yuanzhao321/scientific-writing-zh — explained in plain English
Analysis updated 2026-05-18
Get help tightening and improving a paragraph of a research paper.
Check whether an abstract is specific enough and within length limits.
Draft a response to a peer reviewer's comments.
Write a peer review report for someone else's manuscript.
| yuanzhao321/scientific-writing-zh | afadtc/afa-dtc-skills | alibaba-multimodal-industrial-ai/industrybench | |
|---|---|---|---|
| Stars | 66 | 66 | 66 |
| Language | — | Python | Python |
| Setup difficulty | easy | easy | moderate |
| Complexity | 1/5 | 2/5 | 3/5 |
| Audience | researcher | pm founder | researcher |
Figures from each repo's GitHub metadata at analysis time.
Requires Claude Code, install by cloning the repo into the ~/.claude/skills/ directory.
This repository is a Claude Code skill that acts as a scientific writing coach, with all of its explanations written in Chinese. It is meant for researchers who write in Chinese but usually submit papers to English language international journals, so while the guidance itself is in Chinese, the before and after example sentences stay in English since that is the language being polished. Once installed, you do not need to invoke it by name. If you are writing a paper, abstract, title, figure caption, reviewer response, or report, Claude automatically pulls in the relevant guidance, even from a casual request like asking to tighten up a paragraph or checking whether an abstract is good enough. The skill covers the full cycle of academic communication: deciding whether and where to publish, choosing a target journal, handling proofs, promoting a paper, brainstorming and outlining, drafting, structuring a paper's title, abstract, introduction, methods, results, and discussion, refining prose at the paragraph, sentence, and word level, revising through what it calls a writing and editing funnel, handling figures, tables, equations, and citations, authorship order and research ethics, guidance specifically for non-native English writers, peer review and responding to reviewers, and conference materials like abstracts, slides, and posters. It is built using progressive disclosure, meaning a small SKILL.md file handles routing to the relevant guidance, while separate reference files organized by topic are only loaded when a task actually matches them. The file structure mirrors an existing English language version of the same skill, keeping the same file names but with Chinese content throughout. To install it, you clone the repository directly into Claude Code's skills folder, then restart Claude Code and confirm it shows up when you run the skills command. The skill files themselves are released under the MIT License.
A Claude Code skill that coaches researchers through academic writing and peer review, explained entirely in Chinese.
Use, modify, and distribute freely, including for commercial purposes, as long as you keep the copyright notice.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly researcher.
This repo across BitVibe Labs
double-check against the repo, no cap.