Token导航 LogoToken导航

Agent Skill · 研究检索

content-repurposer(内容再利用者)

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。

content-repurposer 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,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 59,221。

总安装 392周安装 16GitHub Stars 4下载 127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:content-repurposer(内容再利用者)
来源仓库:https://github.com/cdeistopened/opened-vault
仓库路径:skills/content-repurposer
安装命令:
npx skills add https://github.com/cdeistopened/opened-vault --skill content-repurposer
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/cdeistopened/opened-vault --skill content-repurposer

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。

用于将源内容转化为平台优化的社交媒体帖子,适配不同渠道的表达形式。适合品牌内容再创作、跨平台传播和多格式文案生成场景。使用时应加载 opened-identity 上下文,保持品牌身份一致性。通过 GitHub 安装,支持主流宿主环境,需注意品牌语境和发布平台特性。

content-repurposer 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/content-repurposer

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/cdeistopened/opened-vault --skill content-repu…;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

content-repurposer 怎么安装?

在终端执行 npx skills add https://github.com/cdeistopened/opened-vault --skill content-repurposer。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

content-repurposer 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。

content-repurposer 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。

content-repurposer 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。

content-repurposer 用的人多吗?

公开数据是累计安装 392 次,近一周安装 16 次,下载 127 次,GitHub 4 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。

content-repurposer 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

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。
  • 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。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 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
content-repurposer

内容再利用者 · 本页

3924只读未明确有提醒

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 Repurposer

Transform source content into platform-optimized posts using framework fitting.

Core job: Take content from any source → understand context → extract snippets → match to templates → optimize for each platform.


Context Awareness (Critical)

Brand Context

OpenEd is a company, not an individual.

AspectImplication
VoiceBrand account, not personal ("we" not "I")
Authority125,000+ families, 9 states, tuition-free resources
ToneHelpful expert, not salesy. Permission-giving, not preachy.
IdentityPro-child, not anti-school. Mix and match philosophy.

Always load: opened-identity for full brand context.

Source Context

The source determines voice and framing:

SourceVoiceFramingTagging
Podcast guestThird personMake THEM look goodTag guest on all platforms
NewsletterFirst person plural"We found..." "Our team..."Tag mentioned sources
Archive articleAuthorityFresh angle, current relevanceTag original author if external
Staff insightFirst person pluralShare the discoveryInternal attribution

Podcast guest example:

❌ "I learned so much from Ken Danford..."
✅ "Ken Danford (@northstarteens) explains why..."
✅ "@kendanford's insight on self-directed learning..."

The Framework Fitting Process

Step 1: Identify Source Type

What am I repurposing?
├─ Podcast episode → Load guest context, prepare @handles
├─ Newsletter → Extract TTT segments (Thought/Trend/Tool)
├─ Article → Extract key insights, stats, quotes
├─ Archive content → Find fresh angle, check current relevance
└─ Raw insight → Match to template directly

Step 2: Extract Snippets by Type

From any source, identify standalone pieces:

Snippet TypeWhat to ExtractBest Templates
hot_takeOpinion that stands aloneContrarian, Paradox Hook, Call BS
statData point + interpretationAuthority, Commentary, Data Story
how_toActionable adviceList, Thread, Tips, Carousel
quoteMemorable line from sourceQuote + Commentary
storyTransformation arcStory, Day-in-Life

Step 3: Check Nearbound

Before drafting, check for taggable people:

1. Identify all names mentioned
2. Search: Studio/Nearbound Pipeline/people/
3. If found → Get @handle for target platform
4. If not found → Note for future profile creation

Priority tags: Podcast guests, quoted experts, tool/curriculum founders.

Step 4: Route to Platforms

SnippetLinkedInXInstagramFacebook
hot_takeContrarianParadox HookQuote cardAgree/Disagree
statAuthorityCommentaryCarouselQuestion
how_toListThreadCarouselFill-blank
quoteQuote + Hot TakeCommentaryQuote cardValidation
storyTransformationThreadCarouselStory post

Step 5: Apply Platform Constraints

PlatformLengthHashtagsLinksTone
LinkedIn200-500 words3-5OK in bodyProfessional authority
X70-100 chars0-2In replyPunchy, scroll-stopping
Instagram30-150 chars5-10Bio onlyCasual, visual-first
Facebook10-40 words0Comments onlyConversational, questions
TikTok50-150 chars3-5Bio onlyCasual, trend-aware

Template Quick Reference

Load full templates from: .claude/skills/text-content/references/

LinkedIn (6 categories)

  • linkedin/engagement.md - Polls, Agree/Disagree (16 templates)
  • linkedin/story.md - Transformations, Values (24 templates)
  • linkedin/list.md - Tips, Frameworks (17 templates)
  • linkedin/contrarian.md - Hot takes, Rants (20 templates)
  • linkedin/authority.md - Quotes, How-to (26 templates)
  • linkedin/community.md - Shoutouts, Introductions (15 templates)

X/Twitter

  • templates/post-structures.md - 100+ formats
  • templates/one-liners.md - 12 punchy patterns

Instagram/Facebook

  • platforms/instagram-captions.md
  • platforms/facebook.md

Output Format

For each platform, produce:

## [Platform] Post

**Template:** [Template name used]
**Source:** [What this came from]

---

[The actual post content]

---

**Tags:** @handle1 @handle2
**Hashtags:** #tag1 #tag2 (platform-appropriate count)
**Visual direction:** [If needed - quote card, carousel, etc.]

