each panel's bounds first and draws the contents in panel-local coordinates, raising an error when something falls outside its panel, so the second half of a figure doesn't drift or overlap. What sets it apart is a firm rule against passing off a pasted reference image as a rebuild: before delivery it checks that no full-size reference image is left in the file, and every shape should be selectable and editable on its own.

Who it's for:researchers and teachers who write papers or present project work, revise model, architecture and flow diagrams often, and already use Visio.

Before you start: you need a Windows PC with Microsoft Visio and PowerShell, plus Microsoft PowerPoint for PPTX export. Visio and PowerPoint are bought or subscribed to separately; check Microsoft's site for pricing. The README recommends Codex Desktop or another Codex setup that can work with local files and tools, and you install it by cloning the repo into Codex's skills folder. The README says full automated drawing depends on Visio's COM interface on Windows; without Visio you can only inspect a .vsdx's package structure or make limited XML edits. The Skill opens, edits and saves the .vsdx you point it to, so keep your own backup of anything important. The author notes it's meant for model, architecture and flow diagrams; charts with real experimental results are better drawn from the actual data with a script, without guessing numbers from an image. The exported PPTX places the Visio page in a slide as an SVG, so individual elements can't be edited in PowerPoint; the editable master is the .vsdx. If you rebuild a figure from someone else's paper, make sure you have the right to use it. It only connects to Visio for now, and draw.io users would need to adapt the interface themselves. The license is MIT.

Based on the project's public documentation. Not tested by AI Inspiration World.

Source

https://github.com/pengjunchi0/codex-visio-paper-figure-skill ↗

Repository ↗

How rankings work →
AI Inspiration World logo: aiw. picked by humansAI Inspiration World logo: aiw. picked by humansAI Inspiration World

Check in daily. The newest in AI lives here.

Updated every day · Latest arrival: Oct 14, 2026

Site version: Oct 11, 2026

About·Methodology·Trust & Safety·@AIInspoWorld