Skip to content

alchaincyf/x-mentor-skill

x-mastery-mentor

Verified

X/Twitter运营:选题、推文与长文、账号诊断、增长和转化。用户说X运营、推特、写推文、涨粉、tweet、thread、发推时使用。以本人数据和真实经历校正通用经验。

references/writing-workshop.md

View on GitHub
# 写作工坊

> 2026-10-11更新:以下字数、1/3/1、Thread条数与CTA公式仅作可选手法;核心以SKILL.md的Tibo准则、本人认可范例与当前任务为准,不强压篇幅、不强制外链入回复。
> 按需加载:场景A(写推文/Thread)、场景B(选题)、场景C(审阅内容)时读取。

写花叔的中文X短表达时,先读[本人原文与认可范例](huashu_x_expression_examples.md)。用范例校准判断、洞察与可用动作,再选择本工坊中适合当前任务的组织方式。

---

## 短推文写法

**适用场景**:验证想法、日常互动、观点输出

**格式选择**:
- 观点陈述:一个鲜明立场(「大多数人X做错了...」)
- 金句:可截图传播的一句话
- 真诚提问:需要读者经验时使用,不为算法凑回复。
- 个人重定义:「X不是Y,X是Z」

**Dickie Bush的配比**:75%广度内容(viral-friendly,谁写都能传播)+ 25%深度内容(个人故事/独特视角,建立辨识度)

---

## Hook改进示例(Before → After)

**示例1:AI工具推广**
- Before: `I tested 5 AI coding tools. Here's what I found.`
- 问题:无好奇缺口、无可信度锚点、无具体收益
- After: `I mass-tested 5 AI coding tools on the same project (a full-stack app in 48 hours). One saved me 12 hours. The others were useless. A thread:`
- 改了什么:加了具体场景(full-stack app)、时间锚点(48h)、悬念(哪个?)、可信度(实测数据)

**示例2:Build in Public**
- Before: `Just launched my new app. Check it out!`
- 问题:零好奇缺口、自说自话、没回答「为什么我要看」
- After: `I built an iOS app with zero coding experience using only AI tools. It hit #1 Paid on the App Store in 3 days. Here's exactly how (and what almost killed it):`
- 改了什么:加了身份反差(零经验)、结果锚点(#1 Paid)、时间框架(3天)、悬念(almost killed it)

**示例3:争议观点**
- Before: `AI coding tools are overrated.`
- 问题:观点太模糊、没有stake in the ground
- After: `Unpopular opinion: 90% of "AI coding tools" reviews on X are from people who never shipped a real product. I've shipped 3 apps. Here's what actually works vs what's just demo-ware:`
- 改了什么:加了具体数据(90%)、可信度(shipped 3 apps)、对立面(demo-ware)、承诺交付(what works)

---

## Thread写法

**四段结构**(Cole/Bush):

**1. Hook(开头)**
必须回答:给谁看?讲什么?为什么信你?能得到什么?

Hook公式:
- 可信度元素(「我研究了1000个...」「作为X年的Y...」)
- 场景锚点(「In 2024...」「上周...」)
- 核心收益(「How to...」「Why...」)
- 具体交付(「10 bullets on...」「A thread:」)

**2. Main Points(正文)**
- 每条tweet第一句当小标题
- 1/3/1节奏
- 每条能独立成文
- 条数由论证需要决定,不设平台最佳条数。

**3. TL;DR**
- 先写TL;DR当大纲
- 只列要点标题

**4. CTA(行动召唤)**
- 总结核心收获
- 明确下一步(Follow/Bookmark/Newsletter)

**Cole的7种模板**:
1. **Framework**:To solve X, I do Y, To achieve Z
2. **Story**:先给结局→再给开头→读者为找「中间」读完
3. **Actionable**:清单式可执行建议
4. **Curation**:「我读完了{人名}所有内容,这是最好的N条」
5. **Lessons**:从个人经历提取教训
6. **Mistakes**:「X mistakes I made doing Y」
7. **Contrarian**:挑战常见认知

---

## AI赛道特有Thread类型

1. 「我测了X,结果令人惊讶」——新模型/工具实测
2. 「从0到$XK MRR的N个教训」——Build in Public总结
3. 「这篇论文改变了我的认知」——论文解读
4. 「X vs Y:深度对比」——工具/模型横评
5. 「我用AI做了X,省了N小时」——实战案例

---

## 选题系统

**Endless Idea Generator**(Cole/Bush,30分钟100+选题):

**Step 1: 2-Year Test**
问自己:过去两年我解决了什么问题?学了什么?列出3-5个主题桶。

**Step 2: 加限定**
给主题加具体限定直到「不舒服地具体」。目标读者 = 两年前的你自己。

**Step 3: 4A × 话题 = 矩阵**
一个话题 × 4角度(Actionable/Analytical/Aspirational/Anthropological) × 多种格式 = 数百篇内容。

**Step 4: 选3个最让你兴奋的,开写**

**Bush的大小问题策略**:
- 大问题(how to be happier)= 大受众但浅 → 拉新
- 小问题(how to grow podcast from 2K to 10K)= 小受众但深 → 留人
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 194kUpdated Claude CodeCodexVerifiedInstall