Writing Assistant
- 1 installs
- 404 repo stars
- Updated August 5, 2026
- aiskillstore/marketplace
writing-assistant is a skill that provides an interactive workflow for self-media content, guiding topic selection, outlining, drafting, titling, and formatting through to publishing.
About
writing-assistant is an interactive self-media writing helper that guides the full flow from topic selection to publishing. A content creator uses it to start a new article, score a draft against a checklist, or diagnose why a published piece underperformed. It routes through method files for topic, outline, body, title, and formatting and enforces publish thresholds. The documentation is written in Chinese.
- Interactive self-media writing assistant from topic to publish
- Routes scenarios: new article, draft check, low-performance diagnosis
- Scores drafts on a checklist and gates publishing at 80%
Writing Assistant by the numbers
- 1 all-time installs (skills.sh)
- Ranked #1,710 of 1,879 Marketing & SEO skills by installs in the Skillselion catalog
- Data as of Aug 5, 2026 (Skillselion catalog sync)
writing-assistant capabilities & compatibility
- Capabilities
- copywriting · seo · marketing
- Use cases
- copywriting · marketing · seo
What writing-assistant says it does
自媒体写作交互式助手。帮助完成从选题到发布的完整流程。
≥80%发布，60-79%优化，<60%重做
写新文章 | 读 `user-personas.md` 确定≥3类人群 → 按 methods/01→05 执行
npx skills add https://github.com/aiskillstore/marketplace --skill writing-assistantAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 1 |
|---|---|
| repo stars | ★ 404 |
| Last updated | August 5, 2026 |
| Repository | aiskillstore/marketplace ↗ |
What it does
Guide self-media content creation from topic selection through drafting, titling, and publishing.
Who is it for?
Guiding self-media article creation from topic to publish with a scoring checklist
Skip if: English-language workflows, since the guidance is written in Chinese
When should I use this skill?
Starting a new article, checking a draft, or diagnosing poor post performance
What you get
A scored, publish-ready article produced through a staged topic-to-format workflow
- Scored draft and publish-ready article
By the numbers
- Publish threshold at 80% on the checklist
- 5 method files: topic, outline, body, title, formatting
Files
写作助手
⚠️ 每次使用前必读
1. 读 about-me.md — 了解定位、禁区、术语降维 2. 读 user-personas.md — 了解3类用户、4层需求
---
文件索引
| 文件 | 内容 |
|---|---|
about-me.md | 定位、双品牌、5个主题池、禁区、术语降维 |
