Token导航 LogoToken导航

Agent Skill · 研究检索

cardiology-youtube-scriptwriter(心脏病学 youtube 编剧)

cardiology-youtube-scriptwriter 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 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。

cardiology-youtube-scriptwriter 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;许可证为 MIT,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 53,927。

总安装 519周安装 21GitHub Stars 3下载 163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

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:cardiology-youtube-scriptwriter(心脏病学 youtube 编剧)
来源仓库:https://github.com/drshailesh88/integrated_content_os
仓库路径:skills/cardiology-youtube-scriptwriter
安装命令:
npx skills add https://github.com/drshailesh88/integrated_content_os --skill cardiology-youtube-scriptwriter
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/drshailesh88/integrated_content_os --skill cardiology-youtube-scriptwriter

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。

提供从 Hello 到完成脚本的完整 YouTube 内容创作工作流,融合社交监听与研究框架。包含 Discovery、Social Listening、Idea Generation、Approval、Script Writing 五个核心阶段。

适用于创建 engaging 的心脏病学视频内容,支持用户交互选择创意方向与最终脚本。使用前请确认本地环境是否具备必要的依赖项与权限配置,避免因缺失组件导致运行失败。cardiology-youtube-scriptwriter 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/cardiology-youtube-scriptwriter

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/drshailesh88/integrated_content_os --skill car…;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

cardiology-youtube-scriptwriter 怎么安装?

在终端执行 npx skills add https://github.com/drshailesh88/integrated_content_os --skill cardiology-yo…。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

cardiology-youtube-scriptwriter 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。

cardiology-youtube-scriptwriter 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。

cardiology-youtube-scriptwriter 是免费的吗?

来源标注的许可证是 MIT。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。

cardiology-youtube-scriptwriter 用的人多吗?

公开数据是累计安装 519 次,近一周安装 21 次,下载 163 次,GitHub 3 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。

cardiology-youtube-scriptwriter 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。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,商用和分发边界好判断。
  • 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
cardiology-youtube-scriptwriter

心脏病学 youtube 编剧 · 本页

5193只读MIT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

Cardiology YouTube Scriptwriter

Complete workflow from "Hello" to finished script. Combines social listening research with proven content frameworks to create engaging cardiology YouTube content.


Workflow Overview

┌─────────────┐     ┌─────────────┐     ┌─────────────┐     ┌─────────────┐     ┌─────────────┐
│  1. HELLO   │────▶│ 2. RESEARCH │────▶│ 3. IDEATE   │────▶│ 4. APPROVE  │────▶│ 5. SCRIPT   │
│  Discovery  │     │   Social    │     │   Generate  │     │   Select    │     │   Write     │
│             │     │   Listening │     │   Ideas     │     │   Idea      │     │   Full      │
└─────────────┘     └─────────────┘     └─────────────┘     └─────────────┘     └─────────────┘

Phase 1: Discovery

When user starts conversation, gather context:

Ask (pick 2-3 most relevant):

  • What's your channel about / who's your audience?
  • What topics have performed well / poorly for you?
  • Any specific areas you want to focus on?
  • Video length preference? (Shorts, 5-10 min, long-form)
  • What's your content goal? (Grow subs, educate, build authority)
  • Any topics you want to avoid?

Keep it conversational - don't interrogate. If user just wants to dive in, roll with it.


Phase 2: Social Listening Research

Use web search to find what's trending and what people are asking about.

Search Strategy

Execute 3-5 searches mixing these patterns:

1. Reddit: "site:reddit.com [topic] heart" OR "site:reddit.com cardiology"
2. Questions: "[cardiology topic] questions people ask"
3. Trends: "heart health trending 2024" OR "[topic] new research"
4. YouTube: "cardiology YouTube video ideas" (see what's working)
5. Forums: "[topic] patient questions forum"

Extract & Synthesize

Look for:

  • Hot questions people are asking
  • Misconceptions that need correcting
  • Emotional language (fear, confusion, hope)
  • Gaps in existing content
  • Trending hooks and angles

Reference: social-listening.md for detailed research methodology.


Phase 3: Generate Ideas

Combine social listening insights with seed topics and modifiers.

