GB-Power-Market-JJ custom-youtube-summarize

Extract transcript from a YouTube video using Python and summarize it.

install

source · Clone the upstream repo

git clone https://github.com/GeorgeDoors888/GB-Power-Market-JJ

Claude Code · Install into ~/.claude/skills/

T=$(mktemp -d) && git clone --depth=1 https://github.com/GeorgeDoors888/GB-Power-Market-JJ "$T" && mkdir -p ~/.claude/skills && cp -r "$T/openclaw-skills/skills/5eun/custom-youtube-summarize" ~/.claude/skills/georgedoors888-gb-power-market-jj-custom-youtube-summarize && rm -rf "$T"

OpenClaw · Install into ~/.openclaw/skills/

T=$(mktemp -d) && git clone --depth=1 https://github.com/GeorgeDoors888/GB-Power-Market-JJ "$T" && mkdir -p ~/.openclaw/skills && cp -r "$T/openclaw-skills/skills/5eun/custom-youtube-summarize" ~/.openclaw/skills/georgedoors888-gb-power-market-jj-custom-youtube-summarize && rm -rf "$T"

manifest: openclaw-skills/skills/5eun/custom-youtube-summarize/SKILL.md

source content

Description

Extracts subtitles from YouTube videos and generates summaries. Use when users provide YouTube URLs with requests like "summarize this video" or "explain this content".

Setup

Create script at

{baseDir}/transcript_extract.py

with provided code

Usage

When receiving a YouTube URL:

Executes:

python3 {baseDir}/transcript_extract.py "<YOUTUBE_URL>"

Captures output between
```
[TRANSCRIPT-START]
```
and
```
[TRANSCRIPT-END]
```
Uses LLM to generate summary from raw transcript text

Configuration

script_path: "{baseDir}/transcript_extract.py"
supported_languages: ["ko", "en"]

Example

Input:

https://youtu.be/dQw4w9WgXcQ

(with "summarize this" context)

Output:

[TRANSCRIPT-START]
Never gonna give you up, never gonna let you down...
[TRANSCRIPT-END]

This video contains the iconic Rickroll meme track by Rick Astley. The lyrics revolve around themes of loyalty and commitment, featuring the artist's signature bassline and 80s production style. Despite its simple lyrics, it remains one of YouTube's most enduring viral pranks.

Notes

Handles both manually created and auto-generated subtitles
Falls back to English if Korean subtitles aren't available
Returns error messages if no transcript exists or URL is invalid