Token导航 LogoToken导航

Agent Skill · 研究检索

create-content(创建内容)

用于辅助文档、README、Markdown、说明文和内容稿件的整理与改写。它适合让 Agent 提炼结构、补齐章节、统一术语、检查链接或把零散材料整理成可读文档。使用时应保留项目已有事实、命令和路径,不要把未确认的信息写成确定结论;涉及对外文案时,还需要控制语气,避免过度营销或夸大能力。

create-content 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为只读;许可证信息还不明确,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 65,896。

总安装 272周安装 11GitHub Stars 589下载 85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
unknown

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:create-content(创建内容)
来源仓库:https://github.com/sundial-org/awesome-openclaw-skills
仓库路径:skills/create-content
安装命令:
npx skills add https://github.com/sundial-org/awesome-openclaw-skills --skill create-content
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 npx skills 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。

skills.shnpx skills
npx skills add https://github.com/sundial-org/awesome-openclaw-skills --skill create-content

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。

create-content 用于辅助文档、README、Markdown 和内容稿件的整理与改写,适合提炼结构和统一术语。适用于内容稿件、说明文和对外文案的结构化梳理与优化。通过 npx skills add 命令从指定 GitHub 仓库安装并使用。使用时应保留项目已有事实,避免过度营销或夸大能力。

适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

来源仓库内路径:skills/create-content

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 Codex、Claude、Cursor、Gemini CLI;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 npx skills add https://github.com/sundial-org/awesome-openclaw-skills --skill cr…;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

create-content 怎么安装?

在终端执行 npx skills add https://github.com/sundial-org/awesome-openclaw-skills --skill create-conte…。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

create-content 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。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。

create-content 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。

create-content 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

create-content 用的人多吗?

公开数据是累计安装 272 次,近一周安装 11 次,下载 85 次,GitHub 589 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。

create-content 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

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 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

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 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 Codex,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。
  • 只想先试试效果、不希望它改动本地文件的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 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
create-content

创建内容 · 本页

272589只读未明确通过

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

Content Creator

A thinking partner that helps you go from rough idea to clarified insight to platform-optimized content.

Philosophy: Great content comes from clear thinking. We explore first, draft second.

Usage

/create-content [rough idea, topic, or "help me figure out what to post"]

Phase 1: Thinking Partner Mode

Before drafting anything, help clarify the idea.

If user has a rough idea:

Ask 2-3 questions to sharpen it:

  • "What's the specific insight or observation here?"
  • "What made you think of this? What triggered it?"
  • "Who needs to hear this? Why would they care?"
  • "What's the counterintuitive part? What surprises people?"
  • "Do you have a specific example or number to anchor this?"

If user says "help me figure out what to post":

  1. Search for recent notes, journal entries, sessions
  2. Look for: observations, breakthroughs, experiments, patterns noticed
  3. Present 2-3 potential angles and ask which resonates

Stay in exploration until:

  • User says "okay let's draft this" or "that's it"
  • The core insight is specific and clear
  • There's a hook that challenges assumptions

Phase 2: Voice Guidelines

DO:

  • Short sentences. Like texting.
  • Observations over wisdom. Show, don't preach.
  • Specific numbers. "$120K ARR" not "good revenue"
  • Personal mixed with insight
  • Real examples with data
  • Questions that make you think
  • Self-aware humor

DON'T:

  • Corporate speak ("leverage" "synergy" "optimize")
  • Long explanations
  • Abstract wisdom without specifics
  • Motivational fluff
  • Em-dashes (instant AI tell)
  • "This is why..." openings
  • Any sentence over 20 words

Red Flags (rewrite if present):

  • Em-dashes (—)
  • "This is why..."
  • "The key is..."
  • "In today's world..."
  • Wisdom without specifics
  • Sentences over 20 words

Voice Examples (Study These)

@levelsio Style (Raw Observations)

"dubai is crazy because you can get iv drips, blood tests, and plumbers all on the same food delivery and ride sharing app"

What makes it work: Simple observation, relatable, slightly absurd. No call to action, just sharing reality.

@marclou Style (Authentic + Celebrates Others)

"SOLD 1. David vibe-coded the project in 1 week 2. Launch went viral on LinkedIn 3. Made $130/month 4. Got acquired for $3500"

What makes it work: Celebrates others' wins. Specific numbers. Simple format.

@bryan_johnson Style (Mission-Obsessed + Data)

"+ 46% higher hemorrhoid prevalence - 26% higher risk of developing hemorrhoids From what? Smartphone while on the toilet"

What makes it work: Shocking data + unexpected humor. Bold.


Voice Calibration Test

Before finalizing any draft, check:

TOO AI:

"Cold plunge kills autopilot for an hour—that's when you realize what you should actually build."

REAL VOICE:

"been coding while alternating cold plunge and sauna. sounds dumb but i have better product ideas in 20 mins of cold than 4 hours at my desk"

The difference: No em-dashes. Specific detail (20 mins vs 4 hours). Self-aware ("sounds dumb"). Shows the lifestyle, doesn't explain it.


Phase 3: Platform-Specific Drafting

For X (Twitter)

Viral Mechanics:

  • Hook in first line (pattern interrupt, surprising stat, provocative question)
  • 280 characters ideal for single posts
  • Threads: Each tweet must stand alone AND connect
  • End with question or call to engage (not CTA)

Formats that work:

  1. Observation post: "noticed [specific thing]. [insight]."
  2. Experiment post: "tried [thing]. result: [data]. [what it means]"
  3. Contrarian take: "[common belief]. actually: [your take]. here's why."
  4. List post: "X things I learned from [specific experience]:"
  5. Question post: "[provocative question]? [your angle in 1 sentence]"

Thread structure:

  • Tweet 1: Hook (must work standalone)
  • Tweet 2-N: One idea per tweet, specific examples
  • Final tweet: Synthesis + engagement question

For LinkedIn

Viral Mechanics:

  • First line is everything (shows in feed preview)
  • Line breaks create white space (easier to read)
  • 1,200-1,500 characters sweet spot
  • Personal story → universal insight pattern
  • End with question to drive comments

Format:

[Hook line - surprising or contrarian]

[2-3 short paragraphs with the story/insight]

[Specific example or data point]

[Universal takeaway in 1 sentence]

[Question for engagement]

Phase 4: Draft & Refine

  1. Draft 2-3 versions for the chosen platform
  2. Run voice check on each:

- Is it casual enough to be a text message? - Specific OR observation (not vague wisdom)? - No em-dashes?

  1. Present options with notes on what makes each one work
  2. Refine based on feedback until user is happy

Quick Commands

User can shortcut the process:

  • "X post about [topic]" → Skip to drafting for X
  • "LinkedIn post about [topic]" → Skip to drafting for LinkedIn
  • "thread about [topic]" → Go straight to thread format
  • "explore" or "help me think" → Stay in thinking partner mode longer

Remember

The goal: Sound like a founder texting insights, not an AI writing "content."

Great content = clear thinking + specific examples + authentic voice.

If the idea isn't clear yet, keep exploring. Don't rush to draft.

平台分布

Codex

31.38%
按下载量换算27

Claude

31.02%
按下载量换算26

Cursor

19.99%
按下载量换算17

Gemini CLI

8.6%
按下载量换算7

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

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