Back to Details

copywriting-hooks

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Copywriting Hooks

文案钩子创作

The method

创作方法

A hook's only job is to make the reader want sentence 2. Voice, structure, formatting, all of it, follow from that one job.
What makes a reader want sentence 2 is one of five levers:
  1. Open a gap. Pose something incomplete that the reader needs to close. Curiosity gap, question, open loop.
  2. Break a prediction. State something that violates the reader's prior. Contrarian, definition reversal, surprising statistic.
  3. Drop into a scene. Load sensory or specific detail that builds a vivid frame. In medias res, concrete detail, time anchor.
  4. Promise a payoff. Name an outcome the reader wants. Benefit, "if you... then this", direct problem.
  5. Borrow weight. Lean on a name, number, or quote that carries embedded authority. Authority hook, statistic, quote with disagreement.
A strong hook usually pulls two levers at once. "Frank Sinatra, holding a glass of bourbon..." is scene plus open loop. "Most people think X. They're wrong." is prediction-break plus gap. Single-lever hooks can still work but are easier to ignore.
Three further principles:
  • Specific beats abstract every time. Replace "many companies" with "Stripe, Shopify, Vercel". Replace "recently" with a date. Replace "studies show" with the actual finding or cut the claim.
  • The first sentence must force the second. Read each candidate cold. If you would not click sentence 2 after sentence 1, rewrite.
  • Match technique to article type. Personal essay does not open like a tutorial. See the type-fit table below.
钩子的唯一作用就是让读者想要读第二句。语气、结构、格式等所有要素，都要围绕这一核心目标展开。
能让读者想要读第二句的是以下五大心理杠杆之一：
  1. 打开信息缺口。提出读者想要填补的不完整信息，比如好奇心缺口、问题、开放式循环。
  2. 打破预期。陈述违背读者既有认知的内容，比如逆向观点、定义反转、惊人数据。
  3. 场景切入。加入感官或具体细节构建生动场景，比如直入核心、具象细节、时间锚点。
  4. 承诺收益。点明读者想要的结果，比如利益点、“如果你……就会……”、直接痛点。
  5. 借势权威。借助具有权威属性的名字、数字或引语，比如权威钩子、数据、带有争议的引语。
优秀的钩子通常会同时运用两个杠杆。比如“Frank Sinatra，手里拿着一杯波本威士忌……”就是场景切入+开放式循环；“大多数人认为X，但他们错了”就是打破预期+信息缺口。仅用单一杠杆的钩子也能生效，但更容易被读者忽略。
还有三条补充原则：
  • 具象永远优于抽象。把“许多公司”替换成“Stripe、Shopify、Vercel”；把“最近”替换成具体日期；把“研究表明”替换成具体结论，或者直接删掉这种表述。
  • 第一句必须能引导读者读第二句。客观审视每个候选钩子，如果读完第一句你不想读第二句，就重写。
  • 技巧要匹配文章类型。个人随笔的开头不能像教程一样。详见下方的类型适配表。

Behavior

操作流程

When this skill triggers:
  1. Confirm the brief. Topic, audience, target language (EN, FR, or both), approximate length, where it will be published. If any of these is unclear and material, ask before generating.
  2. Pick 3 to 4 hooks from the catalog below that are genuinely different. Different levers, not three flavors of contrarian.
  3. Write 2 candidates per hook, specific to the user's article. The two candidates within one hook should explore different angles (different anecdote, different statistic, different scene), not be rewordings of each other.
  4. Present using the Output format below.
  5. Use 
    ask_user_input_v0
     if available. The choice is a small fixed set, which is what that tool is for.
  6. Wait for the user's pick. Do not pick for them.
  7. After they pick, name what the choice commits the rest of the article to. A contrarian hook commits paragraphs 2 to 3 to defending the non-consensus claim. A scene opener commits the next section to either resolving the scene or productively delaying it.
Diversification rule. Across your 3 to 4 options include at minimum:
  • One intellectual hook (contrarian, definition reversal, historical analogy, curiosity gap)
  • One sensory hook (in medias res, concrete detail)
  • One reader-direct hook (conditional, direct problem, promise)
This guarantees real choice. Three flavors of contrarian is not a choice.
Type-fit reference:
Article typeStrong hooksAvoid
Technical deep-diveconcrete detail, statistic, contrarian, direct problempersonal confession, scene opener
Personal essayin medias res, personal confession, time anchor, definition reversalbold claim, direct problem
Opinion / contrarianbold claim, definition reversal, contrarian, quote + disagreementgentle setup, dictionary opener
Tutorial / how-todirect problem (PAS), promise, conditionalscene opener, historical analogy
Reported / investigativeconcrete detail, time anchor, in medias res, statisticbold claim, definition reversal
Listiclecuriosity gap, counted stakes, conditionalpersonal confession, in medias res
Longform analysishistorical analogy, statistic, contrariandirect problem
Newsletter issuepersonal confession + open loop, conditional, curiosity gapdictionary opener
当本技能触发时：
  1. 确认需求细节。包括主题、受众、目标语言（英文、法文或双语）、文章大致篇幅、发布平台。如果任何关键信息不明确，先询问用户再生成内容。
  2. 从下方钩子目录中挑选3-4种完全不同的钩子。要选用不同杠杆的钩子，而不是三种变体的逆向思维钩子。
  3. 针对用户的文章，为每种钩子撰写2个候选方案。同一钩子下的两个候选要探索不同角度（不同轶事、不同数据、不同场景），而不是简单改写。
  4. 按照下方的输出格式呈现内容。
  5. 如果可用，请使用
    ask_user_input_v0
    工具    。用户的选择是固定的小范围选项，正好适用该工具。
  6. 等待用户选择。不要替用户做决定。
  7. 用户选定后，说明该选择对文章后续内容的要求。比如逆向思维钩子要求第2-3段为非共识观点辩护；场景开头要求下一部分要么收尾场景，要么合理推迟收尾。
多样化规则。在3-4个选项中，至少要包含：
  • 一个理性钩子（逆向思维、定义反转、历史类比、好奇心缺口）
  • 一个感官钩子（直入核心、具象细节）
  • 一个面向读者的钩子（条件句、直接痛点、收益承诺）
这样才能保证用户有真正的选择空间。三个同类的逆向思维钩子不算选择。
类型适配参考表：
文章类型适用钩子避免使用
技术深度解析具象细节、数据、逆向思维、直接痛点个人自白、场景开头
个人随笔直入核心、个人自白、时间锚点、定义反转大胆断言、直接痛点
评论/逆向观点大胆断言、定义反转、逆向思维、引语+争议温和铺垫、字典式开头
教程/指南直接痛点（PAS模式）、收益承诺、条件句场景开头、历史类比
深度报道/调查类具象细节、时间锚点、直入核心、数据大胆断言、定义反转
清单文好奇心缺口、明确 stakes、条件句个人自白、直入核心
长篇分析历史类比、数据、逆向思维直接痛点
通讯刊物个人自白+开放式循环、条件句、好奇心缺口字典式开头

