Analysis updated 2026-05-18
Turn a screenshot of a flowchart or architecture diagram into an editable Visio file.
Recreate model diagrams from a research paper so they can be relabeled and restyled.
Convert an AI generated diagram from GPT or Gemini into a maintainable Visio drawing.
Export a rebuilt diagram as SVG or PNG for quick review before finalizing it.
| rss3208/visiomaster | colossus-lab/openarg_backend | vanquishervohonor25/zapret-4.0 | |
|---|---|---|---|
| Stars | 134 | 135 | 135 |
| Language | Python | Python | Python |
| Setup difficulty | hard | hard | moderate |
| Complexity | 3/5 | 5/5 | 3/5 |
| Audience | researcher | developer | general |
Figures from each repo's GitHub metadata at analysis time.
Requires Windows, a licensed Microsoft Visio desktop install, and pywin32 for COM access.
Visiomaster is a Windows focused tool that rebuilds diagrams as editable Microsoft Visio files. It takes flowcharts, architecture diagrams, or academic paper diagrams, whether they already exist as images or were generated by tools like GPT or Gemini, and turns them into a real .vsdx file you can keep editing, along with SVG and PNG exports for quick review. The goal is not to simply paste the original picture into Visio. Instead, the tool breaks the image down into actual Visio shapes, text, and connecting lines, so the result behaves like a diagram someone built directly in Visio rather than an image pasted on a page. The design follows a similar approach to a project called ppt master: first understand the image, break down its structure and style, then drive a rendering script using a stable intermediate description file. The difference is that Visiomaster's end goal is an editable Visio diagram rather than a slide. The core process moves from an original image to an intermediate scene.json file, through structure checking and module review, then to rendering through Visio itself, producing the vsdx, svg, and png outputs, followed by a visual comparison and manual fixes if needed. Typical uses include recreating model architecture diagrams and flowcharts from research papers, or converting AI generated academic diagrams into something you can relabel, rewire, recolor, and lay out for a slide deck. Running it requires Windows, a desktop copy of Microsoft Visio, Python 3.10 or newer, and the Python packages listed in requirements.txt. It works both as a plain command line tool where you prepare a scene file and run validation and rendering scripts yourself, and as a skill for AI coding assistants like Codex or Claude Code, which can read the included SKILL.md and call the same scripts. The image to scene conversion only produces a starting point, not a guaranteed exact reproduction, so complex diagrams typically need manual or AI assisted adjustments before the final render matches the source closely.
A Windows tool that rebuilds flowcharts and diagrams from images into editable Microsoft Visio files.
Mainly Python. The stack also includes Python, Visio COM, pywin32.
The README does not state a license.
Setup difficulty is rated hard, with roughly 1h+ to a first successful run.
Mainly researcher.
This repo across BitVibe Labs
Verify against the repo before relying on details.