Token导航 LogoToken导航

Agent Skill · 研究检索

youtube-wisdom(YouTube 智慧)

youtube-wisdom 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。

youtube-wisdom 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,权限为操作浏览器;许可证信息还不明确,第三方扫描有未通过项,按安装量在研究检索类的 73,675 个 Skill 里排第 72,323。

总安装 188周安装 8GitHub Stars 125下载 66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:youtube-wisdom(YouTube 智慧)
来源仓库:https://github.com/sammcj/agentic-coding
仓库路径:skills/youtube-wisdom
安装命令:
npx skills add https://github.com/sammcj/agentic-coding --skill youtube-wisdom
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/sammcj/agentic-coding --skill youtube-wisdom

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。

用于基于 YouTube 内容进行知识检索与筛选。可根据关键词快速定位相关视频与关键信息。适合研究、学习或市场调研类任务场景。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。结果准确性依赖搜索词质量与平台内容覆盖度。

youtube-wisdom 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/youtube-wisdom

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/sammcj/agentic-coding --skill youtube-wisdom;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

    它需要操作浏览器,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

youtube-wisdom 怎么安装?

在终端执行 npx skills add https://github.com/sammcj/agentic-coding --skill youtube-wisdom。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

youtube-wisdom 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。

youtube-wisdom 需要什么权限?

权限级别是操作浏览器。该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

youtube-wisdom 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。

youtube-wisdom 用的人多吗?

公开数据是累计安装 188 次,近一周安装 8 次,下载 66 次,GitHub 125 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。

youtube-wisdom 安全吗?

第三方扫描有未通过项,风险较高。除非你确认过源码,否则不建议直接安装到常用环境。

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。

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。

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
youtube-wisdom

YouTube 智慧 · 本页

188125操作浏览器未明确未通过

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

YouTube Wisdom Extraction

Overview

Extract meaningful insights, key learnings, and actionable wisdom from YouTube videos. This skill handles the complete workflow: downloading transcripts using yt-dlp (no video files), parsing subtitle data, performing comprehensive analysis, and saving results to organised markdown files.

When to Use This Skill

This skill is useful when users request:

  • Analysis or summary of YouTube video content
  • Extraction of key insights or learnings from videos
  • Identification of notable quotes or important statements
  • Structured breakdown of video content
  • Actionable takeaways from educational or informational videos

Workflow

Step 1: Download Transcript

Execute the download script to fetch only the transcript (no video file):

bash scripts/download_video.sh <youtube-url>

The script will:

  • Auto-detect your environment (Claude Code or OpenClaw) and OS (macOS or Linux)
  • Download English subtitles or auto-generated transcripts
  • Output the transcript path and next steps

Cookie handling: The script first attempts to download using browser cookies (for access restricted videos). If no browser is found or cookies fail, it automatically falls back to trying cookie-less download.

Step 2: Rename Directory

The script outputs a NEXT_STEP line. Execute the rename command it provides:

mv <OUTPUT_DIR> <OUTPUT_DIR>/../YYYY-MM-DD-<concise-description>
  • Keep the description short (1-6 words)
  • Use hyphens instead of spaces
  • Take the video content and title into consideration

Example:

  • Video: "My Interview With Demis Hassabis, CEO of DeepMind"
  • Rename to: 2026-02-05-Demis-Hassabis-Interview

Step 3: Read Transcript

Read the transcript file from the path provided by the script (TRANSCRIPT_PATH).

Transcripts are cleaned and formatted as continuous text with minimal whitespace.

Note: The download script uses --restrict-filenames to sanitise special characters in filenames for safer handling.

Step 4: Analyse and Extract Wisdom

IMPORTANT: One of your goals is to avoid signal dilution, context collapse, quality degradation and degraded reasoning for future understanding of the content by ensuring you keep the signal to noise ratio high and that domain insights are preserved while not introducing unnecessary filler or fluff in documentation.

Perform comprehensive analysis on the transcript, extracting:

1. Key Insights & Takeaways
  • Identify the main ideas, core concepts, and central arguments
  • Extract fundamental learnings and important revelations
  • Highlight expert advice, best practices, or recommendations
  • Note any surprising or counterintuitive information
2. Notable Quotes
  • Extract memorable, impactful, or particularly well-articulated statements
  • Include context for each quote when relevant
  • Focus on quotes that encapsulate key ideas or provide unique perspectives
  • Preserve the original wording exactly as spoken, except correct American spellings to Australian English
3. Structured Summary
  • Create hierarchical organisation of content
  • Break down into logical sections or themes
  • Provide clear section headings that reflect content structure
  • Include high-level overview followed by detailed breakdowns
  • Note any important examples, case studies, or demonstrations
4. Actionable Takeaways
  • List specific, concrete actions viewers can implement
  • Frame as clear, executable steps
  • Prioritise practical advice over theoretical concepts
  • Include any tools, resources, or techniques mentioned
  • Distinguish between immediate actions and longer-term strategies

