Qinyan Academic Skills
Research skills for literature discovery, manuscript drafting and checking claims against the study's original evidence
- 1,011 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Individual skills run scripts, install libraries and write results to disk
- Search queries and supplied research material may reach external services
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/qinyan-academic-skillsDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Individual skills run scripts, install libraries and write results to disk
- Search queries and supplied research material may reach external services
This is third-party code. Review the repository files before installing.
What it does
Qinyan Academic Skills supplies agent instructions for different stages of research. The collection covers literature discovery, manuscript preparation and research data. Qinyan's scientific-writing suite helps formulate the main conclusion from the author’s data and link it to supporting material. The agent links statements to tables or experiments, records missing facts and drafts manuscript sections. Separate skills cover language revision, pre-submission review and publication figures. The Qinyan paper-search skill calls an external API spanning several scholarly databases. Other parts of the collection describe analysis libraries and specialist resources. Required software and credentials depend on the chosen skill.
Who it is for. For researchers, teachers and scientific editors who need focused workflows for literature and their own study results.
Good fit when
- Build a manuscript argument from existing results and tables without adding unsupported conclusions
- Prepare a pre-submission list of structural and evidence issues
- Find publications for a literature review and preserve their references
- Install a focused figure or statistics workflow rather than the whole collection
Not a fit when
- You expect new experimental results from a description of the desired conclusion
- You need automatic replacement of peer review or guaranteed journal acceptance
- Your environment forbids external requests but you need remote literature search
Example request
Read Qinyan's scientific-writing skill. Use my tables to map claims to evidence, then propose the results-section outline and list facts that still need confirmation.Limitations
Some skills contain scripts and require extra libraries; inspect the chosen directory’s instructions before running them. Qinyan search requires an API key, curl and Python 3, and sends queries to an external operator. The agent can misjudge evidence and references, so the study author must review the manuscript before submission. The install.sh script downloads another repository, qinyan-ai/qinyan-academic-skills; copy the chosen folder manually to use this checkout.
How to disable. Remove the selected directory from your agent's skills folder. Generated manuscripts and tables remain on disk. Revoke the search-service key separately if it is no longer needed.
FAQ
How do I install one skill in Claude Code or Codex?
Copy its complete directory from `skills/` into your agent's skills folder. The collection lists project-local `.claude/skills/` for Claude Code and `.codex/skills/` for Codex. Check that the client discovers the installed name.
What input does the writing skill need?
Provide study results, tables and the limits of the conclusions they support. It maps claims to evidence and marks missing facts as `AUTHOR_INPUT_NEEDED`. Incomplete evidence first produces an outline and a gap list.
Why not copy only `SKILL.md`?
Instructions refer to adjacent resources. The writing skill includes section and submission guides plus a manuscript-checking script. Without these directories, some steps lack their input files.
Where are subjects beyond scientific writing?
Browse the topic directories under `skills/`. They include bioinformatics, molecular analysis and document workflows. Each has its own dependencies; installing the instructions does not install every library.
How does polishing differ from pre-submission review?
Polishing revises structure and wording. Pre-submission review looks for problems in claims, methods and evidence and returns findings. A person remains responsible for evaluating the research.
Is the collection affiliated with Nature?
No. The Nature-style suite is independent. The main repository uses MIT; derived skills retain their own terms.
Discussion
No questions or reviews yet.
Related
SkillsA skill set for research work in the style of Nature journals: finding and reading papers, writing, statistics, figures, reviewer responses
A set of three research skills: paper writing, academic Word and PowerPoint documents, scientific computing and figures
Skill with 30 prompt templates for research papers in English and Chinese, from literature review to grant proposals
Nursing Research Skills (护研搭子)
Nursing Research Skills
18 skills for nursing research, from literature search and statistics to the manuscript and reviewer replies
Seven skills for planning research, reviewing literature, preparing experiment code and drafting a paper with traceable sources
Claude Code skills for the academic pipeline: research, paper drafting, peer review and citation checks
README in short
The collection includes Qinyan’s own material and adaptations of other research skill sets. Attribution and licenses are recorded inside the directories; check those for the skills you choose.
README badge
Original SKILL.md
--- name: qinyan-nature-writing description: 面向 Nature、Nature Communications 及同等级综合期刊的证据驱动论文写作技能。用于从已核验的结果、图表、实验记录或中英文草稿构建标题、摘要、引言、结果、方法、讨论、结论和首次投稿材料；也用于重组论文论证、建立 claim–evidence map、控制声称边界和开展投稿前完整性检查。触发场景包括 Nature 写作、论文起草、章节重写、scientific writing、manuscript drafting、abstract、introduction、results narrative、discussion、cover letter、投稿材料。 --- # 沁言 Nature 论文写作 把论文视为一条可审计的论证链，而不是一组“像顶刊”的句子。先锁定事实和证据，再安排读者路径，最后写作。 ## 工作边界 - 处理从材料到论文章节的起草、重构与首次投稿材料。 - 将纯语言精修交给 `qinyan-nature-polishing`。 - 将投稿前同行评审交给 `qinyan-nature-review`。 - 将统计设计与报告核验交给 `qinyan-nature-statistics`。 - 将科研图设计和导出交给 `qinyan-nature-figures`。 - 不生成不存在的结果、机制、统计显著性、参考文献、实验细节或期刊政策。 ## 必须先建立的写作底稿 从用户材料中提取并明确标注： 1. `Central claim`：论文实际证明或支持的核心命题。 2. `Evidence set`：支撑命题的图、表、实验、模型、数据集和统计结果。 3. `Boundary`：证据不能支持的外推范围、机制解释或因果结论。 4. `Audience`：跨领域读者为何应关心，以及专业读者需要看到什么。 5. `Terminology ledger`：方法、数据集、指标、缩写、符号的唯一标准写法。 6. `Unresolved facts`：仍需作者确认的事实；统一写为 `AUTHOR_INPUT_NEEDED`。 如果核心命题、关键证据或边界缺失，先给出可继续工作的框架与缺口清单，不补造正文事实。 ## 执行流程 1. **分类任务。** 确定论文类型、目标章节、目标期刊、源语言、字数限制和交付格式。 2. **构建论证脊柱。** 用一句话表达“研究对象—关键进展—方法—证据—适用边界”。 3. **建立证据矩阵。** 为每个主要声称绑定至少一个证据位置；把无证据声称降级、删除或标为待补。 4. **设计章节职责。** 让每段只承担一个主要功能：背景、缺口、方法、发现、比较、解释、意义或局限。 5. **按证据顺序起草。** 先写结果与图件逻辑，再写方法和讨论，最后收束摘要与标题；用户指定其他顺序时服从用户。 6. **校准表达强度。** 区分观察、关联、预测、干预和机制证据；使动词强度与证据层级一致。 7. **运行结构审计。** 对文本文件执行 `python scripts/manuscript_audit.py <file>`，处理 blocker 与 warning。 8. **交付并暴露缺口。** 给出可粘贴正文、关键编辑说明、证据风险和作者待确认项。 需要章节级结构时读取 [references/section-blueprints.md](references/section-blueprints.md)。需要完整论证与证据映射方法时读取 [references/argument-workflow.md](references/argument-workflow.md)。需要首次投稿材料时读取 [references/submission-package.md](references/submission-package.md)。 ## 默认输出