user-personas.md | 3类目标用户、4层需求、触达策略 |
content/methods/01-05 | 选题→大纲→正文→标题→排版 |
---
场景路由
| 场景 | 操作 |
|---|---|
| 写新文章 | 读 user-personas.md 确定≥3类人群 → 按 methods/01→05 执行 |
| 检查草稿 | 用检查清单打分，≥80分发布 |
| 诊断数据差 | 检查清单排查 → 读对应 methods/ |
| 取标题 | 读 methods/04-标题.md |
| 排版 | 读 methods/05-排版.md |
| 了解定位/禁区 | 读 about-me.md |
---
检查清单
≥80%发布，60-79%优化，<60%重做
选题 methods/01-选题.md
- [ ] 三维匹配：人×情境×信息定义清晰
- [ ] 市场洼地：搜索<10条
- [ ] 信息增量≥70%
- [ ] 覆盖≥3类人群（参考
user-personas.md）
大纲 methods/02-大纲.md
- [ ] 有"为什么"章节+3个理由
- [ ] 3-5个场景，≥3类人群
- [ ] 承诺：具体+可信+有吸引力
- [ ] ≤15章，40-60行/章
正文 methods/03-正文.md
- [ ] 抽象概念有图，操作有截图
- [ ] ≤5-7行/段
- [ ] 首段无术语（参考
about-me.md术语降维表） - [ ] 开头3秒：痛点/对比/反常识
标题 methods/04-标题.md
- [ ] 数字对比，倍数≥10倍
- [ ] 有戏剧性词汇
- [ ] 你妈妈能理解吗？
排版 methods/05-排版.md
- [ ] 只用H1/H2/H3
- [ ] 有序列表无空格（
1.项目） - [ ] 用
---换行 - [ ] 流程/步骤用 ASCII 图可视化
---
一票否决
- 无"为什么"章节
- <3个理由
- <3类人群
- 违反
about-me.md禁区
关于我
定位
- 领域：[你的领域，如：AI编程、健身、理财等]
- 目标：[你的目标，如：成为XX领域头部创作者]
- 受众：[你的受众，如：小白用户、开发者、职场人等]
双品牌（可选）
| 品牌 | 受众 | 形式 | 内容 |
|---|---|---|---|
| [品牌A] | 小白 | 视频 | 3-5分钟快速教程 |
| [品牌B] | 深度用户 | 图文 | 3000-5000字长文 |
主题池
1. [主题1] 2. [主题2] 3. [主题3] 4. [主题4] 5. [主题5]
禁区
- ❌ [不做的内容1]
- ❌ 高级内容为主（应80%基础+20%进阶）
- ❌ 跳过认知阶段
- ❌ 系列重复信息>30%
- ❌ 无先验调研的选题
- ❌ 单一人群内容（需≥3类）
术语降维
| 术语 | 降维表达 |
|---|---|
| [专业术语1] | [小白能懂的说法] |
| [专业术语2] | [小白能懂的说法] |
| [专业术语3] | [小白能懂的说法] |
01-选题
核心原则
>
1. 三维匹配：价值 = 人×情境×信息
2. 先验调研：不做调研=闭门造车
3. 市场洼地：搜索<10条=早期，>100条=晚期
4. 信息增量≥70%：重复内容=浪费时间
---
三维匹配法
价值 = 人 (Who) × 情境 (Context) × 信息 (What)
任何一维为0，价值就为0维度1：人 (Who)
人 = 认知结构 + 技能基础 + 资源约束
- 认知结构：知道什么，不知道什么
- 技能基础：会做什么，不会做什么
- 资源约束：时间/金钱/注意力维度2：情境 (Context)
情境 = 触发事件 + 当前状态 + 期望状态 + 行动阻碍
- 触发事件：什么事让他需要这个信息
- 当前状态：他现在在哪里
- 期望状态：他想到哪里
- 行动阻碍：什么阻止他到达维度3：信息 (What)
信息 = 信息增量 + 认知门槛 + 行动成本
- 信息增量：缩短"当前→期望"的距离
- 认知门槛：理解需要多少背景知识
- 行动成本：应用需要多少资源---
先验调研
必做步骤：
1. 搜索：小红书/知乎/微信搜一搜
2. 数量判断：
- <10条 = 早期市场（金矿）
- 10-100条 = 中期市场（可做）
- >100条 = 晚期市场（换主题）
3. 质量判断：现有内容好不好？有没有空间？---
市场洼地判断
搜索结果数量：
├─ <10条 → 金矿，快速介入
├─ 10-100条 → 中期，需要差异化
└─ >100条 → 晚期，建议换主题---
信息增量检查
□ 新信息占比 ≥ 70%？
□ 与现有内容有差异化？
□ 提供了新视角/新方法？---
检查清单
□ 三维匹配定义清晰？
□ 做了先验调研？
□ 市场洼地（搜索<10条）？
□ 信息增量≥70%？
□ 覆盖≥3类人群？02-大纲
核心原则
>
1. 销售漏斗4阶段：认知→兴趣→决策→行动
2. 承诺设计：承诺 = 实际价值 × 0.8
3. 必须有"为什么"章节：≥3个理由
4. 章节密度：40-60行/章最佳
---
销售漏斗4阶段
认知 → 兴趣 → 决策 → 行动
│ │ │ │
↓ ↓ ↓ ↓
为什么 适合我 怎么做 下一步
要用？ 吗？ 学什么？阶段1：认知（必须≥3个理由）
用户心理：为什么要用？
回答方式：
- 理由1：[解决什么问题]
- 理由2：[带来什么好处]
- 理由3：[为什么现在要用]
⚠️ 这是最常被跳过的阶段！阶段2：兴趣
用户心理：适合我吗？
回答方式：
- 场景1："你是不是遇到过..."
- 场景2："当你想要..."
- 场景3："如果你正在..."
