/SKILL.md
Use when writing academic papers, theses, or research articles - supports brainstorming, chapter writing, literature review, and LaTeX output
npx skillsauth add norman-bury/articlewriting-skill research-writing-assistantInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
面向本科与研究生论文写作的执行型 Skill。
本 Skill 已升级为模块化技能架构,支持多平台:
| 平台 | 配置 |
|------|------|
| Claude Code | .claude-plugin/plugin.json |
| Cursor | .cursor-plugin/plugin.json |
| Codex | .codex/INSTALL.md |
| OpenCode | .opencode/INSTALL.md |
| Gemini CLI | GEMINI.md |
| 技能 | 路径 | 说明 |
|------|------|------|
| using-research-writing | skills/using-research-writing/ | 入口技能,规则和路由 |
| paper-orchestration | skills/paper-orchestration/ | 中型/整篇任务编排 |
| brainstorming-research | skills/brainstorming-research/ | 头脑风暴,7轮问答 |
| writing-chapters | skills/writing-chapters/ | 章节写作 |
| latex-output | skills/latex-output/ | LaTeX 输出 |
| 技能 | 路径 | 说明 |
|------|------|------|
| writing-core | skills/writing-core/ | 核心写作规范 |
| writing-humanities | skills/writing-humanities/ | 文科写作 |
| writing-medical | skills/writing-medical/ | 医学写作 |
| writing-law | skills/writing-law/ | 法学写作 |
| literature-review | skills/literature-review/ | 文献综述 |
| evidence-driven-writing | skills/evidence-driven-writing/ | 文献驱动引言/相关工作 |
| experiment-results-planning | skills/experiment-results-planning/ | 实验、结果和 mock 数据规划 |
| 技能 | 路径 | 说明 |
|------|------|------|
| verification | skills/verification/ | 验证机制,确保完成声称有证据 |
| figures-python | skills/figures-python/ | Python 数据图表 |
| figures-diagram | skills/figures-diagram/ | 流程图/架构图 |
| peer-review | skills/peer-review/ | 自审检查 |
| statistical-analysis | skills/statistical-analysis/ | 统计分析 |
| prompts-collection | skills/prompts-collection/ | 提示词集合 |
| environment-setup | skills/environment-setup/ | 环境配置 |
| AI的想法 | 正确做法 | |----------|----------| | "用户说得很清楚了,直接开始写" | 必须先完成 brainstorming-research | | "这只是修改一小段" | 检查是否有 plan/,没有则先创建 | | "先写一段看看效果" | 必须先确认论文类型和章节结构 | | "用户很着急,跳过讨论" | 流程可以加速,但不能跳过关键确认 | | "这是简单任务,不需要 plan" | 任何写作任务都需要 plan 记录 | | "我知道怎么写论文" | 必须按用户选择的类型和结构写 | | "先把内容写完再说格式" | 格式在 brainstorming 阶段确定 | | "这章内容很简单,不用确认" | 每章写完都必须让用户确认 |
| AI的想法 | 正确做法 | |----------|----------| | "文献我可以补充一些" | 绝不编造文献,必须可追溯 | | "我记得这个技能的内容" | 技能会更新,必须重新读取当前版本 | | "这个引用看起来很合理" | 没有来源的引用一律不写 | | "用户应该知道这个领域" | 不假设用户知识,问清楚再写 |
| AI的想法 | 正确做法 | |----------|----------| | "应该写完了" | 运行验证命令确认 | | "章节看起来完整" | 执行字数统计和结构检查 | | "引用应该是真的" | 调用 CrossRef API 或搜索验证 | | "格式应该没问题" | 运行格式检查脚本 | | "搜索完成" | 检查结果数量和 DOI 列表 | | "我很确信" | 确信 ≠ 证据,运行验证 | | "就这一次跳过验证" | 没有例外 |
<EXTREMELY-IMPORTANT> 任何论文写作任务开始前,必须先调用 `skills/using-research-writing/` 确定流程。 不允许跳过头脑风暴直接写作。 </EXTREMELY-IMPORTANT>入口路由 → 调用 using-research-writing
paper-orchestration中型任务编排 → 调用 paper-orchestration
头脑风暴 → 调用 brainstorming-research
章节写作 → 调用 writing-chapters
LaTeX 输出(可选)→ 调用 latex-output
两阶段 Review(每章完成后)
每章写作完成后,执行两阶段检查:
阶段一:规范合规检查
阶段二:质量检查
原有 modules/ 目录仍然保留,可通过以下方式使用:
| 场景 | 模块 |
|------|------|
| 通用写作 | modules/writing-core.md |
| 文科写作 | modules/writing-humanities.md |
| 医学写作 | modules/writing-medical.md |
| 法学写作 | modules/writing-law.md |
| 文献综述 | modules/literature-review.md |
| 文献驱动写作 | skills/evidence-driven-writing/ |
| 论文工作流编排 | skills/paper-orchestration/ |
| 实验结果规划 | skills/experiment-results-planning/ |
| 翻译润色 | modules/prompts-collection.md |
| 自审检查 | modules/peer-review.md |
| Python图表 | modules/figures-python.md |
| 流程图 | modules/figures-diagram.md |
| 环境配置 | modules/environment-setup.md |
| LaTeX指南 | modules/latex-guide.md |
development
Use when writing or revising academic papers, especially Chinese journal manuscripts, that need natural prose, de-AI-ification, Markdown formatting, or quality checks
data-ai
Use for translation, polishing, or de-AI-ification of academic text - provides ready-to-use prompt templates
testing
Use when a research-writing task spans multiple sections, medium-sized revisions, full-paper drafting, or repeated quality failures
data-ai
Use when designing experiments, result tables, mock planning data, evaluation protocols, or results sections before real data are final