Output format

输出格式

Always present options exactly like this:
undefined
请严格按照以下格式呈现选项：
undefined

Hook options for: [working title]

Hook options for: [working title]

Option 1: [Hook name] ([lever]) A. [Candidate 1] B. [Candidate 2]
Option 2: [Hook name] ([lever]) A. [Candidate 1] B. [Candidate 2]
Option 3: [Hook name] ([lever]) A. [Candidate 1] B. [Candidate 2]
Which? Reply with letter combination (e.g., "1B") or "more" for different techniques.

If the user says "more" or "none", produce 3 different hooks (not new candidates for the same hooks). If the user says "blend 1A and 2B", write a combined hook and check in again.

---
Option 1: [Hook name] ([lever]) A. [Candidate 1] B. [Candidate 2]
Option 2: [Hook name] ([lever]) A. [Candidate 1] B. [Candidate 2]
Option 3: [Hook name] ([lever]) A. [Candidate 1] B. [Candidate 2]
Which? Reply with letter combination (e.g., "1B") or "more" for different techniques.

如果用户说“more”或“none”，请生成3种全新的钩子（不是同一钩子的新候选）。如果用户说“blend 1A and 2B”，请撰写一个融合后的钩子并再次确认。

---

The hook catalog (18)

钩子目录（18种）

Each hook: what it does, examples (mix EN and FR, real and illustrative), when to use, when to avoid.
每种钩子包含：作用、示例（混合英文和法文，真实案例与演示案例）、适用场景、避免场景。

1. Curiosity Gap

1. 好奇心缺口

Open an information gap the reader wants closed.
  • EN: "How does Shen Yun make any money? Short answer: they don't." (Packy McCormick)
  • EN: "Many years ago, one mustard dominated the supermarket shelves: French's." (Gladwell, "The Ketchup Conundrum")
  • FR: "Trois startups françaises ont franchi le milliard cette année. Aucune n'est dans la tech."
  • EN: "I've been a surgeon for eight years. For the past couple of them, my performance in the operating room has reached a plateau." (Gawande)
Use when: you can honestly close the gap in 2 or 3 sentences. The gap must be one the reader cares about. Avoid: vague gaps ("You won't believe what happened next") that the reader cannot even guess at.
打开读者想要填补的信息缺口。
  • EN: "How does Shen Yun make any money? Short answer: they don't." (Packy McCormick)
  • EN: "Many years ago, one mustard dominated the supermarket shelves: French's." (Gladwell, "The Ketchup Conundrum")
  • FR: "Trois startups françaises ont franchi le milliard cette année. Aucune n'est dans la tech."
  • EN: "I've been a surgeon for eight years. For the past couple of them, my performance in the operating room has reached a plateau." (Gawande)
适用场景：你能在2-3句话内合理填补缺口，且该缺口是读者关心的内容。避免场景：模糊的缺口（比如“你绝对想不到接下来发生了什么”），读者甚至无法猜测缺口内容。

2. Contrarian

2. 逆向思维

Knock down a consensus belief the reader holds.
  • EN: "Prevailing wisdom claims that the best way to achieve what we want in life is to set specific, actionable goals." (Clear, "Forget About Setting Goals", proceeds to argue the opposite)
  • EN: "To do something well you have to like it. That idea is not exactly novel." (Graham, "Do What You Love", proceeds to complicate it)
  • EN: "If you're not saying 'HELL YEAH!' about something, say no." (Sivers)
  • FR: "Le Marketing est TOUT ce que l'IA ne peut pas faire !" (Truphème)
Use when: you have a defensible non-consensus view and 200 to 400 words to defend it. Avoid: strawmen, gratuitous edginess, contrarianism for its own sake.
推翻读者普遍认同的观点。
  • EN: "Prevailing wisdom claims that the best way to achieve what we want in life is to set specific, actionable goals." (Clear, "Forget About Setting Goals", proceeds to argue the opposite)
  • EN: "To do something well you have to like it. That idea is not exactly novel." (Graham, "Do What You Love", proceeds to complicate it)
  • EN: "If you're not saying 'HELL YEAH!' about something, say no." (Sivers)
  • FR: "Le Marketing est TOUT ce que l'IA ne peut pas faire !" (Truphème)
适用场景：你有站得住脚的非共识观点，且能用200-400字为其辩护。避免场景：稻草人谬误、无端尖锐、为了逆向而逆向。

3. Bold Claim / Promise

3. 大胆断言/收益承诺

State the outcome upfront, before the proof.
  • EN: "At 60 miles an hour the loudest noise in this new Rolls-Royce comes from the electric clock." (Ogilvy)
  • EN: "They laughed when I sat down at the piano. But when I started to play!—" (Caples 1925)
  • EN: "Give me 15 minutes and I'll give you a super-power memory." (Schwartz)
  • FR: "En 5 minutes : la structure qui fait passer le taux de complétion d'un article de 18% à 64%."
Use when: you can deliver on the promise concretely in the article. Avoid: promises larger than the payoff. Destroys trust permanently.
在给出证据前，先点明结果。
  • EN: "At 60 miles an hour the loudest noise in this new Rolls-Royce comes from the electric clock." (Ogilvy)
  • EN: "They laughed when I sat down at the piano. But when I started to play!—" (Caples 1925)
  • EN: "Give me 15 minutes and I'll give you a super-power memory." (Schwartz)
  • FR: "En 5 minutes : la structure qui fait passer le taux de complétion d'un article de 18% à 64%."
适用场景：你能在文章中切实兑现承诺。避免场景：承诺大于实际收益，会永久破坏读者信任。

