paper-research
Skill1 starsChanged 32 days agoApache-2.0View on GitHub
带用户完成文献检索、PDF 精读(正文/表格/公式)、参考文献格式整理(国标 GB/T 7714 与 BibTeX)以及英文文献翻译精读。当用户要查文献、找论文、读某篇 PDF、按国标整理参考文献,或要翻译英文论文时使用。
5 other projects use a version of this.
Adopted by
Most recently taken up by
- cabbage2000-lab/paper-tutor-skillstheir own version, first seen Nov 8, 2026
- zhouyiyulibaitian/zcode-paper-research-skilltheir own version, first seen Nov 8, 2026
- stophobia/deerflow2.0-enhancedtheir own version, first seen Oct 8, 2026
- parkjui92/socsci-paper-kittheir own version, first seen Sep 26, 2026
- yogsoth-ai/convergencetheir own version, first seen May 19, 2026
Other versions5 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 5
- stophobia/deerflow2.0-enhanced★ 821same name
- cabbage2000-lab/paper-tutor-skills★ 48same name
- parkjui92/socsci-paper-kit★ 5same name
- yogsoth-ai/convergence★ 2same name
- zhouyiyulibaitian/zcode-paper-research-skill★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository18 other files it gives its agents.
- AGENTS.mdAGENTS.md
- llms.txtllms.txt
- academic-figuresSkill
- citation-managementSkill
- cn-literatureSkill
- cpp-workflowSkill
- data-analysisSkill
- design-materialSkill
- doc-batchSkill
- email-calendarSkill
- lab-reportSkill
- math-modelingSkill
- meeting-notesSkill
- office-docsSkill
- python-workflowSkill
- study-tutorSkill
- thesis-formatSkill
- web-researchSkill
The file
Files
文献研究工作流
四段:检索 → 精读 → 引用格式化 → 翻译。 全程合规:只做公开可得的文献检索,不绕过 CNKI 等收费库的付费墙,不代替用户登录下载。
1. 检索
优先用已配置的 MCP 工具,没有 MCP 时用联网搜索:
arxivMCP:适合预印本与理工科最新论文,可按关键词 / 作者 / 分类(如cs.CL)检索semantic-scholarMCP:适合按主题找论文,并拿到引用数、参考文献、作者信息- 都没配:用网页搜索,把检索式交给用户
构造检索式按「主题词 + 限定」拆:
- 同义词用 OR,不同概念用 AND(例:
("large language model" OR "LLM") AND education) - 限定年份、文献类型(期刊 / 会议 / 预印本)、语种
- 中文库(知网 / 万方 / 维普)只做检索式建议和已公开摘要的整理,不代替用户登录下载
检索后先给候选清单:标题、作者、年份、出处 + 一句话说明为什么相关,让用户挑,再进入精读。
2. PDF 精读
阅读顺序建议:摘要 → 引言最后一段(贡献)→ 方法 → 实验 → 结论。
提取内容时区分三类,分别处理:
- 正文文字:直接抽取文本。扫描版 PDF 没有文本层,需要 OCR —— 先试着抽一段,抽不出字就是扫描版
- 表格:按行列还原成 markdown 表,注意表头与单位;有合并单元格要说明
- 公式:不要凭记忆写,把公式按可见符号逐项转成 LaTeX;不确定的符号标注「原文如此」
精读产物给这几点:研究问题、方法、数据、结论、局限,以及与用户用途的关系。 引用原文句子时标页码,方便用户核对。
3. 参考文献格式
GB/T 7714(国内毕业论文硬性要求,务必优先)
这是中文学位论文、期刊投稿的强制格式,各校模板可能微调,但结构固定。常见类型:
- 期刊论文(J):
[序号] 作者. 题名[J]. 刊名, 年, 卷(期): 起止页码.例:[1] 张三, 李四. 深度学习方法综述[J]. 计算机学报, 2023, 46(5): 1001-1020. - 专著(M):
[序号] 作者. 书名[M]. 版本. 出版地: 出版社, 出版年: 起止页码. - 学位论文(D):
[序号] 作者. 题名[D]. 保存地: 保存单位, 年份. - 会议论文(C):
[序号] 作者. 题名[C]//论文集名. 出版地: 出版社, 年: 页码. - 电子资源(EB/OL):
[序号] 作者. 题名[EB/OL]. (发布日期)[引用日期]. URL. - 预印本可归为
[EB/OL],并注明 arXiv 编号
规则要点:
- 作者:姓在前名在后,中外一致;3 位以内全列,超过 3 位列前 3 位后加「, 等」或「, et al.」
- 文献类型标识必须写:
[J] [M] [D] [C] [EB/OL],漏了就算格式错 - 卷期、年份、页码之间的标点按上面模板;用中文还是西文标点以学校模板为准,先问用户
- 序号用
[1],正文中引用写[1]或[1-3]
BibTeX(投稿 / LaTeX 环境)
@article{zhang2023survey,
author = {Zhang, San and Li, Si},
title = {A Survey of Deep Learning Methods},
journal = {Journal of Computer Science},
year = {2023},
volume = {46},
number = {5},
pages = {1001--1020},
doi = {10.xxxx/xxxxx}
}
@inproceedings{...} % 会议论文
@book{...} % 专著
@misc{...} % 网页;预印本建议加 eprint 与 archivePrefix = {arXiv}- key 命名建议
第一作者姓+年份+关键词,全库唯一 author里多人用and连接,不是逗号;pages用--双连字符- 有 DOI 一定要填
规范化操作:用户给一批题录(或让 MCP / 搜索拿到元数据)后,输出格式化完整的列表,并标注缺哪些字段、需要用户回原文补。
4. 英文文献翻译精读
不做整篇直译堆砌,做「对照精读」:
- 分段落翻译;术语保留原文,首次出现时「中文(English)」并列
- 长句先拆主谓宾再译,避免语序混乱
- 专业术语用领域通行译法;不确定的保留英文并标注
- 译完给一句「本段要点」,帮用户抓重点
- 引用他人译文或机翻时要说明来源,不要冒充原创翻译
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

