Skip to content
Back to search

anti-defensive-writing-Skill

极致轻量|阻止论文的防御性写作 | Anti-defensive writing for academic papers — 学术写作原则 Skill + prompt pack (CN/EN)

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today

Check it deeper on your machine

It has changed since its scan on 11 Nov 2026.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/Adkid-Zephyr/anti-defensive-writing-Skill with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/Adkid-Zephyr/anti-defensive-writing-Skill'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/adkid-zephyr--anti-defensive-writing-skill/badge.svg)](/listings/adkid-zephyr--anti-defensive-writing-skill)

HTML

<a href="/listings/adkid-zephyr--anti-defensive-writing-skill"><img src="/api/listings/adkid-zephyr--anti-defensive-writing-skill/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

plugin.json · Manifest · 52 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. {
  2. "$schema": "https://agent-plugins.org/schemas/1.0.0/plugin.schema.json",
  3. "name": "anti-defensive-writing",
  4. "version": "1.0.0",
  5. "description": "Chinese and English academic-writing skills based on the Press-Release Principle.",
  6. "author": {
  7. "name": "Adkid-Zephyr",
  8. "url": "https://github.com/Adkid-Zephyr"
  9. },
  10. "homepage": "https://github.com/Adkid-Zephyr/anti-defensive-writing-Skill",
  11. "repository": "https://github.com/Adkid-Zephyr/anti-defensive-writing-Skill",
  12. "license": "MIT",
  13. "keywords": [
  14. "academic-writing",
  15. "paper-writing",
  16. "anti-defensive-writing",
  17. "chinese",
  18. "english"
  19. ],
  20. "extensions": {
  21. "com.openai": {
  22. "interface": {
  23. "displayName": "Anti-Defensive Writing",
  24. "shortDescription": "Academic writing, claim first",
  25. "longDescription": "Organize academic papers around their strongest supported contribution. Revise abstracts, introductions, conclusions and rebuttals, reduce unnecessary defensive language, and give each experiment an argumentative role. Includes Chinese and English skills based on the Press-Release Principle. Users supply the text and evidence; the plugin does not run experiments or validate scientific claims. It contains instructions only, with no MCP server, login, telemetry or external data transfer implemented by the publisher.",
  26. "developerName": "Adkid-Zephyr",
  27. "category": "Productivity",
  28. "capabilities": [
  29. "Academic writing",
  30. "Paper restructuring",
  31. "Chinese and English editing"
  32. ],
  33. "websiteURL": "https://github.com/Adkid-Zephyr/anti-defensive-writing-Skill",
  34. "supportURL": "https://github.com/Adkid-Zephyr/anti-defensive-writing-Skill/issues",
  35. "privacyPolicyURL": "https://github.com/Adkid-Zephyr/anti-defensive-writing-Skill/blob/main/PRIVACY.md",
  36. "termsOfServiceURL": "https://github.com/Adkid-Zephyr/anti-defensive-writing-Skill/blob/main/LICENSE",
  37. "defaultPrompt": [
  38. "帮我重写这段摘要,围绕最强的证据组织贡献,保留准确的数字与研究范围。",
  39. "Revise this introduction around its strongest supported contribution. Preserve facts and essential qualifications.",
  40. "帮我检查这段 rebuttal,直接回答审稿人的问题,并指出哪些表述缺少证据。"
  41. ],
  42. "brandColor": "#234B8C",
  43. "brandColorDark": "#A8C7FA",
  44. "composerIcon": "./assets/icon.svg",
  45. "logo": "./assets/icon.svg"
