Token导航 LogoToken导航

Agent Skill · 待分类

social-content-creation(社交内容创作)

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。

social-content-creation 是待分类类 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,按安装量在待分类类的 11,209 个 Skill 里排第 9,612。

总安装 235周安装 10GitHub Stars 9下载 82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:social-content-creation(社交内容创作)
来源仓库:https://github.com/cdeistopened/skill-stack
仓库路径:skills/social-content-creation
安装命令:
npx skills add https://github.com/cdeistopened/skill-stack --skill social-content-creation
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/skill-stack --skill social-content-creation

这个 Skill 是做什么的

这个 Skill 的分类还没最终确定,用法以来源仓库的说明为准。

social-content-creation 用于辅助文档、README、Markdown、说明文和内容稿件的整理与改写。适用于社交媒体文案策划、品牌故事撰写与传播材料组织。支持结构提炼、术语统一与零散材料整合成可读文档。使用时应保留项目已有事实与路径,避免将未确认信息写成确定结论。

涉及对外文案时需控制语气,防止过度营销或夸大能力。

来源仓库内路径:skills/social-content-creation

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/skill-stack --skill social-conten…;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

social-content-creation 怎么安装?

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

social-content-creation 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。

social-content-creation 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。

social-content-creation 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。

social-content-creation 用的人多吗?

公开数据是累计安装 235 次,近一周安装 10 次,下载 82 次,GitHub 9 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。

social-content-creation 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。
  • 需要处理待分类类任务、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。

待分类类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
infsh-cli

infsh CLI · Codex、Claude

4,374,576382权限未标注未明确未收录
ai-video-generation

AI 视频生成 · Codex、Claude

4,313,376382调用外部服务未明确有提醒
lark-openapi-explorer

云雀 OpenAPI 浏览器 · Codex、Claude

2,064,1209,011需要联网未明确有提醒
lark-workflow-standup-report

Lark 工作流程站立报告 · Codex、Claude

2,041,6328,978需要联网未明确通过
use-my-browser

使用我的浏览器 · Codex、Claude

1,848,49140操作浏览器未明确未通过
social-content-creation

社交内容创作 · 本页

2359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

Social Content Creator

Transform any source material into high-performing social media posts using the framework fitting method: extract standalone concepts, match to proven templates, generate volume, select best, execute, and proliferate winners.

Purpose

Content creation isn't about constantly coming up with new ideas. It's about turning 1 idea into 1000 variations using proven frameworks.

Core Philosophy: Good content follows proven formats. The skill is in framework fitting - matching concepts to the delivery mechanisms that amplify them best.

When to Use This Skill

  • Repurposing newsletter content into social posts
  • Transforming long-form articles into multi-platform campaigns
  • Creating LinkedIn posts from any concept or story
  • Generating multiple post options quickly
  • Multiplying high-performing posts using SCAMPER or desire reframing

Not for: Short-form video captions (use video-caption-creation), podcast clip selection (separate workflow)


The Framework Fitting Method

Core Principle

Good content follows proven formats/frameworks/templates. The skill is in framework fitting - matching concepts to the delivery mechanisms that amplify them best.

Critical Rule: Avoid First-Match Bias

Do NOT default to the first template that seems to fit.

"Analyze, Then Select" - Treat template selection as a distinct step. Consider ALL applicable frameworks before selecting.

The best framework might not be the obvious one.