4. Scene Opener / In Medias Res

4. 场景切入/直入核心

Drop the reader inside a specific moment, no setup.
  • EN: "Frank Sinatra, holding a glass of bourbon in one hand and a cigarette in the other, stood in a dark corner of the bar between two attractive but fading blondes who sat waiting for him to say something." (Talese)
  • EN: "We were somewhere around Barstow on the edge of the desert when the drugs began to take hold." (Thompson)
  • EN: "The center was not holding." (Didion)
  • FR: "Le 14 mai 2024, à 6h12, trois hommes en civil sonnent à la porte d'un appartement du XVe arrondissement."
Use when: longform, profile, reported piece, essay. Avoid: short technical pieces where the reader has not earned the scene yet.
直接将读者带入特定场景，无需铺垫。
  • EN: "Frank Sinatra, holding a glass of bourbon in one hand and a cigarette in the other, stood in a dark corner of the bar between two attractive but fading blondes who sat waiting for him to say something." (Talese)
  • EN: "We were somewhere around Barstow on the edge of the desert when the drugs began to take hold." (Thompson)
  • EN: "The center was not holding." (Didion)
  • FR: "Le 14 mai 2024, à 6h12, trois hommes en civil sonnent à la porte d'un appartement du XVe arrondissement."
适用场景：长篇内容、人物特写、深度报道、随笔。避免场景：短篇技术内容，读者还没做好接受场景的准备。

5. Surprising Statistic

5. 惊人数据

Lead with a number that violates the prior.
  • EN: "You have five seconds to get people's attention." (Housel, self-demonstrating)
  • EN: "Thirty-seven thousand Americans died in car accidents in 1955, six times today's rate adjusted for miles driven." (Housel)
  • EN: "Approximately 40 percent of the actions you perform each day are habits, not actual decisions." (Clear)
  • FR: "Tu as moins de 3 secondes pour convaincre tes audiences que ton contenu est intéressant." (Truphème)
Use when: the number is genuinely surprising and you can cite it accurately. Avoid: vague stats ("studies show 90%..."), stats that confirm the reader's prior (no surprise equals no hook).
以违背读者既有认知的数字开篇。
  • EN: "You have five seconds to get people's attention." (Housel, self-demonstrating)
  • EN: "Thirty-seven thousand Americans died in car accidents in 1955, six times today's rate adjusted for miles driven." (Housel)
  • EN: "Approximately 40 percent of the actions you perform each day are habits, not actual decisions." (Clear)
  • FR: "Tu as moins de 3 secondes pour convaincre tes audiences que ton contenu est intéressant." (Truphème)
适用场景：数字确实令人惊讶，且你能准确引用来源。避免场景：模糊数据（比如“研究表明90%的人……”）、符合读者既有认知的数据（没有惊喜就没有钩子效果）。

6. Question

6. 提问式

Pose a question the reader actually wants answered.
  • EN: "If you collected lists of techniques for doing great work in a lot of different fields, what would the intersection look like?" (Graham)
  • EN: "Why do we make kids go to school?" (Caplan)
  • FR: "Pourquoi 80% des articles LinkedIn ne dépassent jamais 100 vues ? Spoiler : ce n'est pas l'algorithme."
  • FR: "Avez-vous le courage de gagner la moitié d'un million de dollars cette année ?" (Schwartz)
Use when: the reader was implicitly carrying the question. Avoid: "Have you ever wondered...?", "Did you know...?", "What if I told you...?". These presuppose curiosity not yet formed.
提出读者真正想要得到答案的问题。
  • EN: "If you collected lists of techniques for doing great work in a lot of different fields, what would the intersection look like?" (Graham)
  • EN: "Why do we make kids go to school?" (Caplan)
  • FR: "Pourquoi 80% des articles LinkedIn ne dépassent jamais 100 vues ? Spoiler : ce n'est pas l'algorithme."
  • FR: "Avez-vous le courage de gagner la moitié d'un million de dollars cette année ?" (Schwartz)
适用场景：读者内心已经隐含这个问题。避免场景：“你有没有想过……？”“你知道吗……？”“如果我告诉你……？”这类预设读者好奇心的表述，而此时读者的好奇心尚未形成。

7. Quote + Disagreement

7. 引语+争议

Borrow weight, then twist.
  • EN: "Vernor Vinge: 'We are on the edge of change comparable to the rise of human life on Earth.'" (Urban setup)
  • EN: "Steve Jobs said people don't know what they want until you show it to them. For SaaS, this is exactly backwards."
  • EN: "Henry Ford supposedly said his customers would have asked for faster horses. For most products, that excuse is wrong."
  • FR: "Henry Ford disait qu'on lui aurait demandé des chevaux plus rapides. Pour 90% des produits, c'est faux."
Use when: you have a real quote that supports or genuinely contrasts your point. Avoid: misattributed Einstein, Seneca, Confucius, Bouddha platitudes. Cliché.
借势权威，再反转观点。
  • EN: "Vernor Vinge: 'We are on the edge of change comparable to the rise of human life on Earth.'" (Urban setup)
  • EN: "Steve Jobs said people don't know what they want until you show it to them. For SaaS, this is exactly backwards."
  • EN: "Henry Ford supposedly said his customers would have asked for faster horses. For most products, that excuse is wrong."
  • FR: "Henry Ford disait qu'on lui aurait demandé des chevaux plus rapides. Pour 90% des produits, c'est faux."
适用场景：你有真实的引语能够支撑或真正对比你的观点。避免场景：错误归因于爱因斯坦、塞内卡、孔子、佛陀的陈词滥调，过于俗套。

8. Personal Confession

8. 个人自白

Admit something vulnerable, then universalize.
  • EN: "I cheated on my husband." (Strayed)
  • EN: "I've been thinking about my parents, who are in their mid-60s. During my first 18 years, I spent some time with my parents during at least 90% of my days." (Urban, "The Tail End")
  • EN: "It's been a minute. This is as long as I've gone without writing an essay since starting Not Boring." (McCormick)
  • FR: "À 34 ans, j'ai démissionné sans plan B. Voici ce que personne ne dit sur la suite."
Use when: personal byline, essay, newsletter. Avoid: corporate byline, technical articles where the author voice is not personal, performative vulnerability ("I almost didn't write this...").
承认脆弱的内容，再将其普遍化。
  • EN: "I cheated on my husband." (Strayed)
  • EN: "I've been thinking about my parents, who are in their mid-60s. During my first 18 years, I spent some time with my parents during at least 90% of my days." (Urban, "The Tail End")
  • EN: "It's been a minute. This is as long as I've gone without writing an essay since starting Not Boring." (McCormick)
  • FR: "À 34 ans, j'ai démissionné sans plan B. Voici ce que personne ne dit sur la suite."
适用场景：个人署名文章、随笔、通讯刊物。避免场景：企业署名文章、作者语气非个人化的技术文章、表演式脆弱（比如“我差点没写这篇文章……”）。

