my-fetch-youtube

Extract YouTube captions, correct errors via Web Search, and generate summary, insights, and translation.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/Kwondongkyun/claude-code-settings --skill my-fetch-youtube
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: my-fetch-youtube
Source: https://github.com/Kwondongkyun/claude-code-settings/tree/main/skills/my-fetch-youtube
Command: npx skills add https://github.com/Kwondongkyun/claude-code-settings --skill my-fetch-youtube

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

YouTube URL에서 자막을 추출하고, Web Search로 자동 자막 오류를 보정한 뒤, 요약-인사이트-전체 번역의 3단계 파이프라인으로 제공하는 스킬.

Core Features & Use Cases

  • 자막 추출 및 정제: 영상의 자막을 텍스트로 변환하고 불필요한 정보 제거
  • 메타데이터 기반 보정: 영상 제목/설명을 활용한 용어 교정 및 정합성 향상
  • 요약-인사이트-번역 파이프라인: 3단계로 핵심 내용, 시사점, 번역된 아티클 제공

Quick Start

유튜브 URL을 입력하면 자막을 추출하고 웹 검색으로 자동 자막 보정을 수행한 뒤, 요약-인사이트-전체 번역의 결과를 제공합니다.

Frequently Asked Questions about my-fetch-youtube

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I extract and summarize YouTube captions automatically?

The pipeline corrects YouTube caption errors by running web searches for proper terminology, using video metadata like titles and descriptions as context to fix transcription inaccuracies.

Can I translate YouTube transcripts into another language after extraction?

You need yt-dlp installed to fetch caption data from the video URL, alongside web search capability to verify terminology and correct transcription errors during the processing pipeline.

Why does automated YouTube transcription contain terminology errors?

Automated YouTube transcription contains terminology errors because speech recognition algorithms misinterpret specialized words, which requires web search-based correction against video metadata to restore accuracy.

What is the best way to get insights from a YouTube video without watching it?

The best way to get insights from a YouTube video without watching it is using a caption extraction pipeline that outputs structured insights derived from the corrected transcript text.