目标：3-5个场景，覆盖≥3类人群阶段3：决策
用户心理：难学吗？
回答方式：
- 详细步骤
- 截图演示
- 强调"简单"、"只需X步"阶段4：行动
用户心理：接下来学什么？
回答方式：
- 80%基础内容
- 20%进阶内容
- 标注：必学/可选---
承诺设计
承诺 = 实际价值 × 0.8
原则：稍微低承诺，超额交付承诺强度
| 承诺类型 | 强度 | 示例 |
|---|---|---|
| 强承诺 | 高 | "3分钟学会XX" |
| 中承诺 | 中 | "XX完整指南" |
| 弱承诺 | 低 | "聊聊XX" |
---
章节密度
最佳密度：40-60行/章
<40行/章：内容太稀疏
>60行/章：信息过载---
检查清单
□ 有"为什么"章节？
□ 认知阶段≥3个理由？
□ 3-5个场景？
□ 覆盖≥3类人群？
□ 承诺：具体+可信+有吸引力？
□ ≤15章？
□ 40-60行/章？03-正文
核心原则
>
1. 视觉优先：图像处理50ms，文字200ms
2. 块化阅读：段落≤5-7行
3. 术语降维：首段无术语或有解释
4. 开头3秒：痛点/对比/反常识
---
视觉优先
处理速度：
图像 50ms >> 文字 200ms
结论：抽象概念必须有图执行标准：
□ 抽象概念有图？
□ 操作步骤有截图？
□ 流程有流程图？---
块化阅读
1块 = 1个核心观点
1块 ≤ 5-7行
太长 → 拆分
太短 → 合并段落结构：
观点（1行）
↓
解释（2-3行）
↓
例子/证据（2-3行）---
术语降维
原则：首段无术语，或有即时解释
降维模板：
[术语]，就是[小白能懂的说法]
示例：
- API → "程序之间的对话接口"
- Token → "AI的阅读成本"检查：
□ 首段无术语？
□ 必须用术语时有解释？
□ 用类比让小白理解？---
开头3秒
三种开头方式：
方式1：痛点前置
"你是不是遇到过这个问题：[具体痛点]"方式2：对比冲击
"以前需要[旧方法]，现在只需要[新方法]"方式3：反常识
"大多数人以为[错误认知]，其实[正确认知]"---
检查清单
□ 抽象概念有图？
□ 操作有截图？
□ ≤5-7行/段？
□ 首段无术语？
□ 开头3秒：痛点/对比/反常识？04-标题
核心原则
>
1. 数字冲击：倍数≥10倍
2. 戏剧性词汇：激发好奇心
3. 门槛降低：小白能理解
---
核心公式
标题吸引力 = 数字对比 × 戏剧性 × 低门槛三要素缺一不可：
- 只有数字 → 缺乏情绪
- 只有戏剧性 → 像标题党
- 门槛太高 → 看不懂
---
数字对比
大脑处理速度：
数字 <100ms
术语 500-1000ms
描述 1000-2000ms
结论：数字 = 最快触达倍数要求：
<10倍：冲击力不足
≥10倍：数量级差异，足够震撼示例：
❌ "效率提升2倍"
✅ "效率提升10倍"
✅ "从3小时→15分钟"---
戏剧性词汇
高吸引力词汇：
- 秘密、真相、颠覆
- 终极、完整、一文搞定
- 踩坑、避坑、血泪教训示例：
❌ "XX使用教程"
✅ "XX终极指南：3分钟上手"---
门槛降低
原则：你妈妈能理解吗？
方法：
1. 避免直接用术语
2. 用结果代替过程
3. 用场景代替概念示例：
❌ "MCP Server配置教程"
✅ "让AI帮你自动打车"---
标题模板
模板1：数字对比型
从[旧数字]到[新数字]：[主题]模板2：时间承诺型
[X分钟]学会[主题]模板3：问题解决型
[痛点]？[方案]帮你搞定---
检查清单
□ 有数字对比？
□ 倍数≥10倍？
□ 有戏剧性词汇？
□ 你妈妈能理解吗？05-排版
核心原则
>
1. 三级标题：只用H1/H2/H3
2. 公众号适配：有序列表无空格
3. 换行标记：用 --- 换行4. 流程可视化：用 ASCII 图
---
三级标题系统
只用3级标题：H1 / H2 / H3
四级以上 → 层级混乱标题规范：
| 标题 | 用途 |
|---|---|
| H1 | 主要章节 |
| H2 | 子章节 |
| H3 | 小节 |
---
公众号适配
有序列表格式
✅ 正确：
1.项目一（数字后直接接内容）
2.项目二
❌ 错误：
1. 项目一（数字后有空格）
2. 项目二换行标记
用 `---` 产生大间距
段落1
---
段落2---
ASCII 可视化
流程/步骤必须用 ASCII 图：
步骤1 → 步骤2 → 步骤3复杂流程：
┌─────────┐ ┌─────────┐
│ 输入 │ → │ 输出 │
└─────────┘ └─────────┘循环流程：
┌──────────────┐
│ │
↓ │
步骤1 → 步骤2 → 步骤3---
检查清单
□ 只用H1/H2/H3？
□ 有序列表无空格（`1.项目`）？
□ 用 `---` 换行？
□ 流程/步骤用 ASCII 图？Writing Assistant Skill
自媒体写作助手 Skill，帮助完成从选题到发布的完整流程。
安装
# 复制到 Claude Code skills 目录
cp -r writing-assistant ~/.claude/skills/文件结构
writing-assistant/
├── SKILL.md # 入口（场景路由+检查清单）
├── about-me.md # 关于我（定位+禁区+术语降维）⬅️ 需自定义
├── user-personas.md # 用户画像（3类用户+4层需求）⬅️ 需自定义
└── content/methods/
├── 01-选题.md # 三维匹配法
├── 02-大纲.md # 销售漏斗+承诺设计
├── 03-正文.md # 认知科学+可视化
├── 04-标题.md # 数字冲击力法
└── 05-排版.md # 公众号适配使用前必做
1. 编辑 `about-me.md` — 填入你的定位、禁区、术语降维表 2. 编辑 `user-personas.md` — 定义你的3类目标用户
使用流程
┌─────────────────┐ ┌─────────────────┐ ┌─────────────────┐
│ 1. 读基础文件 │ → │ 2. 按阶段执行 │ → │ 3. 检查打分 │
│ │ │ │ │ │
│ about-me.md │ │ methods/01→05 │ │ ≥80% = 发布 │
│ user-personas │ │ 选题→大纲→正文 │ │ 60-79% = 优化 │
└─────────────────┘ └─────────────────┘ └─────────────────┘核心方法论
| 阶段 | 方法论 | 公式 |
|---|---|---|
| 选题 | 三维匹配法 | 价值 = 人 × 情境 × 信息 |
| 大纲 | 承诺设计法 | 承诺 = 实际价值 × 0.8 |
| 正文 | 认知科学 | 图像(50ms) >> 文字(200ms) |
| 标题 | 数字冲击力 | 吸引力 = 数字对比 × 戏剧性 × 低门槛 |
场景示例
"帮我检查一下这篇文章" → 用检查清单打分
"帮我写一篇关于XX的文章" → 按 methods/01→05 执行
"这篇数据不好，帮我诊断" → 检查清单排查License
MIT
{
"schema_version": "2.0",
"meta": {
"generated_at": "2026-01-16T21:13:11.259Z",
"slug": "ceeon-writing-assistant",
"source_url": "https://github.com/Ceeon/writing-assistant-skill/tree/main/",
"source_ref": "main",
"model": "claude",