The 4-Phase Workflow

Phase 1: Extract Standalone Concepts

Goal: Identify the most shareable "standalone" concepts from source material

Step 1: Read Source Material Thoroughly
  • Full newsletter, article, or podcast transcript
  • Look for bundles of ideas within their context
  • Self-contained insights that can live independently
Step 2: Extract Concepts

For each standalone concept, capture:

  • Core Insight (1-2 sentences)
  • Context Needed (what background info is required?)
  • Emotion/Value (what does this unlock for the reader?)
  • Human Desire(s) (which of the 8 desires does this address?)
  • Potential Angles (initial framework ideas)

How Many Concepts to Extract:

  • Newsletter segment: 1-3 concepts per segment
  • Full newsletter: 5-10 concepts total
  • Long-form article: 5-15 concepts
  • Podcast episode: 3-8 concepts
Step 3: Create Concepts Document
# [Source] - Extracted Concepts

## Concept 1: [Title]
**Core Insight:** [1-2 sentences]
**Context:** [What background is needed]
**Emotion/Value:** [What this unlocks for reader]
**Human Desires:** [Which of the 8]
**Potential Angles:** [Framework ideas]

---

## Concept 2: [Title]
[Repeat structure...]

The 8 Human Desires

Reference these when extracting concepts:

  1. Safety of Tribe - Belonging, community, fitting in
  2. Survival & Success - Progress, achievement, security
  3. Life Enjoyment - Pleasure, experiences, freedom
  4. Social Acceptance - Status, respect, recognition
  5. Sexual Companionship - Connection, attraction (often not applicable)
  6. Comfort & Clarity - Simplicity, ease, certainty
  7. Freedom From Fear - Safety, protection, security
  8. Perceived Status - Respect, authority, expertise

Phase 2: Framework Matching (THE CRITICAL PHASE)

Goal: Match extracted concepts to best-fit frameworks using volume generation → selection

The Framework Matching Process

For each concept:

  1. Brainstorm First: How COULD this concept be presented?

- Think beyond the obvious - Consider different platforms - Imagine different emotional angles

  1. Review ALL Applicable Frameworks:

- LinkedIn templates (86+ options) - Text post structures (100+ options) - Don't stop at first match - Consider at least 5-10 frameworks minimum

  1. Test the Fit:

- Does concept naturally fit framework structure? - What would make this concept fit BETTER in another framework? - Which framework amplifies the core emotion/value most?

  1. Generate Volume:

- Quick social plan: 2-4 framework matches per concept - High-volume campaign: 10-20 framework matches per concept

  1. Select Best:

- Which frameworks amplify the concept most? - Which platforms matter most? - Which formats are most efficient to execute?

Framework Categories

LinkedIn Templates (see references/linkedin-frameworks.md):

  • 86+ templates from LinkedIn swipe files
  • 50+ templates from proven creators
  • Best for: Thought leadership, educational content, stories

Text Post Structures (see references/post-structures.md):

  • 100+ proven tweet/post frameworks
  • Best for: X/Twitter, Facebook, Instagram captions
  • Categories: Comparisons, Lists, Challenges, Observations, Analogies
Document Framework Matches
## Concept: "[Title]"

### Framework Matches Generated (10 total):

**1. Pattern Recognition Template**
- **Fit Assessment:** ✅ STRONG
- **Why it fits:** [Explanation]
- **Structure:** "I've noticed a pattern: [Common Belief] isn't what leads to [Success]... It's [Core Insight]"
- **Platform:** LinkedIn
- **Execution Preview:** "[Draft of how it would read]"

**2. Contrarian Truth Template**
- **Fit Assessment:** ✅ STRONG / ⚠️ MODERATE / ❌ WEAK
- **Why it fits:** [Explanation]
[Continue...]

### SELECTED FOR EXECUTION (Top 3):
1. [Framework] - [Rationale]
2. [Framework] - [Rationale]
3. [Framework] - [Rationale]

Quality Check:

  • Generated at least 5-10 framework options per concept
  • Assessed fit strength (STRONG/MODERATE/WEAK)
  • Explained WHY concept fits each framework
  • Selected best 2-4 with clear rationale
  • Considered platform appropriateness
  • Avoided first-match bias