9. Concrete Specific Detail

9. 具象细节

Replace abstraction with a single vivid detail.
  • EN: "John Laroche is a tall guy, skinny as a stick, pale-eyed, slouch-shouldered, and sharply handsome, despite the fact that he is missing all his front teeth." (Orlean)
  • EN: "Charles Bukowski was an alcoholic, a womanizer, a chronic gambler, a lout, a cheapskate, a deadbeat, and on his worst days, a poet." (Manson)
  • FR: "Quand on la voit arriver dans les locaux du Bondy Blog, ce samedi 24 février, difficile de se dire qu'elle est fichée illégalement par les services de renseignement français."
  • FR: "Un Post-it collé sur mon écran : 'Supprime ton premier paragraphe.'"
Use when: profile, reported piece, contrarian biographical setup. Avoid: specificity that does not advance the thesis (clutter).
用单一生动细节替代抽象表述。
  • EN: "John Laroche is a tall guy, skinny as a stick, pale-eyed, slouch-shouldered, and sharply handsome, despite the fact that he is missing all his front teeth." (Orlean)
  • EN: "Charles Bukowski was an alcoholic, a womanizer, a chronic gambler, a lout, a cheapskate, a deadbeat, and on his worst days, a poet." (Manson)
  • FR: "Quand on la voit arriver dans les locaux du Bondy Blog, ce samedi 24 février, difficile de se dire qu'elle est fichée illégalement par les services de renseignement français."
  • FR: "Un Post-it collé sur mon écran : 'Supprime ton premier paragraphe.'"
适用场景：人物特写、深度报道、逆向观点的传记式铺垫。避免场景：无助于推进主题的细节（冗余内容）。

10. Pattern Interrupt

10. 节奏打断

Break expected rhythm with a fragment.
  • EN: "Wait." (Godin style)
  • EN: "Stop."
  • EN: "This is not an article about productivity. It's an article about identity."
  • FR: "Non. Ce que vous avez lu cette semaine sur l'IA est faux. Voici pourquoi."
Use when: rhythm-driven content, opinion piece, when the reader expects flow and you want to interrupt it. Avoid: every article. It becomes its own pattern fast.
用片段打破预期节奏。
  • EN: "Wait." (Godin style)
  • EN: "Stop."
  • EN: "This is not an article about productivity. It's an article about identity."
  • FR: "Non. Ce que vous avez lu cette semaine sur l'IA est faux. Voici pourquoi."
适用场景：注重节奏的内容、评论文章，当读者预期流畅内容时，你想要打断节奏。避免场景：所有文章都用，很快会形成新的俗套。

11. Direct Problem (PAS)

11. 直接痛点（PAS模式）

Name the pain, sharpen it, hint at solution.
  • EN: "Your articles aren't read. The data says you lose 80% of readers by paragraph two. There's a fix, and it's not what you've been doing."
  • EN: "If you're like most marketing managers, you don't have enough time to write your white papers, and the ones you outsource come back generic." (Bly)
  • FR: "Vos emails ne sont pas ouverts. Trois mois, taux plafonné à 12%. Le problème n'est pas l'objet, c'est l'expéditeur."
  • FR: "Vos articles ne sont pas lus. 80% des lecteurs décrochent dès le deuxième paragraphe. Et la solution n'est pas celle que vous croyez."
Use when: tutorial, how-to, sales-adjacent content. Avoid: manufactured problems. Reads as fear-mongering. The pain must be real and recognizable.
点明痛点、强化痛点、暗示解决方案。
  • EN: "Your articles aren't read. The data says you lose 80% of readers by paragraph two. There's a fix, and it's not what you've been doing."
  • EN: "If you're like most marketing managers, you don't have enough time to write your white papers, and the ones you outsource come back generic." (Bly)
  • FR: "Vos emails ne sont pas ouverts. Trois mois, taux plafonné à 12%. Le problème n'est pas l'objet, c'est l'expéditeur."
  • FR: "Vos articles ne sont pas lus. 80% des lecteurs décrochent dès le deuxième paragraphe. Et la solution n'est pas celle que vous croyez."
适用场景：教程、指南、销售相关内容。避免场景：制造虚假痛点，读起来像恐惧营销，痛点必须真实且读者能共情。

12. Promise / Benefit

12. 收益承诺

State a specific, bounded outcome.
  • EN: "Read this in 5 minutes and you'll never write a weak opening again."
  • EN: "By the end of this article, you'll know exactly when to use goroutines and when not to."
  • EN: "Three things separate writers who get read from writers who don't. None involve writing every day."
  • FR: "En 5 minutes : la structure qui fait passer le taux de complétion d'un article de 18% à 64%."
Use when: tutorial, how-to. Avoid: vague promises ("Learn how to be more productive"). Add a time bound or a number to anchor it.
点明具体、有限的结果。
  • EN: "Read this in 5 minutes and you'll never write a weak opening again."
  • EN: "By the end of this article, you'll know exactly when to use goroutines and when not to."
  • EN: "Three things separate writers who get read from writers who don't. None involve writing every day."
  • FR: "En 5 minutes : la structure qui fait passer le taux de complétion d'un article de 18% à 64%."
适用场景：教程、指南。避免场景：模糊承诺（比如“学会如何提高效率”），要添加时间限制或数字来锚定承诺。