Step 5: Write Analysis to Markdown File

Write the complete analysis to a markdown file in the renamed directory:

File location: <renamed-directory>/<video-title> - analysis.md

Format the analysis using this structure:

# Video Analysis: [Video Title]

**Video URL:** [YouTube URL]
**Analysis Date:** [Current date in YYYY-MM-DD]

## Summary
[Brief 2 sentence overview of the video's main topic and purpose]

### Simplified Explanation
[Explain It Like I'm 10: A simple 1-2 sentence explanation of the core concept in a way a 10-year-old could understand]

### Key Takeaways
- [Concise takeaway 1]
- [Concise takeaway 2]
- [Concise takeaway 3]

## Key Insights
- [Insight 1]
  - [Supporting detail]
- [Insight 2]
  - [Supporting detail]
- [Insight 3]
  - [Supporting detail]
- etc..

## Notable Quotes (Only add this if there are notable quotes / statements)
> "[Quote 1]"

Context: [Brief context if needed]

> "[Quote 2]"

Context: [Brief context if needed]

## Structured Breakdown
### [Section 1 Title]
[Content summary]

### [Section 2 Title]
[Content summary]

## Actionable Takeaways
1. [Specific action item 1]
2. [Specific action item 2]
3. [Specific action item 3]

## Additional Resources
[Any tools, links, or references mentioned in the video]

_Wisdom Extraction Skill Generated: [Current date in YYYY-MM-DD]_

After writing the analysis file, inform the user of the location.

OpenClaw note: When running via OpenClaw (in a chat session), also return the full analysis content to the user in your response — don't just save it to file. Users often want to read the insights immediately without opening a separate file.

Step 6: Critical Self-Review

Conduct a critical self-review of YOUR summarisation and analysis.

Create tasks to track the following (mechanical checks first, then content quality):

  • No American English spelling - check and fix (e.g. judgment→judgement, practicing→practising, organize→organise)
  • No em-dashes, smart quotes, or non-standard typography
  • Proper markdown formatting
  • Accuracy & faithfulness to the original content
  • Completeness
  • Concise, clear content with no fluff, marketing speak, filler, or padding (ensure the content has a high signal to noise ratio)
  • Logical organisation & structure

Re-read the analysis file, verify each item, fix any issues found, then mark tasks completed.

Step 7: Send Completion Notification (Claude Code Only & Optional)

Use scripts/send_notification.sh to send a desktop notification:

TITLE="Wisdom Extracted" MESSAGE="<short description>" PLAY_SOUND=true DIR="<renamed-directory>" bash scripts/send_notification.sh

Then stop unless further instructions are given.

Additional Capabilities

Multiple Video Analysis

When analysing multiple videos:

  • Process each video sequentially using the workflow above
  • Each video gets its own directory by video ID
  • Create comparative analysis highlighting common themes or contrasting viewpoints
  • Synthesise insights across multiple sources in a separate summary file
  • Notify once only at the end of the entire batch process

Time-Stamped Analysis

If timestamps are needed:

  • Note that basic transcripts don't preserve timestamps
  • Can reference general flow (beginning, middle, end) of content
  • For precise timestamps, may need to cross-reference with the actual video

Topic-Specific Focus

When user requests focused analysis on specific topics:

  • Search transcript for relevant keywords and themes
  • Extract only content related to specified topics
  • Provide concentrated analysis on areas of interest

Tips

  • Don't add new lines between items in a list
  • Avoid marketing speak or fluff
  • Always use Australian English spelling
  • Do not use em-dashes or smart quotes
  • Only use bold where emphasis is truly needed
  • Ensure clarity and conciseness in summaries and takeaways
  • Consider if the text you're adding actually adds value to the analysis, don't add filler or padding
  • If the video mentions a specific tool, resource or website, task a sub-agent to look it up and provide a brief summary, then include it in the Additional Resources section

Resources

scripts/

  • download_video.sh: Bash script that downloads YouTube transcripts (no video files) using yt-dlp with optimised settings. Auto-detects environment (Claude Code/OpenClaw) and OS, outputs paths and next steps for the agent.
  • send_notification.sh: Sends desktop notifications when analysis is complete (Claude Code only, optional).

Installation Paths

This skill works with both Claude Code and OpenClaw:

  • Claude Code: ~/.claude/skills/youtube-wisdom/
  • OpenClaw: ~/.openclaw/workspace/skills/youtube-wisdom/

平台分布

Codex

37.8%
按下载量换算25

Claude

28.54%
按下载量换算19

Cursor

21.06%
按下载量换算14

Gemini CLI

10.49%
按下载量换算7

安全审计

Gen Agent Trust Hub

未通过

Socket

通过

Snyk

可疑

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

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