  46. },
  47. "publication": {
  48. "release_notes": "Initial Codex plugin packaging of the existing Chinese and English Anti-Defensive Writing skills."
  49. }
  50. }
  51. }
  52. }
README.md · README · 102 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. # Anti-Defensive Writing · 学术写作原则
  2. **阻止论文的防御性写作 —— 一个轻量级、零依赖的 AI 写作助手 Skill + 提示词,复制即用**
  3. [English README](README_EN.md) · [中文 Skill](skills/anti-defensive-writing/SKILL.md) · [English Skill](skills/anti-defensive-writing-en/SKILL.md)
  4. [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
  5. [![PRs Welcome](https://img.shields.io/badge/PRs-welcome-brightgreen.svg)](https://github.com/Adkid-Zephyr/anti-defensive-writing-Skill/pulls)
  6. ---
  7. ### Codex 插件安装
  8. 作者:**Adkid-Zephyr**。本仓库同时提供中文和英文 Skill。Codex 可以安装 Skill,也可以安装打包后的插件。
  9. 安装本仓库插件:
  10. ```bash
  11. codex plugin marketplace add Adkid-Zephyr/anti-defensive-writing-Skill
  12. codex plugin add anti-defensive-writing@adkid-zephyr-writing
  13. ```
  14. 安装后在新对话中选择 Anti-Defensive Writing 插件,并使用中文或英文 Skill 修改论文。
  15. 也可以把 [中文 Skill 链接](skills/anti-defensive-writing/SKILL.md) 交给 Codex,要求它安装这个 Skill。
  16. 插件发布包包含相同的两份 Skill,没有 MCP 服务或额外运行依赖。通过 `python3 scripts/build_plugin.py` 生成可上传到 OpenAI 发布后台的 ZIP;[隐私政策](PRIVACY.md)。官方目录发布状态以 OpenAI 后台为准,GitHub release 不是官方目录上架记录。
  17. ## 这是什么
  18. 很多人用 AI 改论文,改完越看越心虚:审稿人还没开口,自己先把「遗憾的是」「效果有限」「仍明显落后」写满了。结构是工作汇报式的,语言是自我审查式的——**等于提前把刀子递到审稿人手里,还附赠一份使用说明。**
  19. 这叫**防御性写作(Defensive Writing)**。
  20. 解药只有一条原则:
  21. > **论文是一场学术发布会,不是项目总结。**
  22. > 发布会只讲一件事:你最强的那个优势。
  23. 打不赢的指标不设为比赛,不占优的结果不说输,实验不是结果仓库、是论证工具。论文的说服力来自主张与证据高度一致,而不是比较项目最多。
  24. ## 效果对比
  25. | | 防御性写作 | 发布会原则 |
  26. |---|---|---|
  27. | 劣势表述 | 虽然本方法在两个同领域数据集上提高了准确率,但必须承认,我们尚未测试跨领域泛化,因此应谨慎理解这些结果。 | 本方法在两个同领域数据集上提高了准确率;跨领域泛化尚未评估。 |
  28. | 结构 | 我们首先尝试了 A,然后尝试了 B,最后选择了 C | 问题 X 很关键,现有方法缺 Y,本文提出 Z,证据如下 |
  29. | 实验 | 附录式堆结果,全覆盖无重点 | 每个实验承担一个论证职责 |
  30. | 结论 | 最后一段突然自我否定 | 只强化记忆点:解决了什么、证明了什么、为什么重要 |
  31. ## 快速开始
  32. ### 方式一:复制提示词(任何 AI 都能用)
  33. 直接复制 [`prompts/精简版提示词.txt`](prompts/精简版提示词.txt)([English](prompts/quick-prompt-en.txt)),粘贴到你和 AI 的对话开头,然后发给它你的论文段落。复制即用,零门槛。
  34. ### 方式二:安装 Skill(Claude Code / Cursor / 其他 Agent 工具)
  35. 把对应语言的 skill 目录整个拷进你的 skills 目录即可:
  36. ```bash
  37. # 中文版
  38. cp -r skills/anti-defensive-writing ~/.claude/skills/
  39. # English version
  40. cp -r skills/anti-defensive-writing-en ~/.claude/skills/
  41. ```
  42. 之后对 AI 说「帮我改改这段论文」「润色一下 abstract」,它会自动按发布会原则工作。
  43. ## 十二条规则速览
  44. - **叙事**:只围绕优势组织 · 不写工作汇报 · 打不过的维度不设为比赛 · 优势必须明说 · 控制比较范围 · 允许彻底重构故事
  45. - **语言**:禁用自我削弱表达 · 不占优的结果不说输 · 不把局部现象上升为对整体方法的否定
  46. - **实验**:每个实验必须有论证职责
  47. - **结构**:摘要引言 = 发布会开场 · 结论只强化记忆点
  48. 完整规则与决策流程见 [SKILL.md](skills/anti-defensive-writing/SKILL.md)。
  49. ## 仓库结构
  50. ```
  51. anti-defensive-writing/
  52. ├── skills/
  53. │ ├── anti-defensive-writing/ # 中文 Skill
  54. │ └── anti-defensive-writing-en/ # English Skill
  55. ├── prompts/
  56. │ ├── 精简版提示词.txt # 中文精简提示词(复制即用)
  57. │ └── quick-prompt-en.txt # English quick prompt
  58. └── README.md / README_EN.md
  59. ```
  60. ## 适用场景
  61. - 写/改论文摘要、引言、结论
  62. - 压缩论文篇幅(砍哪段,这里给了决策优先级)
  63. - 组织实验章节(每个实验该承担什么职责)
  64. - 写 rebuttal 前的自查(不给审稿人递刀子清单)
  65. ## License
  66. MIT。欢迎 PR,欢迎转发给你的同门。

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.