Token导航 LogoToken导航

Agent Skill · 研究检索

marketing-hooks(营销挂钩)

marketing-hooks 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

marketing-hooks 是研究检索类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为只读;许可证为 MIT-0,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 17,138。

总安装 6,774周安装 288GitHub Stars 1下载 2,373核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
0.1.0

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:marketing-hooks(营销挂钩)
来源仓库:https://github.com/yuqingnicole/marketing-hooks
安装命令:
openclaw skills install marketing-hooks
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install marketing-hooks

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

营销挂钩技能基于谜题驱动模型生成吸引眼球的社交媒体文案与内容结构模板。适用于内容创作者、运营人员提升帖子点击率与用户参与度。通过 clawhub 安装后,在 OpenClaw 中部署,支持输入主题关键词自动生成初稿。使用前应评估生成内容的合规性,避免误导性或侵权表述出现。

适合效率类 Agent 嵌入至内容管理系统,实现批量草稿生产与优化建议。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install marketing-hooks;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

marketing-hooks 怎么安装?

在终端执行 openclaw skills install marketing-hooks。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

marketing-hooks 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

marketing-hooks 需要什么权限?

权限级别是只读。该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

marketing-hooks 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

marketing-hooks 用的人多吗?

公开数据是累计安装 6,774 次,近一周安装 288 次,下载 2,373 次,GitHub 1 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

marketing-hooks 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。
  • 标注为只读权限,试用阶段不容易误改本地文件。

先想清楚的地方

  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
marketing-hooks

营销挂钩 · 本页

6,7741只读MIT-0通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

name
marketing-hooks
description
Create compelling marketing hooks and content structures using the Puzzle-Driven Model. Use when creating social media posts, newsletters, video scripts, or any marketing content that needs to capture attention. Triggers include "write a hook", "create a title", "make this more engaging", "content structure", "wow factor", or when analyzing why content performs well.

Marketing Hooks - Puzzle-Driven Content Model

Create attention-grabbing marketing content using proven psychological frameworks.

Core Principle: Questions Drive Engagement

Good content is driven by good questions. Good questions are puzzles.

When your opening creates a puzzle, the audience's brain automatically tries to solve it. This creates engagement and retention.

Four Question Types

1. How-Type (Method)

Pattern: "How to [achieve desired outcome]?"

Examples:

  • "How to become wealthy quickly?"
  • "From white-label to brand?"
  • "How to start cross-border e-commerce?"
  • "How to 10x your productivity with AI?"

Best for: Tutorials, guides, practical content

2. Why-Type (Reason)

Pattern: "Why [surprising fact/outcome]?"

Examples:

  • "Why did they become CEO?"
  • "Why 90% of AI tools fail?"
  • "Why top creators don't post daily?"

Best for: Analysis, insights, thought leadership

3. What-Type (Definition)

Pattern: "What is [concept/term]?"

Examples:

  • "What is supply-side reform?"
  • "What is CPS?"
  • "What is Product-Market Fit?"

Best for: Educational content, explainers

4. WoW-Type (Extreme Case) ⭐ HIGHEST ENGAGEMENT

Pattern: Create probability-defying extreme cases

Core insight: WoW = Manufacturing extremely low-probability extreme cases

Formula:

  • Discover extremes → Get traffic
  • Become extreme → Build brand

Examples:

  • "Gen-Z makes $100K/month with AI"
  • "10K followers in 3 days - the secret"
  • "I replaced my entire team with one tool"
  • "This 19-year-old built a $1M business with ChatGPT"

Why it works: Brain automatically fills in missing logic, creating engagement stickiness.

Visual/Info/Case combinations:

  • Extreme visual + surprising data
  • Impossible timeline + real results
  • Tiny input + massive output

Content Tree Structure

Every piece of content should follow this hierarchy:

Question (Root)
    ↓
Viewpoint (Trunk)
    ↓
Reasoning (Branches)
    ↓
Evidence (Leaves)

Gan Method Content Model

Structure:

  1. Core viewpoint (Trunk)
  2. Sub-arguments (Branches)
  3. Evidence (Leaves) - Must be:

- Unexpected - Memorable - Vivid

Quality check: A good 15-min video or well-evidenced article = a tree with lush branches and leaves.

Platform-Specific Applications

Xiaohongshu (Little Red Book)

Best formats:

  • WoW-type titles: "Gen-Z AI tools $100K/month"
  • How-type for tutorials: "5 AI tools for 10x efficiency"

Structure:

  1. Hook (5-10s): WoW statement
  2. Pain point (10-15s): "Are you also facing..."
  3. Solution (120-180s): Step-by-step with demos
  4. Results (20-30s): Proof
  5. CTA (5-10s): "Follow for more"

X (Twitter)

Best formats:

  • Start with puzzle: "Do you know why 90% of people fail with AI tools?"
  • Then reveal answer in thread

Structure:

  • Tweet 1: Puzzle/Hook
  • Tweet 2-4: Reasoning + Evidence
  • Final tweet: Conclusion + CTA

Newsletter

Best formats:

  • Question (title) → Viewpoint (intro) → Reasoning (body) → Evidence (examples)

Structure:

  • Subject line: WoW or Why-type
  • Opening: Establish puzzle
  • Body: Tree structure (trunk → branches → leaves)
  • Closing: Actionable takeaway

Hook Generation Workflow

When asked to create hooks:

  1. Identify the core message

- What's the main point? - What outcome does the audience want?

  1. Choose question type

- Practical guide? → How - Surprising insight? → Why - New concept? → What - Extreme case? → WoW (prioritize this!)

  1. Apply WoW amplification

- Can you make it more extreme? - Can you add surprising numbers? - Can you create contrast (small input → big output)?

  1. Test the puzzle

- Does it make you think "why?" or "how?" - Does it create information gap? - Would you click it?

Examples Library

See EXAMPLES.md for detailed case studies and templates.

Quality Checklist

Before finalizing content:

  • [ ] Does the opening create a puzzle?
  • [ ] Is there a clear tree structure (root → trunk → branches → leaves)?
  • [ ] Are the evidence points unexpected, memorable, and vivid?
  • [ ] For WoW-type: Is the case extreme enough?
  • [ ] Does it make the brain want to "fill in the gaps"?

References

  • "Writing is a Craft" - "Questions are the engine of articles"
  • Gan Method - Content tree structure
  • WoW Psychology - Extreme case engagement mechanics

平台分布

OpenClaw

84.04%
按下载量换算1,994

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills