论文常用的tex格式转换为word格式，核心是能转换数学公式（非图片），表格，图表等等
ReviewAudited by ClawScan on May 10, 2026.
Overview
The converter is mostly purpose-aligned, but its script silently inserts a hard-coded company address and email into generated Word documents.
Before installing or using this skill, edit scripts/tex2docx.py to remove or parameterize the hard-coded company/address/email block, install dependencies only from trusted sources, run the converter on copies of your LaTeX project, and review the final DOCX carefully before sharing or submitting it.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
A user could unknowingly produce or submit a Word document containing someone else's affiliation and email address.
The converter adds a fixed company/location/email block after the title and author, but the skill description only promises format conversion.
P('State Grid Quanzhou Power Supply Company\nFengze, Quanzhou, Fujian, China\nEmail: huangchuzhi81@gmail.com',Remove this hard-coded block, replace it with data extracted from the user's LaTeX source, or clearly prompt the user before adding any affiliation/contact text.
The tool will process local document content and images from the selected paper directory.
The script reads the requested LaTeX file, invokes pandoc conversion, and embeds local figures, which is expected for this converter but still affects local files.
tex = open(tex_path, encoding='utf-8').read(); pypandoc.convert_file(...); fp.add_run().add_picture(found, width=Inches(3.3))
Run it only on intended paper directories, preferably on a copy of the source files, and inspect the generated DOCX before sharing.
Installing dependencies from untrusted or unexpected sources could affect the user's local environment.
The skill relies on external manually installed software and unpinned Python packages; this is normal for a converter, but users must trust those sources.
- **pandoc** (system install): `winget install pandoc` or pandoc.org - **Python packages**: `pip install python-docx lxml pypandoc_binary`
Install dependencies from official package sources, consider pinning versions, and use an isolated Python environment.