13. Historical Analogy

13. 历史类比

Open with a vignette from history, pivot to the present.
  • EN: "On a Tuesday morning in March 1976, a 21-year-old college dropout sold his Volkswagen Bus for 1,500 dollars. He used the money to build the first Apple computer."
  • EN: "In 1965, Robert Lucas wrote a four-page paper that broke macroeconomics."
  • EN: Wright Brothers vignette (Housel, "Three Big Things"), pivots to demographics and interest rates.
  • FR: "En 1903, à Kitty Hawk, deux frères réparateurs de vélos volent pendant 12 secondes. Personne ne remarque. Voici ce que cette indifférence dit de l'adoption technologique aujourd'hui."
Use when: longform analysis, opinion piece, idea essay. Avoid: tutorials, news pieces. Reads as indulgent.
以历史片段开篇，再转向当下。
  • EN: "On a Tuesday morning in March 1976, a 21-year-old college dropout sold his Volkswagen Bus for 1,500 dollars. He used the money to build the first Apple computer."
  • EN: "In 1965, Robert Lucas wrote a four-page paper that broke macroeconomics."
  • EN: Wright Brothers vignette (Housel, "Three Big Things"), pivots to demographics and interest rates.
  • FR: "En 1903, à Kitty Hawk, deux frères réparateurs de vélos volent pendant 12 secondes. Personne ne remarque. Voici ce que cette indifférence dit de l'adoption technologique aujourd'hui."
适用场景：长篇分析、评论文章、观点随笔。避免场景：教程、新闻内容，会显得过于自我 indulgent。

14. Definition Reversal

14. 定义反转

"X is not what you think. It's Y."
  • EN: "Procrastination isn't laziness. It's a fight between two parts of your brain."
  • EN: "This is not an article about productivity. It's an article about identity."
  • EN: "Generics were the most requested Go feature for a decade. Three years in, the people who pushed hardest for them are telling you to stop using them."
  • FR: "Le copywriting n'est pas de l'écriture. C'est de la psychologie déguisée en phrases."
Use when: opinion piece, contrarian deep dive. Avoid: when your reframe is just a slight angle. Sounds gimmicky.
“X不是你想的那样，它是Y。”
  • EN: "Procrastination isn't laziness. It's a fight between two parts of your brain."
  • EN: "This is not an article about productivity. It's an article about identity."
  • EN: "Generics were the most requested Go feature for a decade. Three years in, the people who pushed hardest for them are telling you to stop using them."
  • FR: "Le copywriting n'est pas de l'écriture. C'est de la psychologie déguisée en phrases."
适用场景：评论文章、逆向深度解析。避免场景：你的重构只是轻微角度调整，会显得 gimmicky。

15. Authority Borrow

15. 权威借势

Lead with a name plus a specific action.
  • EN: "Warren Buffett, at 91, still reads 500 pages a day."
  • EN: "When Steve Jobs returned to Apple in 1997, he killed 70% of the product line in his first year."
  • EN: "A boy once asked Charlie Munger..." (Housel narrative variant)
  • FR: "Quand Bernard Arnault a racheté Tiffany pour 16 milliards, trois analystes ont prédit un échec. Ils s'étaient tous trompés sur la même chose."
Use when: business piece, profile, analytical essay. Avoid: name-dropping without payoff. The action must be specific and relevant.
以知名人物加具体行为开篇。
  • EN: "Warren Buffett, at 91, still reads 500 pages a day."
  • EN: "When Steve Jobs returned to Apple in 1997, he killed 70% of the product line in his first year."
  • EN: "A boy once asked Charlie Munger..." (Housel narrative variant)
  • FR: "Quand Bernard Arnault a racheté Tiffany pour 16 milliards, trois analystes ont prédit un échec. Ils s'étaient tous trompés sur la même chose."
适用场景：商业文章、人物特写、分析随笔。避免场景：无意义的名人提及，行为必须具体且相关。

16. Time Anchor

16. 时间锚点

Lead with a specific date, hour, or moment.
  • EN: "October 2005. Three journalists set up a blog in an apartment of the cité Blanqui, in Bondy."
  • EN: "In 2022, the cost of writing a competent article dropped to zero. Most writers haven't noticed."
  • EN: "On a Tuesday morning in March 1976, a 21-year-old dropout sold his Volkswagen Bus for 1,500 dollars."
  • FR: "Octobre 2005, Bondy. Trois journalistes installent un blog dans un appartement de la cité Blanqui."
Use when: reported piece, retrospective, "why now" framing. Avoid: vague time anchors ("recently", "these past few years"). Use a specific date or cut the time framing.
以具体日期、小时或时刻开篇。
  • EN: "October 2005. Three journalists set up a blog in an apartment of the cité Blanqui, in Bondy."
  • EN: "In 2022, the cost of writing a competent article dropped to zero. Most writers haven't noticed."
  • EN: "On a Tuesday morning in March 1976, a 21-year-old dropout sold his Volkswagen Bus for 1,500 dollars."
  • FR: "Octobre 2005, Bondy. Trois journalistes installent un blog dans un appartement de la cité Blanqui."
适用场景：深度报道、回顾性内容、“为何是现在”的框架。避免场景：模糊的时间锚点（比如“最近”“过去几年”），要么用具体日期，要么删掉时间框架。

17. Conditional ("If you... then this")

17. 条件句（“如果你……就会……”）

Self-segment the reader. Pre-target the curiosity.
  • EN: "If you write for a living, you've probably been taught to start with context. Don't."
  • EN: "If you've ever read a blog post and forgotten what it said within the hour, this article is the diagnostic."
  • EN: "If you're not saying 'HELL YEAH!' about something, say no." (Sivers)
  • FR: "Si vos articles sont lus à 80% par vos collègues et à 8% par vos prospects, le problème n'est pas le contenu. C'est l'accroche."
Use when: tutorial, advice piece, segmented audience. Avoid: conditions too broad ("If you've ever felt stuck..."). Segments nobody.
自我筛选读者，精准激发好奇心。
  • EN: "If you write for a living, you've probably been taught to start with context. Don't."
  • EN: "If you've ever read a blog post and forgotten what it said within the hour, this article is the diagnostic."
  • EN: "If you're not saying 'HELL YEAH!' about something, say no." (Sivers)
  • FR: "Si vos articles sont lus à 80% par vos collègues et à 8% par vos prospects, le problème n'est pas le contenu. C'est l'accroche."
适用场景：教程、建议类文章、细分受众内容。避免场景：条件过于宽泛（比如“如果你曾感到迷茫……”），无法筛选出目标读者。