Dimension Handling (Visual Assets)

When source includes a 16:9 infographic or thumbnail:

Original (16:9 - article thumbnail)
    │
    ├─→ Instagram Feed (4:5)
    │   Prompt: "Adapt this infographic to 4:5 portrait,
    │            prioritize [key element], crop or extend background"
    │
    ├─→ Instagram/Facebook Square (1:1)
    │   Prompt: "Center the key message, square crop"
    │
    └─→ Stories/Reels Cover (9:16)
        Prompt: "Vertical adaptation, text readable on mobile"

Use: nano-banana-image-generator with re-input of original image.


Voice Constraints (Always Apply)

From ai-tells - HARD BLOCKS:

❌ No correlative constructions:

  • "X isn't just Y - it's Z"
  • "It's not about X, it's about Y"

❌ No forbidden words:

  • delve, comprehensive, crucial, leverage, landscape, navigate, foster, facilitate, realm, paradigm, embark, journey, tapestry, myriad, multifaceted, seamless, cutting-edge

❌ No setup phrases:

  • "The best part?", "What if I told you", "Here's the thing", "Let's be honest"

❌ No staccato patterns:

  • "No fluff. No filler. Just results."

Dash usage: Hyphens with spaces - like this. Never em dashes.


Batch Processing Mode

When repurposing multiple pieces (e.g., full newsletter → social):

1. Read entire source
2. Extract ALL snippets (usually 3-5 per newsletter)
3. For each snippet:
   ├─ Identify type (hot_take, stat, how_to, quote, story)
   ├─ Check nearbound for tags
   └─ Generate 2-3 platform drafts
4. Output organized by platform:
   ├─ LinkedIn (2-3 posts)
   ├─ X (2-3 posts)
   ├─ Instagram (1-2 posts + visual direction)
   └─ Facebook (1-2 posts)

Expected output: 6-9 total posts per newsletter/article.


When to Use This Skill

Use content-repurposer when:

  • Newsletter is complete → need social posts
  • Podcast is published → need clip promotions
  • Article is live → need distribution posts
  • Archive content identified → need fresh framing
  • Any content needs multi-platform adaptation

Don't use for:

  • Original content creation (use specific content skills)
  • Video production (use video-caption-creation)
  • Newsletter writing (use opened-daily-newsletter-writer)

Related Skills

  • text-content - Full template library (this skill routes TO those templates)
  • newsletter-to-social - Automated newsletter → social (uses this methodology)
  • opened-identity - Brand voice (always load)
  • ai-tells - Hard blocks (always apply)
  • nano-banana-image-generator - Visual adaptations

Example: Podcast Guest Repurposing

Source: Podcast with Ken Danford about self-directed learning

Step 1: Identify source type → Podcast guest

Step 2: Extract snippets:

  • hot_take: "The best thing a school can do is get out of the way"
  • stat: "75% of North Star teens go to college - without transcripts"
  • story: Ken quit teaching to prove schools are optional

Step 3: Check nearbound → Ken Danford profile exists, @kendanford on X

Step 4: Draft posts:

LinkedIn (Authority):

Ken Danford spent 10 years as a public school teacher.

Then he quit to prove schools are optional.

His self-directed learning center, North Star, has helped hundreds of teens:
- No grades
- No transcripts
- 75% college attendance rate

The secret? "Get out of the way and let them lead."

Full conversation on the OpenEd podcast: [link]

#SelfDirectedLearning #AlternativeEducation #Homeschooling

X:

"The best thing a school can do is get out of the way."

@kendanford quit teaching to prove it. 75% of his students go to college - without transcripts.

[link to episode]

Instagram (Quote card direction):

Visual: Quote card with Ken's face + "Get out of the way and let them lead."
Caption: When a 10-year teacher quits to prove schools are optional... and it works. Link in bio.

#homeschool #selfdirectedlearning #alternativeeducation #unschooling #parentingtips

*Framework fitting is the core technique. This skill specializes in it.*

平台分布

Codex

34.13%
按下载量换算43

Claude

33.07%
按下载量换算42

Cursor

17.04%
按下载量换算22

Gemini CLI

8.98%
按下载量换算11

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

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。来源安全扫描存在提醒或未通过项,安装前请自行复核。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