Formula

[Seed Topic] + [Modifier(s)] + [Social Insight] = Targeted Video Concept

Resources

  • Seeds: seeds.md - 300 cardiology topics across 15 categories
  • Modifiers: modifiers.md - 200+ angles, audiences, and hooks

Generate 3-5 Ideas

For each idea, provide:

### Idea [#]: [Working Title]

**Hook Angle:** [One-sentence hook]
**Target Audience:** [Who this is for]
**Why Now:** [Social listening insight that validates this]
**Format:** [Listicle / Explainer / Myth-Buster / Story / etc.]
**Length:** [Estimated minutes]
**Seed + Modifiers Used:** [Show the combination]

Quality Filters

Each idea must pass:

  • Addresses real question/concern from research
  • Clear audience and intent
  • Not fear-mongering without actionable advice
  • Differentiated from obvious existing content
  • Matches user's channel focus and style

Phase 4: Approval

Present ideas and let user choose. Be ready to:

  • Refine: Adjust angle, audience, or scope
  • Combine: Merge elements from multiple ideas
  • Regenerate: Create fresh options if none fit
  • Clarify: Explain why certain angles work

Get clear approval before moving to script.


Phase 5: Script Writing

Reference: script-templates.md for structures and retention techniques.

Select Structure Based on Video Type

Video TypeUse Template
Educational explainerStructure A
List/rankingStructure B
Myth vs factStructure C
Case study / storyStructure D
Short-form (<60s)Structure E

Script Sections

Every script includes:

  1. HOOK (0:00-0:30)

- Pattern interrupt or surprising fact - Clear promise of value - "Stay for..." tease (optional)

  1. INTRO (0:30-1:30)

- Brief credibility (don't over-explain) - Context: Why this matters now - Preview: What viewer will learn

  1. MAIN CONTENT

- Chunked into clear sections - Retention elements every 2-3 min - Evidence + stories + practical tips balanced

  1. WRAP-UP

- 3 key takeaways (numbered) - Clear CTA - Engagement prompt

Script Format

Deliver scripts in this format:

# [VIDEO TITLE]
## [Subtitle / Hook Line]

**Target Length:** X minutes
**Style Notes:** [Tone, pacing, B-roll suggestions]

---

### HOOK [0:00-0:30]

[Script text with suggested delivery cues]

**[GRAPHIC: Key stat or visual]**

---

### INTRO [0:30-1:30]

[Script text]

---

### SECTION 1: [Title] [Timestamp]

[Script text]

**💡 RETENTION MOMENT:** [Pattern interrupt or question]

---

[Continue sections...]

---

### WRAP-UP + CTA [Timestamp]

**3 KEY TAKEAWAYS:**
1. [Takeaway]
2. [Takeaway]
3. [Takeaway]

[CTA script]

---

## Production Notes

**Suggested B-Roll:**
- [Scene suggestions]

**Graphics/Text Overlays:**
- [Key stats or terms to display]

**Thumbnail Concept:**
- [Visual idea for thumbnail]

Safety & Quality Guardrails

Always Include:

  • Medical disclaimer (education, not personal advice)
  • "Talk to your doctor" for decision-making
  • Balanced presentation of evidence
  • Emergency warning signs when relevant

Never Include:

  • Guaranteed outcomes or cure claims
  • Anti-doctor or anti-medicine rhetoric
  • Unsubstantiated alternative treatments as fact
  • Content that could delay emergency care
  • Fear-mongering without actionable guidance

Handle Carefully:

  • Controversial topics: Present evidence fairly
  • Supplements: Acknowledge evidence levels
  • Medications: Side effects AND benefits
  • Alternative approaches: Neither dismiss nor endorse without evidence

Quick Reference

For idea generation: Read seeds.md + modifiers.md For script structures: Read script-templates.md For research approach: Read social-listening.md


Example Workflow

User: "Hey, I want to create some YouTube content this week"

Claude: [Phase 1] "Great! Quick questions - what's your audience like,
        and any topics you're excited about or want to avoid?"

User: "My audience is mostly 40-60, worried about heart disease.
       Open to anything but want to stay positive."

Claude: [Phase 2] *Performs 4-5 web searches for trending topics,
        Reddit discussions, recent news*

Claude: [Phase 3] "Based on what I found, here are 4 ideas..."
        *Presents structured ideas with hooks and validation*

User: "Love idea #2, but can we make it more for beginners?"

Claude: [Phase 4] "Absolutely, here's the refined version..."

User: "Perfect, let's do it!"

Claude: [Phase 5] *Delivers full script with hook, structure,
        timestamps, and production notes*

平台分布

Claude Code

31.27%
按下载量换算51

OpenCode

20.42%
按下载量换算33

Antigravity

16.74%
按下载量换算27

Gemini CLI

13.04%
按下载量换算21

windsurf

7.96%
按下载量换算13

Codex

3.16%
按下载量换算5

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。

继续浏览同类 Skills