18. Open Loop

18. 开放式循环

Start something, withhold the resolution.
  • EN: Talese's Sinatra waiting for him to say something. He never says.
  • EN: "He pressed Send and waited. Forty-seven seconds later, the company was worth 4 billion dollars less."
  • EN: "Now here's where it gets really interesting..." (Sugarman, transition phrase, also valid as opener)
  • FR: "J'ai été contacté par un éditeur la semaine dernière. Ce qu'il m'a demandé m'a fait reconsidérer 15 ans de pratique."
Use when: longform, pieces where the journey matters as much as the answer. Avoid: unresolved loops. Creates disproportionate betrayal when the article ends without paying off.
开启一件事，但不给出结局。
  • EN: Talese's Sinatra waiting for him to say something. He never says.
  • EN: "He pressed Send and waited. Forty-seven seconds later, the company was worth 4 billion dollars less."
  • EN: "Now here's where it gets really interesting..." (Sugarman, transition phrase, also valid as opener)
  • FR: "J'ai été contacté par un éditeur la semaine dernière. Ce qu'il m'a demandé m'a fait reconsidérer 15 ans de pratique."
适用场景：长篇内容，过程和结果同样重要的文章。避免场景：未闭合的循环，当文章结束仍未兑现时，会引发读者强烈的背叛感。

Anti-patterns (never propose any of these)

反模式（绝对不要推荐以下内容）

Cliché openers that immediately disqualify the writer:
  • "In today's fast-paced world..." / "À l'heure du tout-numérique" / "À l'ère de l'IA" / "Dans un monde où..."
  • "Have you ever wondered...?" / "Vous êtes-vous déjà demandé...?"
  • Dictionary opener played straight ("Productivity, defined as...")
  • "In this article, I'll discuss..." / "Dans cet article, nous allons voir..."
  • Generic stats without source ("90% of people...", "Les études montrent...")
  • Misattributed Einstein / Seneca / Confucius / Bouddha quotes
  • "I'm not an expert, but..." / "Je ne suis pas spécialiste mais..."
  • Three rhetorical questions in a row
  • "Imagine waking up..." without a specific scene
  • "Hot take:", "Unpopular opinion:", "Voici la vérité que personne ne veut entendre..."
  • "At [Company], we believe..." / "Chez [Entreprise], nous pensons..."
  • "Recently,..." / "Récemment,..." without a specific date
  • "You're not alone."
Current AI tells (refresh yearly):
  • "It's not just X, it's Y" (the formula construction)
  • "Picture this:", "Imagine a world where...", "What if I told you..."
  • "Whether you're a seasoned X or a curious newcomer..."
  • "In the realm of...", "Navigating the landscape of..."
  • "Unlock the power of...", "Dive into...", "Buckle up,", "Let's dive in"
  • "Crucially,", "Notably,", "Importantly,", "Essentially," as sentence openers
  • French AI tells: "Dans un monde en constante évolution", "Plongez dans...", "Découvrez comment...", "Par ailleurs,...", "Notamment,...", "Il est crucial de..."
Run every candidate through this list before presenting. If a candidate matches, rewrite.
For the extended anti-pattern list, see 
references/anti-patterns.md
.
会立刻让作者显得不合格的俗套开头：
  • "In today's fast-paced world..." / "À l'heure du tout-numérique" / "À l'ère de l'IA" / "Dans un monde où..."
  • "Have you ever wondered...?" / "Vous êtes-vous déjà demandé...?"
  • 直白的字典式开头（比如“Productivity, defined as...”）
  • "In this article, I'll discuss..." / "Dans cet article, nous allons voir..."
  • 无来源的通用数据（比如“90% of people...”, “Les études montrent...”）
  • 错误归因于爱因斯坦/塞内卡/孔子/佛陀的引语
  • "I'm not an expert, but..." / "Je ne suis pas spécialiste mais..."
  • 连续三个反问句
  • "Imagine waking up..."但没有具体场景
  • "Hot take:", "Unpopular opinion:", "Voici la vérité que personne ne veut entendre..."
  • "At [Company], we believe..." / "Chez [Entreprise], nous pensons..."
  • "Recently,..." / "Récemment,..."但没有具体日期
  • "You're not alone."
当前AI生成内容的典型特征（每年更新）：
  • "It's not just X, it's Y"（公式化结构）
  • "Picture this:", "Imagine a world where...", "What if I told you..."
  • "Whether you're a seasoned X or a curious newcomer..."
  • "In the realm of...", "Navigating the landscape of..."
  • "Unlock the power of...", "Dive into...", "Buckle up,", "Let's dive in"
  • "Crucially,", "Notably,", "Importantly,", "Essentially,"作为句子开头
  • 法语AI典型特征："Dans un monde en constante évolution", "Plongez dans...", "Découvrez comment...", "Par ailleurs,...", "Notamment,...", "Il est crucial de..."
在呈现候选内容前，请对照此列表检查。如果候选内容符合任何一条，请重写。
如需完整的反模式列表，请查看
references/anti-patterns.md
。

Language handling

语言处理

If the audience is French: write in French. Apply the attaque journalistique register: concrete scene-setting, restrained tone, dated anchors, formal "vous" or restrained tutoiement. Do not translate American hype tropes literally. "You'll never believe..." becomes "Vous n'allez pas en croire vos yeux", which reads as scam in French. French marketing-skepticism baseline is higher than English; high promises trigger réactance faster.
If English: default to direct-response register for marketing or tutorial content, longform register for essays and reported pieces.
If bilingual: produce hooks in both languages, label clearly.
For deeper register guidance, see 
references/anglophone-vs-francophone.md
.
如果受众是法语使用者：用法语写作。采用_attaque journalistique_风格：具象场景构建、克制语气、带日期的锚点、正式的“vous”或克制的“tu”。不要直译美式夸张表述。比如“You'll never believe...”翻译成“Vous n'allez pas en croire vos yeux”，在法语中会显得像诈骗。法语使用者对营销的怀疑程度高于英语使用者，过高的承诺会更快引发抵触情绪。
如果是英语：营销或教程类内容默认采用直接响应风格，随笔和深度报道采用长篇风格。
如果是双语：生成两种语言的钩子，并清晰标注。
如需更详细的风格指导，请查看
references/anglophone-vs-francophone.md
。

Post Titles

文章标题

A title is what the reader sees before clicking. A hook is what they read after. Both earn attention through different mechanisms: the title competes for clicks in a feed or search result; the hook earns continued reading after the click.
标题是读者点击前看到的内容，钩子是读者点击后读到的内容。两者通过不同机制吸引注意力：标题在信息流或搜索结果中竞争点击量；钩子在点击后吸引读者继续阅读。