"analysis_version": "3.0.0",
"source_type": "community",
"content_hash": "42282bc0370c8ea741efcda08914b5cc240112467872d17c993fa614bddc9525",
"tree_hash": "e3b5aa5aaedbae72e52af8802b8e5a462e47639c2b90bf69b79da7469f4778e2"
},
"skill": {
"name": "writing-assistant",
"description": "自媒体写作交互式助手。帮助完成从选题到发布的完整流程。适用于：开始写新文章、检查草稿质量、诊断发布后数据不佳的原因。",
"summary": "自媒体写作交互式助手。帮助完成从选题到发布的完整流程。适用于：开始写新文章、检查草稿质量、诊断发布后数据不佳的原因。",
"icon": "✍️",
"version": "1.0.0",
"author": "Ceeon",
"license": "MIT",
"category": "writing",
"tags": [
"content-creation",
"self-media",
"article-writing",
"workflow"
],
"supported_tools": [
"claude",
"codex",
"claude-code"
],
"risk_factors": [
"external_commands",
"filesystem",
"network"
]
},
"security_audit": {
"risk_level": "safe",
"is_blocked": false,
"safe_to_publish": true,
"summary": "Pure markdown documentation skill with zero executable code. All static findings are false positives: markdown code blocks flagged as shell execution, Chinese documentation entropy flagged as obfuscation, and metadata URLs flagged as network calls. No security risks present.",
"risk_factor_evidence": [
{
"factor": "external_commands",
"evidence": [
{
"file": "content/methods/01-选题.md",
"line_start": 14,
"line_end": 18
},
{
"file": "content/methods/01-选题.md",
"line_start": 18,
"line_end": 22
},
{
"file": "content/methods/01-选题.md",
"line_start": 22,
"line_end": 28
},
{
"file": "content/methods/01-选题.md",
"line_start": 28,
"line_end": 32
},
{
"file": "content/methods/01-选题.md",
"line_start": 32,
"line_end": 39
},
{
"file": "content/methods/01-选题.md",
"line_start": 39,
"line_end": 43
},
{
"file": "content/methods/01-选题.md",
"line_start": 43,
"line_end": 49
},
{
"file": "content/methods/01-选题.md",
"line_start": 49,
"line_end": 57
},
{
"file": "content/methods/01-选题.md",
"line_start": 57,
"line_end": 64
},
{
"file": "content/methods/01-选题.md",
"line_start": 64,
"line_end": 70
},
{
"file": "content/methods/01-选题.md",
"line_start": 70,
"line_end": 75
},
{
"file": "content/methods/01-选题.md",
"line_start": 75,
"line_end": 81
},
{
"file": "content/methods/01-选题.md",
"line_start": 81,
"line_end": 85
},
{
"file": "content/methods/01-选题.md",
"line_start": 85,
"line_end": 91
},
{
"file": "content/methods/01-选题.md",
"line_start": 91,
"line_end": 97
},
{
"file": "content/methods/02-大纲.md",
"line_start": 14,
"line_end": 20
},
{
"file": "content/methods/02-大纲.md",
"line_start": 20,
"line_end": 24
},
{
"file": "content/methods/02-大纲.md",
"line_start": 24,
"line_end": 33
},
{
"file": "content/methods/02-大纲.md",
"line_start": 33,
"line_end": 37
},
{
"file": "content/methods/02-大纲.md",
"line_start": 37,
"line_end": 46
},
{
"file": "content/methods/02-大纲.md",
"line_start": 46,
"line_end": 50
},
{
"file": "content/methods/02-大纲.md",
"line_start": 50,
"line_end": 57
},
{
"file": "content/methods/02-大纲.md",
"line_start": 57,
"line_end": 61
},
{
"file": "content/methods/02-大纲.md",
"line_start": 61,
"line_end": 68
},
{
"file": "content/methods/02-大纲.md",
"line_start": 68,
"line_end": 74
},
{
"file": "content/methods/02-大纲.md",
"line_start": 74,
"line_end": 78
},
{
"file": "content/methods/02-大纲.md",
"line_start": 78,
"line_end": 92
},
{
"file": "content/methods/02-大纲.md",
"line_start": 92,
"line_end": 97
},
{
"file": "content/methods/02-大纲.md",
"line_start": 97,
"line_end": 103
},
{
"file": "content/methods/02-大纲.md",
"line_start": 103,