Phase 3: Execute Selected Frameworks

Goal: Create polished, platform-optimized posts from selected frameworks

Determine Output Scope
  • Quick Social Plan (1-3 posts): Best framework match per concept
  • Multi-Platform Campaign (5-10 posts): Multiple frameworks across concepts
  • High-Volume Repurposing (10-20 posts): 2-3 variations per framework
Execute Each Framework
  1. Load Framework Structure - Reference the specific template
  2. Fill with Concept Details - Insert insight, context, examples
  3. Apply Voice - Match brand or individual voice style
  4. Platform-Specific Optimization - Apply rules below
  5. Generate Variations - Create 1-3 variations of same framework
Platform-Specific Guidelines

LinkedIn:

  • Length: Long-form value-dense posts outperform (200-500 words)
  • Links: Always in comments (not main post)
  • Hashtags: 3-5 maximum, minimal
  • Voice: Professional yet conversational, thought leadership
  • Hook: First 2 lines critical (shows before "see more")
  • Structure: Hook → Body → Insight → Question or CTA
  • Tagging: Tag mentioned people/orgs (increases reach)

X (Twitter):

  • Length: 70-100 characters optimal (despite 280 limit)
  • Hashtags: 1-2 maximum
  • Voice: Witty, concise, to-the-point
  • Threads: Use for longer concepts
  • Current strategy: Reply game important for growth

Facebook:

  • Length: 40-140 characters optimal
  • Links: NO external links in main post (kills reach)
  • Hashtags: NO hashtags (Facebook doesn't reward them)
  • Voice: Informal, playful, personable
  • Format: Question posts with images perform best

Instagram:

  • Length: 30-125 characters for feed captions
  • Hashtags: 5-10 relevant hashtags
  • Voice: Friendly, authentic, micro-story format
  • Formats: Reels outperform static, carousels for educational
Post Output Format
## POST 1: [Framework Name]

[Full post text]

---

**Framework:** [Template used]
**Voice:** [Brand/Individual]
**Word Count:** [Number]
**Hashtags:** [If applicable]
**Platform:** [Target platform]
**Link in Comments:** [URL if applicable]
**Tags:** [People/orgs to tag]
Quality Check

Voice Consistency:

  • Sounds like a real person, not corporate
  • Would I say this to a friend?
  • Respects reader intelligence

Platform Compliance:

  • Adheres to character limits
  • Hashtags appropriate for platform
  • Links placed correctly

Framework Fidelity:

  • Actually follows selected framework structure
  • Doesn't drift into generic post
  • Template name documented

Strategic Alignment:

  • Would target audience stop scrolling?
  • Creates curiosity/emotion/value in first 2 seconds?
  • Clear what action reader should take?

Phase 4: Proliferate Winners (OPTIONAL)

Goal: Multiply high-performing posts using SCAMPER, Human Desires, and Vision reframing

When to use:

  • Monthly: Study top performers and multiply them
  • Pre-posting: Generate many options, select best
  • Content drought: Multiply existing winners
Method 1: SCAMPER Variations

Generate 7 variations from one post (one per letter):

S - Substitute

  • Replace main idea with adjacent concept
  • Swap the method for accomplishing benefit
  • Change examples while keeping structure

C - Combine

  • Merge with personal experience or story
  • Combine two concepts for more power
  • Add reply to someone else's post as your own content

A - Adapt

  • Expand one-liner into full post or thread
  • Add "missing piece" not in original
  • Adapt to different platform

M - Modify

  • Magnify specific pain points
  • Use same idea in different format
  • Make more punchy by removing words

P - Purpose (Put to other use)

  • Angle for different audience segment
  • Different benefit/pain point this solves

E - Eliminate

  • Remove words to make punchier
  • Simplify complex explanation
  • Cut nuance and uncertainty

R - Reverse

  • Flip the idea completely
  • "With X you get Y" → "Without X you stay stuck with Z"
  • Positive framing → negative (or vice versa)

SCAMPER Example:

Original:

"The most successful founders don't start with funding. They start with customers."

