Token导航 LogoToken导航TokenDH.com
研究检索需要联网github未标认证来源可访问clear审计通过

paper-polish纸抛光

Agent Skill

paper-polish 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

6,805

周安装

278

GitHub Stars

5

下载量

2,202
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

3

许可证

MIT

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:paper-polish(纸抛光)
来源仓库:https://github.com/eyh0602/skillshub
仓库路径:skills/paper-polish
安装命令:
npx skills add https://github.com/eyh0602/skillshub --skill paper-polish
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/eyh0602/skillshub --skill paper-polish

简介

paper-polish 用于基于关键词查找、检索和筛选相关信息。

  • 适用于在学术或技术研究中快速定位目标文献或参考资料。
  • 通过 GitHub 仓库安装,使用标准 npx 命令即可集成。
  • 使用前请确认数据源权限和网络访问策略,避免不必要的资源调用。
  • 适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

SKILL.md

Polishing and reviewing research papers in LaTeX

On writing

Contents

  • Your paper should be easily comprehensible by its reviewers. They are far less familiar with your work than you. They may not be an expert on the topic and may not be able to afford much time on your paper.
  • The introduction should convey curiosity or excitement (new problem, new solution, improved solution, impressive results, or high impact), the design novelty, substantiality, and correctness, and the evaluation relevancy and comprehensiveness.
  • Conciseness: Remove every word that contributes no meaning, such as kind of.
  • Use proper tenses:

- 1 - 2 - 3

Grammar

- Title Case: Capitalize the first and last words and all major words in between. - Good: Introduction to Fuzzing with LLMs - Sentence case: Capitalize only the first word and proper nouns. - Good: Introduction to fuzzing with LLMs - Choose one style and be consistent throughout the paper.

  • Avoid passive voice unless strongly justifiable. Passive voice is ambiguous because it has no subject unless followed by "by...".

- Bad: LLM was applied to fuzzing. (Who applied it? The authors or someone else?) - Good: We applied LLM to fuzzing.

- Bad: He made a proposal to use Rust. - Good: He proposed to use Rust.

  • Avoid "There is/are".

- Bad: There are many developers of Rust. - Good: Many developers use Rust. - Good: Rust has many developers.

  • "Which" vs "that": Use "which" in a nonrestrictive clause and "that" in a restrictive clause. More...

- Wrong: Rust that is safe is popular. (This is wrong because there is only one Rust.) - Right: Rust, which is safe, is popular.

- Wrong: C is dangerous, Rust is safe. (Cannot join two sentences by a comma) - Right: C is dangerous, but Rust is safe. - Wrong: C is dangerous, however Rust is safe. - Right: C is dangerous; however, Rust is safe.

  • "Fewer" modifies countable nouns whereas "less" uncountable nouns.

- Wrong: ten items or less - Right: ten items or fewer - Wrong: fewer feedback - Right: less feedback

  • Use articles (a, an, the) properly.

- A singular countable noun must be preceded by an article. - Wrong: I wrote Rust program. - Right: I wrote a Rust program. - Right: I wrote Rust programs. - The must have a reference that is unique either by fact or in the context. - Right: the first Rust programmer (unique by fact) - Right: Our team has a Rust and a C++ programmer. The Rust programmer produces the fastest, most robust code. (unique in the context) - Wrong: Our team has two Rust and two C++ programmers. The Rust programmer is more productive than the C++ programmer.

- Right: Rust is safer compared with C. - Right: Some people compare Rust to a panacea for memory safety problems.

On LaTeX

  • Use modern implementations of LaTeX to take advantage of Unicode and other useful features.

- Use LuaLaTeX instead of LaTeX or pdfLaTeX - Use BibLaTeX for acmart and ieeetrans templates instead of BibTex

  • Use correct typefaces. Particularly, *italics* should be used for variables but not for descriptive terms. More...

- Wrong: $t_{max}$ - Right: $t_\text{max}$

  • Do not manually add separators in large numbers. \usepackage{siunitx}. Then, wrap large numbers in \num{}.

- Bad: 12,345 - Good: \num{12345}

  • Do not manually type reference names, such as Table, Figure, Theorem. Instead, \usepackage{hyperref}, and then \autoref{fig:xxx}, \autoref{sec:xxx}, \autoref{table:xxx}.

- Not recommended: In Figure~\ref{fig:overview} - Good: In \autoref{fig:overview}

Structure

  • main.tex is the entry root of the LaTeX project, containing the usepackage commands and the document structure.
  • src/ contains all the sections.
  • tables/ contains all the tables.
  • figures/ or fig/ contains the figures. Figures are PDF files or .tex files. .tex in figures are algorithms.
  • code/ contains code listings used in the text.

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

04

需要参考平台分布和安装热度时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenCode

27.33%
按下载量换算602

Claude Code

20.74%
按下载量换算457

windsurf

18.23%
按下载量换算401

cline

12.65%
按下载量换算279

Cursor

7.31%
按下载量换算161

Codex

2.95%
按下载量换算65

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。

来源信息

继续浏览同类 Skills