"line_end": 111
},
{
"file": "content/methods/03-正文.md",
"line_start": 14,
"line_end": 19
},
{
"file": "content/methods/03-正文.md",
"line_start": 19,
"line_end": 22
},
{
"file": "content/methods/03-正文.md",
"line_start": 22,
"line_end": 26
},
{
"file": "content/methods/03-正文.md",
"line_start": 26,
"line_end": 32
},
{
"file": "content/methods/03-正文.md",
"line_start": 32,
"line_end": 38
},
{
"file": "content/methods/03-正文.md",
"line_start": 38,
"line_end": 41
},
{
"file": "content/methods/03-正文.md",
"line_start": 41,
"line_end": 47
},
{
"file": "content/methods/03-正文.md",
"line_start": 47,
"line_end": 56
},
{
"file": "content/methods/03-正文.md",
"line_start": 56,
"line_end": 62
},
{
"file": "content/methods/03-正文.md",
"line_start": 62,
"line_end": 65
},
{
"file": "content/methods/03-正文.md",
"line_start": 65,
"line_end": 69
},
{
"file": "content/methods/03-正文.md",
"line_start": 69,
"line_end": 78
},
{
"file": "content/methods/03-正文.md",
"line_start": 78,
"line_end": 80
},
{
"file": "content/methods/03-正文.md",
"line_start": 80,
"line_end": 83
},
{
"file": "content/methods/03-正文.md",
"line_start": 83,
"line_end": 85
},
{
"file": "content/methods/03-正文.md",
"line_start": 85,
"line_end": 88
},
{
"file": "content/methods/03-正文.md",
"line_start": 88,
"line_end": 90
},
{
"file": "content/methods/03-正文.md",
"line_start": 90,
"line_end": 96
},
{
"file": "content/methods/03-正文.md",
"line_start": 96,
"line_end": 102
},
{
"file": "content/methods/04-标题.md",
"line_start": 13,
"line_end": 15
},
{
"file": "content/methods/04-标题.md",
"line_start": 15,
"line_end": 26
},
{
"file": "content/methods/04-标题.md",
"line_start": 26,
"line_end": 33
},
{
"file": "content/methods/04-标题.md",
"line_start": 33,
"line_end": 36
},
{
"file": "content/methods/04-标题.md",
"line_start": 36,
"line_end": 39
},
{
"file": "content/methods/04-标题.md",
"line_start": 39,
"line_end": 42
},
{
"file": "content/methods/04-标题.md",
"line_start": 42,
"line_end": 46
},
{
"file": "content/methods/04-标题.md",
"line_start": 46,
"line_end": 53
},
{
"file": "content/methods/04-标题.md",
"line_start": 53,
"line_end": 57
},
{
"file": "content/methods/04-标题.md",
"line_start": 57,
"line_end": 60
},
{
"file": "content/methods/04-标题.md",
"line_start": 60,
"line_end": 63
},
{
"file": "content/methods/04-标题.md",
"line_start": 63,
"line_end": 72
},
{
"file": "content/methods/04-标题.md",
"line_start": 72,
"line_end": 76
},
{
"file": "content/methods/04-标题.md",
"line_start": 76,
"line_end": 79
},
{
"file": "content/methods/04-标题.md",
"line_start": 79,
"line_end": 82
},
{
"file": "content/methods/04-标题.md",
"line_start": 82,
"line_end": 89
},
{
"file": "content/methods/04-标题.md",
"line_start": 89,
"line_end": 91
},
{
"file": "content/methods/04-标题.md",
"line_start": 91,
"line_end": 94
},
{
"file": "content/methods/04-标题.md",
"line_start": 94,
"line_end": 96
},
{
"file": "content/methods/04-标题.md",
"line_start": 96,
"line_end": 99
},
{
"file": "content/methods/04-标题.md",
"line_start": 99,
"line_end": 101
},
{
"file": "content/methods/04-标题.md",
"line_start": 101,
"line_end": 107
},
{
"file": "content/methods/04-标题.md",
"line_start": 107,
"line_end": 112
},
{
"file": "content/methods/05-排版.md",
"line_start": 7,
"line_end": 7
},
{
"file": "content/methods/05-排版.md",
"line_start": 14,
"line_end": 18
},
{
"file": "content/methods/05-排版.md",
"line_start": 18,
"line_end": 34
