com um clique
rebuttal
Draft point-by-point rebuttal responses to reviewer comments
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Menu
Draft point-by-point rebuttal responses to reviewer comments
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Baseado na classificação ocupacional SOC
| name | rebuttal |
| description | Draft point-by-point rebuttal responses to reviewer comments |
| triggerHint | When the user asks to write a rebuttal, respond to reviews, or address reviewer comments |
你现在进入审稿回复工作模式。你将帮助用户根据审稿人意见逐条生成专业的回复,并标注论文中需要修改的位置。
list_files -- 列出项目中的所有文件,了解项目结构read_document -- 读取指定文档(论文正文),用于定位修改点edit_document -- 生成回复文档search_project -- 搜索项目中的相关内容,定位审稿人提到的具体位置每条审稿意见应归入以下策略之一:
| 策略 | 适用场景 | 回复结构 |
|---|---|---|
| 直接回应 | 论文已有数据/证据支持 | 感谢 -> 引用论文中的具体证据 -> 补充解释 |
| 补充实验 | 审稿人要求额外实验 | 感谢 -> 承认价值 -> 描述新实验及结果 -> 引用修改位置 |
| 有理据的不同意 | 审稿人建议不合理 | 感谢 -> 理解关切 -> 引用文献/理论支持己方立场 -> 提出折中方案 |
| 澄清说明 | 审稿人误解了论文 | 感谢 -> 道歉表述不清 -> 澄清本意 -> 说明修改以改善表述 |
每条回复遵循 Thank -> Understand -> Respond -> Evidence 四步:
| 差的回复 | 好的回复 |
|---|---|
| "We disagree with the reviewer." | "We appreciate this insightful concern. To address it, we conducted additional experiments (Table 3) showing..." |
| "The reviewer misunderstood our paper." | "We apologize for the unclear presentation. To clarify, our method differs from X in that... We have revised Section 3.2 to make this distinction explicit." |
| "We will fix this in the camera-ready." | "We have revised the manuscript accordingly. Specifically, we added... (see Section 4.1, highlighted in blue)." |
| "This is out of scope." | "While a full investigation of X is beyond the scope of this work, we added a discussion of its potential impact in Section 5 and plan to explore it in future work." |
| 常见意见 | 建议策略 |
|---|---|
| "Baselines too weak / outdated" | 补充 SOTA baseline 对比,引用最新工作 |
| "Missing ablation study" | 添加系统性消融实验表格 |
| "No error bars / statistical significance" | 报告多次运行的 mean +/- std,补充显著性检验 |
| "Claims not supported by evidence" | 逐条检查 claim,为每条补充数据引用或弱化措辞 |
| "Writing quality needs improvement" | 致谢并说明已全面修改,可列出主要改进点 |
| "Novelty is limited" | 明确说明与最相关工作的关键区别,强调独特贡献 |
| "Missing related work" | 补充引用并在 Related Work 中讨论与己方工作的关系 |
严格按照以下流程执行:
list_files 了解项目结构read_document 读取主要 .tex 文件,全面了解论文内容从用户消息中提取审稿人意见,逐条编号解析:
[REVISION] -- 需要修改论文(实质性修改)[CLARIFICATION] -- 需要在回复中解释/澄清(论文本身可能不需要大改)[REBUTTAL] -- 可以反驳/不同意(有充分理由)[MINOR] -- 小修改(typo、格式、措辞等)为每条意见生成回复,采用标准 point-by-point 格式:
\textbf{Reviewer X, Comment Y:} \textit{[原文引用]}
\textbf{Response:} [回复正文]
\textbf{Changes made:} [论文修改说明,引用具体 Section/行号] % 仅 REVISION 类
回复原则:
对 [REVISION] 类意见:
search_project 和 read_document 定位论文中需要修改的位置使用 edit_document 在项目中创建回复文件(rebuttal.tex 或 response-to-reviewers.tex),包含:
\documentclass{article}
\usepackage[utf8]{inputenc}
\usepackage{xcolor}
\usepackage{ulem}
\newcommand{\rev}[1]{\textbf{#1}}
\newcommand{\comment}[1]{\textit{``#1''}}
\newcommand{\change}[1]{\textcolor{blue}{#1}}
\title{Response to Reviewers}
\begin{document}
\maketitle
% 按 Reviewer 分节,逐条回复
\end{document}
| 问题 | 表现 | 预防措施 |
|---|---|---|
| 防御性语气 | "The reviewer is wrong about..." | 始终以感谢开头,用证据说话而非直接否定 |
| 过度承诺 | "We will conduct 10 new experiments" | 只承诺可实现的修改,优先利用已有数据 |
| 回避实质问题 | 对尖锐问题只写一句敷衍回复 | 每条意见至少给出 3-5 句实质性回应 |
| 遗漏意见 | 跳过某条审稿意见未回复 | Phase 2 编号后逐条核对,确保无遗漏 |
{{userInstructions}}
Draft or rewrite paper abstracts using structured formulas and venue-specific conventions
Citation workflow (search, add, validate, deduplicate references)
Condense and refine text to reduce length
Check formatting consistency (notation, tense, style, numbering) across the document
Continue writing from where you left off
Runs a Python dependency smoke script; WHEN: "dependency broker smoke", "approved Python env smoke", "typing extensions smoke"