Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add guliqianxun/research-skills --skill latex-to-wordgit clone --depth 1 https://github.com/guliqianxun/research-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/guliqianxun/research-skills/latex-to-word)<a href="https://agentmods.dev/skills/guliqianxun/research-skills/latex-to-word"><img src="https://agentmods.dev/badge/skills/guliqianxun/research-skills/latex-to-word/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/guliqianxun/research-skills/latex-to-word"><img src="https://agentmods.dev/badge/skills/guliqianxun/research-skills/latex-to-word.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00180 | $0.03035 |
| Opus 5.5 | $0.00072 | $0.01214 |
| Sonnet 5.5 | $0.00036 | $0.00607 |
| Haiku 4.5 | $0.00018 | $0.00303 |
Grade A, and why
latex-to-word scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.
LaTeX → Word
A naive pandoc manuscript.tex -o manuscript.docx almost never produces a usable Word file: figure/table cross-reference lists are empty, equation numbers disappear, bibliography numbering is wrong, ctex-specific layout commands (\zihao{}, \kaishu, \hangindent=...) leak through as literal text, and the styling drifts from the target template. This skill implements a four-stage pipeline (ctex-aware preprocess → pandoc with reference doc → SEQ-field postprocess → equation + style postprocess) that fixes all four problems at once.
1. Pick the right path
Ask the user these three questions before running anything — the answers determine which path to use:
- Does the output need to be editable in Word, or just look like the PDF?
- Do cross-references to figures/tables/equations/bibliography need to work inside Word?
- Should it match an existing Word template / house style (e.g. 气象与环境学报)?
| Path | When | Command |
|---|---|---|
| Editable (main) | Editable deliverable; cross-refs must work; template styling matters | scripts/latex_to_docx.py |
| Review | Quick internal read-through of text only; not a deliverable | pandoc src.tex -o out.docx (plain) |
| Fidelity | Must be .docx but visually identical to PDF; editing not needed |
Compile to PDF, then rasterize pages into a .docx (not in this skill — tell the user to use pdftoppm + python-docx if they need this) |
Default to Editable unless the user explicitly asks for one of the others.
2. Why this takes four stages
LaTeX and Word have different reference systems, and ctex adds Chinese-academic layout commands pandoc can't parse:
\cite{key}— LaTeX computes the number at compile time. Word expects an actual numbered list that cross-reference dialogs can see.\ref{fig:x}/\ref{tab:x}— In Word, cross-references require both a bookmark on the caption and aSEQ Figure/SEQ Tablefield in the caption paragraph. Pandoc creates the bookmark but not the SEQ field.\eqref{eq:y}— Pandoc silently drops equation numbering. Cross-refs need us to (a) number each display equation with aSEQ Equationfield, (b) bookmark the number, (c) emitREFfields in the body.\zihao{},\kaishu,\bfseries,\hangindent=...,\noindent,\hfill,\vspace{},\rule{}— ctex-only layout commands; pandoc either ignores them or leaks them as literal text (=2.4em 摘要:is the classic symptom).
What ships with it
6 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2mo ago First seen · 129 lines · 180 tokens per session scan A 80ca9bff3b27
latex-to-word is a skill published in the GitHub repository guliqianxun/research-skills (6 stars, last pushed 7d ago), licensed MIT. It adds 180 tokens to every session and 3,035 once invoked, about $0.0007 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
latex-to-word compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| latex-to-wordguliqianxun/research-skills | Skill | 6 | 7d ago | MIT | A | 180 | Claude Code, Codex |
| orbit-notionnexu-io/open-design | Skill | 106k | yesterday | Apache-2.0 | A | 117 | Claude Code, Codex |
| instrument-data-to-allotropeanthropics/knowledge-work-plugins | Skill | 35k | 2d ago | Apache-2.0 | A | 123 | Claude Code |
| apify-youtube-transcript-scrapersickn33/agentic-awesome-skills | Skill | 49k | yesterday | MIT | A | 48 | Claude Code |
| legal-holdanthropics/claude-for-legal | Skill | 9.9k | 3d ago | Apache-2.0 | A | 68 | Claude Code |
| ppt-reviewOrkas-AI/Orkas | Skill | 9.0k | yesterday | MIT | A | 2 | Claude Code, Codex |
| feishuHmbown/CodeWhale | Skill | 41k | 6d ago | MIT | A | 33 | Claude Code, Codex |