},
{
"file": "content/methods/05-排版.md",
"line_start": 34,
"line_end": 42
},
{
"file": "content/methods/05-排版.md",
"line_start": 42,
"line_end": 46
},
{
"file": "content/methods/05-排版.md",
"line_start": 46,
"line_end": 47
},
{
"file": "content/methods/05-排版.md",
"line_start": 47,
"line_end": 54
},
{
"file": "content/methods/05-排版.md",
"line_start": 54,
"line_end": 62
},
{
"file": "content/methods/05-排版.md",
"line_start": 62,
"line_end": 64
},
{
"file": "content/methods/05-排版.md",
"line_start": 64,
"line_end": 67
},
{
"file": "content/methods/05-排版.md",
"line_start": 67,
"line_end": 71
},
{
"file": "content/methods/05-排版.md",
"line_start": 71,
"line_end": 74
},
{
"file": "content/methods/05-排版.md",
"line_start": 74,
"line_end": 79
},
{
"file": "content/methods/05-排版.md",
"line_start": 79,
"line_end": 85
},
{
"file": "content/methods/05-排版.md",
"line_start": 85,
"line_end": 87
},
{
"file": "content/methods/05-排版.md",
"line_start": 87,
"line_end": 88
},
{
"file": "content/methods/05-排版.md",
"line_start": 88,
"line_end": 90
},
{
"file": "README.md",
"line_start": 7,
"line_end": 10
},
{
"file": "README.md",
"line_start": 10,
"line_end": 14
},
{
"file": "README.md",
"line_start": 14,
"line_end": 25
},
{
"file": "README.md",
"line_start": 25,
"line_end": 29
},
{
"file": "README.md",
"line_start": 29,
"line_end": 30
},
{
"file": "README.md",
"line_start": 30,
"line_end": 34
},
{
"file": "README.md",
"line_start": 34,
"line_end": 41
},
{
"file": "README.md",
"line_start": 41,
"line_end": 54
},
{
"file": "README.md",
"line_start": 54,
"line_end": 58
},
{
"file": "SKILL.md",
"line_start": 10,
"line_end": 10
},
{
"file": "SKILL.md",
"line_start": 11,
"line_end": 11
},
{
"file": "SKILL.md",
"line_start": 19,
"line_end": 19
},
{
"file": "SKILL.md",
"line_start": 20,
"line_end": 20
},
{
"file": "SKILL.md",
"line_start": 21,
"line_end": 21
},
{
"file": "SKILL.md",
"line_start": 29,
"line_end": 29
},
{
"file": "SKILL.md",
"line_start": 32,
"line_end": 32
},
{
"file": "SKILL.md",
"line_start": 33,
"line_end": 33
},
{
"file": "SKILL.md",
"line_start": 34,
"line_end": 34
},
{
"file": "SKILL.md",
"line_start": 42,
"line_end": 42
},
{
"file": "SKILL.md",
"line_start": 46,
"line_end": 46
},
{
"file": "SKILL.md",
"line_start": 48,
"line_end": 48
},
{
"file": "SKILL.md",
"line_start": 54,
"line_end": 54
},
{
"file": "SKILL.md",
"line_start": 57,
"line_end": 57
},
{
"file": "SKILL.md",
"line_start": 60,
"line_end": 60
},
{
"file": "SKILL.md",
"line_start": 65,
"line_end": 65
},
{
"file": "SKILL.md",
"line_start": 67,
"line_end": 67
},
{
"file": "SKILL.md",
"line_start": 68,
"line_end": 68
},
{
"file": "SKILL.md",
"line_start": 78,
"line_end": 78
}
]
},
{
"factor": "filesystem",
"evidence": [
{
"file": "README.md",
"line_start": 9,
"line_end": 9
},
{
"file": "README.md",
"line_start": 9,
"line_end": 9
}
]
},
{
"factor": "network",
"evidence": [
{
"file": "skill-report.json",
"line_start": 6,
"line_end": 6
}
]
}
],
"critical_findings": [],
"high_findings": [],
"medium_findings": [],
"low_findings": [],
"dangerous_patterns": [],
"files_scanned": 10,
"total_lines": 960,
"audit_model": "claude",
"audited_at": "2026-01-16T21:13:11.259Z"
},
"content": {
"user_title": "Structure your writing with proven workflow methodology",