S - Substitute:

"The most successful founders don't start with a perfect product. They start with a painful problem."

C - Combine:

"I spent years thinking I needed funding to start. Then I talked to 3 founders who bootstrapped to $10M. The most successful don't start with funding. They start with customers."

R - Reverse:

"Without customers, your funding means nothing. Ask any startup that raised $50M and still failed."
Method 2: Human Desires Reframing

Spin concept through 8 desire lenses:

Example concept: "Customers over funding"

Safety of Tribe:

"You don't need VCs to validate your idea. You need customers who get it. Build your tribe first."

Survival & Success:

"The path to success isn't through investors. Every successful founder knows: customers first, funding later."

Comfort & Clarity:

"Starting a company feels overwhelming. Here's what brings clarity: successful founders don't start with funding. They start with customers. Everything else follows."

Freedom From Fear:

"The fear of 'not having enough runway' stops more companies than lack of funding. Get customers first. The fear disappears."
Proliferation Output

Create file: [Post Title]_Proliferations.md

Include:

  • 7 SCAMPER variations
  • 6-8 Human Desire reframes
  • Total: 13-15 new variations from 1 original

Output Files & Organization

Standard Output Structure

[Working Folder]/
├── [Source]_Concepts.md              # Phase 1
├── [Source]_Framework_Matches.md     # Phase 2
├── [Source]_Social_Posts.md          # Phase 3
└── [Winner]_Proliferations.md        # Phase 4 (optional)

Alternate: Combined Output

# [Source] - Social Content

## Concept 1: [Name]
[Concept details]

### Framework Matches
[2-4 options with fit assessment]

### Selected Posts
[Executed posts ready to publish]

---

## Concept 2: [Name]
[Repeat...]

Common Pitfalls to Avoid

Content Issues

❌ First-Match Bias - Picking first framework that fits ❌ Generic Posts - Not using any framework ❌ Too Many Concepts - Multiple ideas in one post ❌ Missing Context - Concept doesn't stand alone

Framework Issues

❌ Framework Drift - Abandoning structure mid-post ❌ Mismatched Fit - Forcing concept into wrong framework ❌ No Volume - Creating only 1 option instead of 5-10+

Platform Issues

❌ Wrong Link Placement - External links in Facebook/LinkedIn main posts ❌ Hashtag Overload - Too many for platform ❌ Voice Mismatch - Too formal for Facebook, too casual for LinkedIn

Process Issues

❌ Skipping Concept Extraction - Writing without identifying standalone concepts ❌ Not Documenting Framework - Can't reference what worked later


Success Metrics

A successful social post:

✅ Follows proven framework - Has clear structure ✅ Stops the scroll - Hook creates curiosity/emotion ✅ Standalone value - Works without reading source ✅ Platform-optimized - Follows rules for hashtags, links, length ✅ Framework documented - Template name noted ✅ Shareable - Audience would forward/tag others ✅ Authentic voice - Sounds like real person


Bundled Resources

Template Libraries

  • references/linkedin-frameworks.md - 136 LinkedIn templates
  • references/post-structures.md - 100+ text post frameworks

Proliferation Methods

  • references/scamper-guide.md - SCAMPER method with examples
  • references/human-desires-guide.md - 8 desires with reframing examples

Platform Guidelines

  • references/platform-guidelines.md - Detailed platform rules

Related Skills

  • hook-and-headline-writing - Optimize post hooks
  • anti-ai-writing - Humanize posts that sound too polished
  • voice-[style] - Match posts to specific brand voice
  • transcript-polisher - Clean source transcripts first

*Content creation is framework fitting. Extract concepts, match to templates, generate volume, select best. Turn 1 idea into 1000 variations.*

平台分布

Codex

36.44%
按下载量换算30

Claude

31.65%
按下载量换算26

Cursor

16.93%
按下载量换算14

Gemini CLI

9.84%
按下载量换算8

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