Analysis updated 2026-05-18
Convert a LaTeX manuscript into a polished Word document for a journal that requires Word submissions.
Keep citation fields in the Word document live and synced with a Zotero library.
Audit a document to check that every citation key matches an entry in the bibliography.
Preserve figure, table, and equation numbering as native Word cross references.
| hajimi-kun/latex-to-word-workflow | fainshare/liam-skills | michaelliao/miniweibo | |
|---|---|---|---|
| Stars | 83 | 83 | 83 |
| Language | Python | Python | Python |
| Last pushed | — | — | 2012-12-28 |
| Maintenance | — | — | Dormant |
| Setup difficulty | moderate | easy | moderate |
| Complexity | 3/5 | 3/5 | 3/5 |
| Audience | researcher | developer | developer |
Figures from each repo's GitHub metadata at analysis time.
Requires Pandoc, LaTeX, Python dependencies, desktop Word, and optionally Zotero with Better BibTeX for live citations.
This project is a skill for AI coding agents like Claude Code or Codex that converts academic papers written in LaTeX into properly formatted, editable Microsoft Word documents. Rather than a rough conversion, it aims to preserve figures, equations, tables, custom document styles, and citations so the resulting Word file looks like a real polished manuscript rather than something dumped out of a converter. Citations can be handled two ways. If you use the reference manager Zotero along with a plugin called Better BibTeX, the skill can insert live citation fields that stay connected to your Zotero library and update if your references change. If you do not use Zotero, it can instead insert static citations formatted according to a citation style. It also checks that citation keys in your document actually match entries in your bibliography file, and it uses Word's native cross reference features for figures, tables, and numbered equations so those references stay correct if you reorder content later. The tool relies on Pandoc to do the underlying LaTeX to Word conversion, plus Python libraries for working with Word files, and it needs a working LaTeX and BibTeX setup on your machine. You also need a desktop copy of Microsoft Word so you, the human, can open the finished document, refresh any live citation fields, and do a final visual check, since the skill does not claim to produce a document that needs no human review at all. To install it, you copy or clone the inner skill folder into whatever skills directory your AI agent tool uses, such as the Claude Code skills folder, then install the Python dependencies listed in the requirements file. On first use with an unconfigured setup, the agent is meant to check that all the required tools are installed, confirm a Word template to use as a base, run a small test conversion, and only then convert your full manuscript. The README also lists several similar existing projects for context, noting that this one's particular niche is combining an agent skill, support for both live and static citations, native Word cross references, and a final human review step. The project is released under the MIT license.
An AI agent skill that converts LaTeX academic papers into polished, editable Word documents with working citations, figures, and cross references.
Mainly Python. The stack also includes Python, Pandoc, LaTeX.
Use freely for any purpose, including commercial use, as long as you keep the copyright notice.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly researcher.
This repo across BitVibe Labs
Verify against the repo before relying on details.