"value_statement": "Content creators often struggle with inconsistent quality and poor reader engagement. This skill provides a complete framework from topic selection to final formatting using proven methodologies and quality checklists.",
"seo_keywords": [
"content creation",
"article writing",
"writing workflow",
"self-media assistant",
"content marketing",
"Claude",
"Codex",
"Claude Code",
"自媒体写作",
"writing templates"
],
"actual_capabilities": [
"Guide topic selection using three-dimensional matching method",
"Structure outlines using sales funnel methodology",
"Write body text with cognitive science principles",
"Create engaging titles with numeric impact formulas",
"Format content for platform-specific optimization",
"Evaluate drafts against quality checklists"
],
"limitations": [
"Requires manual input of user context and goals",
"Templates require customization for specific niches",
"No automatic content generation from keywords",
"Platform-specific formatting focuses on WeChat public accounts"
],
"use_cases": [
{
"target_user": "Content creators",
"title": "New article workflow",
"description": "Follow complete workflow from topic research to final formatting for new articles"
},
{
"target_user": "Editors",
"title": "Quality assessment",
"description": "Use checklists to evaluate draft quality and identify improvement areas"
},
{
"target_user": "Marketing teams",
"title": "Content diagnosis",
"description": "Diagnose why published content underperforms using systematic analysis"
}
],
"prompt_templates": [
{
"title": "New article",
"scenario": "Start writing process",
"prompt": "Help me write an article about [topic]. First, I will read user-personas.md to understand target audiences."
},
{
"title": "Draft review",
"scenario": "Evaluate draft quality",
"prompt": "Review my draft using the checklist in SKILL.md. Check each item and give me a score with improvement suggestions."
},
{
"title": "Title creation",
"scenario": "Generate headlines",
"prompt": "Create 5 title options for my article about [topic] using the title methodology in methods/04-标题.md."
},
{
"title": "Performance diagnosis",
"scenario": "Analyze poor metrics",
"prompt": "My article about [topic] has low engagement. Use the checklist in SKILL.md to help me identify what went wrong."
}
],
"output_examples": [
{
"input": "Help me write an article about using AI for productivity",
"output": [
"Topic Assessment: Research first - search for existing content",
"Three dimensions: Define WHO, CONTEXT, and INFORMATION for this topic",
"User personas: Target beginners, professionals, managers",
"Outline: Start with why (3+ reasons to use AI)",
"Title options: Include 10x comparisons and dramatic words"
]
},
{
"input": "Review my draft about remote work tips",
"output": [
"Opening hook: Check if first 3 seconds grab attention",
"Structure: Verify why section with 3+ reasons",
"Audience coverage: Minimum 3 user personas addressed",
"Title impact: Numeric comparison with 10x倍数",
"Formatting: H1/H2/H3 only, no extra spaces in lists"
]
},
{
"input": "My article about Python tutorials has low views",
"output": [
"Check title: Is there numeric comparison with 10x impact?",
"Verify opening: Does it start with pain point or counter-intuitive fact?",
"Assess audience: Are at least 3 user personas covered?",
"Review structure: Is there a why chapter with 3 reasons?",
"Examine formatting: Are lists properly formatted without spaces?"