The core mechanic: calibrated curiosity

核心机制：校准好奇心

A title must open a gap without closing it — but the gap must feel real, not manufactured. Research on 8,977 A/B experiments (Upworthy, Scientific Reports 2024) found a curvilinear relationship: too vague produces confusion (no foothold for curiosity); too specific removes motivation to click. The sweet spot: name the stakes, withhold the resolution.
The craft test. Does the title open a gap the reader cares about, and does the article genuinely close it? If yes, that's craft. If the content doesn't deliver what the title implied, that's clickbait — it destroys trust for future clicks.
标题必须打开缺口但不闭合——但缺口必须真实，而非刻意制造。针对8977组A/B测试的研究（Upworthy，《Scientific Reports》2024）发现了曲线关系：过于模糊会引发困惑（没有好奇心的立足点）；过于具体会消除点击动机。最佳平衡点：点明利害，隐瞒结局。
专业测试：标题是否打开了读者关心的缺口，且文章是否真正填补了这个缺口？如果是，这就是专业创作；如果内容没有兑现标题暗示的承诺，那就是标题党——会破坏未来点击的信任。

Core formulas

核心公式

Curiosity / Gap
  • The [Adjective] Truth About [Topic]
    — "The Counterintuitive Truth About Go Generics"
  • What [Group] Won't Tell You About [Topic]
  • The Real Reason [Phenomenon]
    — "The Real Reason Most Content Gets Zero Shares"
  • [Number] Things Every [Audience] Gets Wrong About [Topic]
Contrarian / Negative (negative superlatives: +63% CTR vs. positive — Outbrain, 65k titles)
  • Stop [Doing X]. Here's Why.
    / 
    Why [Common Belief] Is Wrong
  • [Number] [Myths/Mistakes] That Are Killing Your [Result]
Specificity / Data
  • I Analyzed [Number] [Things] — Here's What I Found
  • [Number]% of [Group] Does This Wrong
  • Brackets signal format honestly and add +38% CTR: 
    How I Cut Build Time by 60% [Benchmark]
List / Number (numbers: +36% CTR — Conductor; odd numbers: +20% CTR — CMI)
  • [Number] [Adjective] Ways to [Goal]
    / 
    [Number] Mistakes to Avoid When [Task]
How-To (3× more B2B shares than other formats — BuzzSumo, 10M LinkedIn articles)
  • How to [Task] Without [Painful Constraint]
    / 
    How to [Task] Even If [Limiting Belief]
好奇心/缺口
  • The [Adjective] Truth About [Topic]
    — "The Counterintuitive Truth About Go Generics"
  • What [Group] Won't Tell You About [Topic]
  • The Real Reason [Phenomenon]
    — "The Real Reason Most Content Gets Zero Shares"
  • [Number] Things Every [Audience] Gets Wrong About [Topic]
逆向/负面表述（负面最高级：点击率比正面高63%——Outbrain，6.5万个标题）
  • Stop [Doing X]. Here's Why.
    / 
    Why [Common Belief] Is Wrong
  • [Number] [Myths/Mistakes] That Are Killing Your [Result]
具象化/数据
  • I Analyzed [Number] [Things] — Here's What I Found
  • [Number]% of [Group] Does This Wrong
  • 括号能清晰表明格式并提升38%点击率：
    How I Cut Build Time by 60% [Benchmark]
清单/数字（数字：点击率提升36%——Conductor；奇数：点击率提升20%——CMI）
  • [Number] [Adjective] Ways to [Goal]
    / 
    [Number] Mistakes to Avoid When [Task]
指南类（B2B分享量是其他格式的3倍——BuzzSumo，1000万篇LinkedIn文章）
  • How to [Task] Without [Painful Constraint]
    / 
    How to [Task] Even If [Limiting Belief]

The "putaclic léger" zone

“轻度标题党”区间

Slightly clickbait but honest — maximum tension with a real promise:
  • Replace "won't believe" → "surprised to learn": same curiosity, honest register
  • Add specificity: "I saved €500/month" beats "I saved money"
  • Add a constraint: "without quitting your job" creates the gap
  • Use "the real reason" or "what nobody tells you" — they signal a non-consensus view you must substantiate
  • Add a bracket 
    [Étude]
    , 
    [Template]
    , 
    [Benchmark]
    to signal payoff type
FR — putaclic carries stronger pejorative weight. Apply the 70/30 rule: 70% information, 30% mystery. Tension and contradiction work; faux drama ("J'arrête tout 😱") destroys credibility immediately in French professional contexts. Canonical FR structure: "Tout le monde vous dit de faire A. Voici pourquoi je fais B." Formulas saturated on LinkedIn FR in 2025+: "J'ai fait une erreur, voici ce que j'ai appris", "J'arrête [plateforme]".
略带标题党但诚实——在真实承诺的前提下最大化张力：
  • 把“won't believe”替换成“surprised to learn”：同样的好奇心，更诚实的风格
  • 添加具体细节：“I saved €500/month”比“I saved money”效果更好
  • 添加限制条件：“without quitting your job”能制造缺口
  • 使用“the real reason”或“what nobody tells you”——它们暗示非共识观点，你必须证实
  • 添加括号标注
    [Étude]
    、
    [Template]
    、
    [Benchmark]
    来表明收益类型
法语环境——标题党带有更强的贬义。遵循70/30规则：70%信息，30%神秘感。张力和矛盾有效；虚假戏剧性（比如“J'arrête tout 😱”）会立即破坏法语职场环境中的可信度。标准法语结构：“Tout le monde vous dit de faire A. Voici pourquoi je fais B.”2025年LinkedIn法语区饱和公式：“J'ai fait une erreur, voici ce que j'ai appris”、“J'arrête [plateforme]”。

Platform constraints

平台限制

PlatformMax displayKey rule
Blog / SEO50–60 charsFront-load keyword. Power words reduce CTR by ~14% in search.
LinkedIn article150 chars hardUnder 100 for clean display.
LinkedIn feed120–140 charsOpening line = the title. Engineer a "see more" click.
Newsletter subject30–50 charsFirst 30 chars carry the signal. Avoid "Newsletter" (−18.7% opens).
X / Twitter≤200 charsFirst tweet = the title. Must stand alone.
平台最大显示长度核心规则
博客/SEO50–60字符前置关键词。强力词汇会降低约14%的搜索点击率。
LinkedIn文章最多150字符控制在100字符以内以保证清晰显示。
LinkedIn动态120–140字符开头行即标题。设计“查看更多”的点击诱因。
通讯主题30–50字符前30字符是核心信号。避免使用“Newsletter”（打开率降低18.7%）。
X/Twitter≤200字符第一条推文即标题。必须独立成立。

