writing-assistant
Skill760 starsChanged 3 months agoMITView on GitHub
写作助手 - 当用户说"我想写XX"、"帮我梳理选题"、"怎么形成框架"、"给我组织思路"时触发。根据观点清晰度自动选择最优路线:清晰观点走"框架→内容",模糊观点走"挖掘→选题→框架→内容"。
9 other projects use a version of this.
Adopted by
Most recently taken up by
- abhichandra21/dotclaudetheir own version, first seen Nov 8, 2026
- ZhouYinLong-lab/Dog-Skillstheir own version, first seen Nov 8, 2026
- eddieoz/skillstheir own version, first seen Nov 1, 2026
- infranodus/skillstheir own version, first seen Oct 31, 2026
- borgr/paper-sharpenertheir own version, first seen Oct 17, 2026
- risukisu/marketing-skillstheir own version, first seen Sep 28, 2026
Other versions9 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 9
- infranodus/skills★ 126same name
- Agentchengfeng/writing-assistant-skill★ 114same name
- borgr/paper-sharpener★ 19same name
- ZhouYinLong-lab/Dog-Skills★ 7same name
- risukisu/marketing-skills★ 6same name
- xiaoYuan928/claude-code-skills★ 4same name
- eddieoz/skills★ 1same name
- abhichandra21/dotclaude★ 0same name
- Ivolga13/cursor-russian-toolkit★ 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 repository37 other files it gives its agents.
- article-studySkill
- auto-taskSkill
- backlog-managerSkill
- case-radarSkill
- design-explorationSkill
- dual-agent-collaborationSkill
- find-top-threeSkill
- github-repo-searchSkill
- git-pushSkill
- goal-setterSkill
- hermes-persona-builderSkill
- image-assistantSkill
- issue-poolSkill
- issue-triageSkill
- lesson-builderSkill
- logo-designSkill
- macos-product-designSkill
- memory-initSkill
- multi-perspective-analysisSkill
- organizeSkill
- page-solution-designSkill
- plan-reportSkill
- prd-auto-test-loopSkill
- prd-doc-writerSkill
- prd-test-writerSkill
- priority-judgeSkill
- product-namingSkill
- project-map-builderSkill
- readable-outputSkill
- req-change-workflowSkill
- system-studySkill
- thinking-partnerSkill
- thought-miningSkill
- ui-designSkill
- version-plannerSkill
- vision-explorationSkill
- weekly-reportSkill
The file
Files
写作助手
核心流程
三个关键步骤:选题 → 框架 → 内容
但根据你对选题的清晰度,会走不同的分支:
用户提出主题或想法
│
↓
[阶段00] 诊断:观点清不清楚?
│
├─→ 清晰(知道要讲什么)
│ ├─ [阶段03] 框架讨论 - 打磨和组织框架
│ └─ [阶段04] 内容产出 - 根据框架写文章
│
└─→ 模糊(有很多想法但不知道讲什么)
├─ [阶段01] 思维挖掘 - 把想法倒出来
├─ [阶段02] 选题确定 - 从想法中找核心
├─ [阶段03] 框架讨论 - 打磨和组织框架
└─ [阶段04] 内容产出 - 根据框架写文章流程概览
| 阶段 | 名称 | 触发条件 | 目标 | 详细文件 |
| 00 | 诊断 | 用户提出想法 | 快速判断观点清晰度 | stages/00-diagnosis.md |
| 01 | 思维挖掘 | 观点模糊 | 把零散想法倒出来,记录成洞察 | stages/01-mining.md |
| 02 | 选题确定 | 洞察足够 | 从洞察中锁定核心选题和灵魂句 | stages/02-topic.md |
| 03 | 框架讨论 | 选题确定(无论哪个分支) | 打磨和组织文章框架,确保逻辑清晰 | stages/03-framework.md |
| 04 | 内容产出 | 框架确定 | 根据框架写成1000字左右的文章 | stages/04-writing.md |
调度规则
当前阶段如何判断:
- 进入阶段00(诊断) — 用户刚开始,说出了想法或选题
- 进入阶段01(思维挖掘) — 诊断判断:观点还不够清晰,有很多想法但不知道讲什么
- 进入阶段02(选题确定) — 洞察收集足够,需要从中提炼出核心选题
- 进入阶段03(框架讨论) — 选题已经清晰,需要打磨框架结构
- 进入阶段04(内容产出) — 框架已经确定,准备写文章
每个阶段开始时:
- 告诉用户当前在哪个阶段
- 读取对应的阶段文件,按照里面的步骤执行
- 这个阶段的目标是什么、会做什么事
文件结构
writing-assistant/
├── SKILL.md # 主文件(触发、流程、调度规则)
├── stages/
│ ├── 00-diagnosis.md # 诊断阶段
│ ├── 01-mining.md # 思维挖掘(仅当观点模糊时)
│ ├── 02-topic.md # 选题确定(仅当观点模糊时)
│ ├── 03-framework.md # 框架讨论(通用)
│ └── 04-writing.md # 内容产出(通用)
└── templates/
├── framework-template.md # 框架讨论的记录模板
└── article-template.md # 内容产出时参考核心原则
- 不浪费时间:观点清晰就不挖掘,直接框架
- 保证质量:框架讨论是必须的,确保逻辑和表达
- 模块化复用:框架和内容两个模块通用,无论哪个分支都会用到
- 用户掌控:用户随时可以说"继续"或"停止",进度由用户控制
注意事项
- 阶段00的诊断要快速,3-5个问题就能判断清晰度
- 思维挖掘不要急,让用户尽量倒干净想法
- 框架讨论时,重点是打磨"读者为什么要读"和"逻辑顺序"
- 内容产出时,保持用户的原话风格和口吻
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.