]
}
],
"best_practices": [
"Always read user-personas.md first to understand target audiences",
"Use the checklist before publishing - aim for 80% or higher score",
"Apply term simplification to make content accessible to beginners"
],
"anti_patterns": [
"Skipping the topic research phase leads to redundant content",
"Using technical jargon without explanation alienates readers",
"Publishing without title testing reduces click-through rates"
],
"faq": [
{
"question": "What platforms does this skill support?",
"answer": "Focuses on WeChat public accounts but principles apply to any content platform like Zhihu, Xiaohongshu, and blogs."
},
{
"question": "How long does the complete workflow take?",
"answer": "Plan 1-2 hours for new articles including research. Draft review takes 15-30 minutes depending on length."
},
{
"question": "Can I use this with other writing tools?",
"answer": "Yes. This skill provides methodology and structure. Combine with any editor or AI writing tool for content generation."
},
{
"question": "Is my data safe when using this skill?",
"answer": "This skill contains only markdown templates. No data is stored, transmitted, or accessed beyond your conversation."
},
{
"question": "What if my draft scores below 60%?",
"answer": "Review the failing checklist items systematically. Common issues include missing why section, too few audience perspectives, and weak titles."
},
{
"question": "How is this different from general writing AI?",
"answer": "Provides specific methodology with proven frameworks. Unlike general AI, it enforces quality standards through checklists and structured workflows."
}
]
},
"file_structure": [
{
"name": "content",
"type": "dir",
"path": "content",
"children": [
{
"name": "methods",
"type": "dir",
"path": "content/methods",
"children": [
{
"name": "01-选题.md",
"type": "file",
"path": "content/methods/01-选题.md",
"lines": 98
},
{
"name": "02-大纲.md",
"type": "file",
"path": "content/methods/02-大纲.md",
"lines": 112
},
{
"name": "03-正文.md",
"type": "file",
"path": "content/methods/03-正文.md",
"lines": 103
},
{
"name": "04-标题.md",
"type": "file",
"path": "content/methods/04-标题.md",
"lines": 113
},
{
"name": "05-排版.md",
"type": "file",
"path": "content/methods/05-排版.md",
"lines": 91
}
]
}
]
},
{
"name": "about-me.md",
"type": "file",
"path": "about-me.md",
"lines": 35
},
{
"name": "README.md",
"type": "file",
"path": "README.md",
"lines": 63
},
{
"name": "SKILL.md",
"type": "file",
"path": "SKILL.md",
"lines": 79
},
{
"name": "user-personas.md",
"type": "file",
"path": "user-personas.md",
"lines": 35
}
]
}
用户画像
核心原则
- 必须≥3类人群
- 必须包含小白
- 从小白视角出发
---
3类目标用户
1. [用户类型1 - 小白]（必须包含）⭐
- [特征描述]
- [痛点描述]
- 触达：[触达策略]
2. [用户类型2]
- [特征描述]
- 触达：[触达策略]
3. [用户类型3]
- [特征描述]
- 触达：[触达策略]
---
4层需求
| 层次 | 用户心理 | 我们的回答 |
|---|---|---|
| 认知 | 为什么要用？ | [你的回答] |
| 兴趣 | 适合我吗？ | [你的回答] |
| 决策 | 难学吗？ | [你的回答] |
| 行动 | 接下来学什么？ | [你的回答] |
Related skills
Seo AuditRun structured SEO audits on their SaaS site or content hub and receive a prioritized action plan.173k42.2kCopywritingGenerate, rewrite, or strengthen persuasive website and landing-page copy that converts visitors into users.164k42.2k
Viral Short FormQuickly generate high-retention hooks, scripts, and outlines for TikTok, Reels, YouTube Shorts, and carousels.132k67Viral HooksWrite and critique viral hooks for short-form video opening sequences.123k67Viral Captions And CtasOptimize social media captions and CTAs for viral short-form video reach and saves.123k67Viral Youtube ShortsWrite and diagnose YouTube Shorts for Shorts Feed and long-form funnel.123k67FAQ
What is the publish threshold?
Score at least 80% on the checklist to publish, 60 to 79% to optimize, and below 60% to redo.
What scenarios does it handle?
Writing a new article, checking a draft, diagnosing poor data, titling, and formatting.