Anti-patterns for titles

标题反模式

  • Generic superlatives without specificity: "The Ultimate Guide to X" — no gap
  • "You won't believe..." / "Vous n'allez pas y croire..." — overused, signals hype
  • Power words (Secret, Amazing, Ultimate) in SEO titles — signal hype, lower CTR
  • Titles that give away the full answer — if the title answers the question, there is no reason to click
  • Recycling saturated fingerprints: "I analyzed X things, here's what I found" — effective 2022–2023, saturating 2025–2026
  • 无具体内容的通用最高级：“The Ultimate Guide to X”——没有缺口
  • “You won't believe...” / “Vous n'allez pas y croire...”——过度使用，暗示夸张
  • SEO标题中的强力词汇（Secret、Amazing、Ultimate）——暗示夸张，降低点击率
  • 泄露完整答案的标题：如果标题已经回答了问题，读者就没有点击的理由
  • 过度使用饱和格式：“I analyzed X things, here's what I found”——2022–2023年有效，2025–2026年已饱和

Behavior for title requests

标题请求的操作流程

  1. Confirm topic, audience, language, platform
  2. Propose 3–5 titles across distinct formulas — not three variants of the same type
  3. Include at minimum: one curiosity/gap, one contrarian or data, one list or how-to
  4. Use the same output format as hooks: numbered options, wait for the pick
  5. After the pick, note what the title commits the article's opening to
For the full 58-formula catalog (EN + FR) with data citations, see 
references/title-formulas.md
.
  1. 确认主题、受众、语言、平台
  2. 提出3–5个不同公式的标题——不是同一类型的三个变体
  3. 至少包含：一个好奇心/缺口型、一个逆向或数据型、一个清单或指南型
  4. 使用与钩子相同的输出格式：编号选项，等待用户选择
  5. 用户选定后，说明该标题对文章开头的要求
如需完整的58种公式目录（英文+法文）及数据引用，请查看
references/title-formulas.md
。

Reference files (load only when needed)

参考文件（仅在需要时加载）

  • references/anti-patterns.md
    : extended cliché and AI-tell list.
  • references/anglophone-vs-francophone.md
    : register conventions, what does not translate.
  • references/formulas.md
    : 30 reusable templates (EN + FR) for quick template-based hooks.
  • references/title-formulas.md
    : 58-formula post title catalog (EN + FR), platform constraints table, data summary.
The catalog above is sufficient for most invocations. Load references only when the user asks for templates, deeper register guidance, or to sanity-check anti-patterns.
  • references/anti-patterns.md
    ：扩展的俗套和AI生成特征列表。
  • references/anglophone-vs-francophone.md
    ：风格惯例、不可直译内容。
  • references/formulas.md
    ：30个可复用模板（英文+法文），用于快速生成钩子。
  • references/title-formulas.md
    ：58种文章标题公式目录（英文+法文）、平台限制表、数据汇总。
上述目录足以应对大多数调用场景。仅当用户要求模板、更详细的风格指导或检查反模式时，才加载参考文件。

Integration with other ghostwriting skills

与其他代笔技能的集成

If invoked from inside another skill's workflow (e.g., 
technical-article-writer
reaches the hook step), produce options, ask, wait. Return the chosen hook to the parent skill for the rest of the article.
If the user has already drafted an opening and asks for "a better one", treat that draft as Option 0 (their current version) and propose 3 alternatives across different hooks. Do not silently discard their version.
如果从其他技能的工作流中触发（比如
technical-article-writer
进入钩子创作步骤），请生成选项、询问用户、等待回复。将选定的钩子返回给父技能，以便继续完成文章。
如果用户已经起草了开头并要求“更好的版本”，请将其草稿视为选项0（当前版本），并提出3种不同杠杆的替代方案。不要直接丢弃用户的版本。

Worked example

实操示例

User: "I'm writing a Substack piece on why most Go developers misuse generics. About 1500 words, intermediate Go devs, English."
Correct skill response:
undefined
用户：“我正在写一篇Substack文章，主题是为什么大多数Go开发者误用泛型。约1500字，受众是中级Go开发者，语言为英语。”
技能正确回复：
undefined

Hook options for: "Why Most Go Developers Misuse Generics"

Hook options for: "Why Most Go Developers Misuse Generics"

Option 1: Contrarian (prediction-break) A. Most articles on Go generics explain when to use them. This one explains when to delete them. B. Generics were the most requested Go feature for a decade. Three years in, the people who pushed hardest for them are the ones telling you to stop using them.
Option 2: Concrete detail (sensory + specific) A. samber/lo ships 87 generic helpers. Six of them account for 94 percent of the binary-size complaints. The other 81 are fine. B. A function called 
Map[T, U any]
looks innocent. Compile a project that imports it 40 times and your build time triples. Here is why.
Option 3: Direct problem (PAS) (reader-direct) A. Your Go code compiles slower since you adopted generics. The cause is not what the linter is telling you. B. If your team adopted generics last year and your build times doubled, you are not alone. The diagnosis is more boring than the fix.
Which? Reply with letter combination (e.g., "1B") or "more" for different techniques.

Then wait for the pick. Do not write the rest of the article until the user has chosen.
Option 1: Contrarian (prediction-break) A. Most articles on Go generics explain when to use them. This one explains when to delete them. B. Generics were the most requested Go feature for a decade. Three years in, the people who pushed hardest for them are the ones telling you to stop using them.
Option 2: Concrete detail (sensory + specific) A. samber/lo ships 87 generic helpers. Six of them account for 94 percent of the binary-size complaints. The other 81 are fine. B. A function called 
Map[T, U any]
looks innocent. Compile a project that imports it 40 times and your build time triples. Here is why.
Option 3: Direct problem (PAS) (reader-direct) A. Your Go code compiles slower since you adopted generics. The cause is not what the linter is telling you. B. If your team adopted generics last year and your build times doubled, you are not alone. The diagnosis is more boring than the fix.
Which? Reply with letter combination (e.g., "1B") or "more" for different techniques.

然后等待用户选择。在用户选定前，不要撰写文章的其余部分。