agentleFS
Sign inStart free
tumftumf/skills/youtube-summarizer/SKILL.md

youtube-summarizer

Skill4 starsChanged 7 months agoMITView on GitHub

Summarize YouTube videos by extracting and analyzing auto-generated subtitles. Use when the user provides a YouTube URL and asks to summarize, explain, or analyze a video's content. Triggers on patterns like "this video summarize", "summarize this YouTube", "what does this video say", or any request involving a youtube.com or youtu.be URL combined with summarization/analysis intent. Supports both Japanese and English videos.

  • Installs packages

We split this skill into the rules it gives an agent

5 rules. 0 are common, 4 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 6 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run the bundled script to fetch and clean the transcript:

    line 22
  2. Distinctive:

    If no subtitles exist, the script exits with code 1.

    line 32
  3. Very specific:

    Read /tmp/yt_meta.json for the title and language, then read /tmp/yt_transcript.txt for the full text.

    line 36
  4. Distinctive:

    Keep the summary concise but comprehensive.

    line 47
  5. Distinctive:

    Remove temporary files after summarization:

    line 51

“Common” is measured: kept by at least 6 in 10 of the 6 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

YouTube Summarizer

Prerequisites

  • yt-dlp must be installed (pip install yt-dlp)

Workflow

1. Extract transcript

Run the bundled script to fetch and clean the transcript:

python scripts/fetch_transcript.py "<YOUTUBE_URL>" --lang <LANG> 2>/tmp/yt_meta.json > /tmp/yt_transcript.txt
  • --lang ja (default) for Japanese videos, --lang en for English.
  • The script auto-detects original-language captions first, then falls back.
  • stdout: cleaned plain-text transcript.
  • stderr: JSON metadata {"video_id", "title", "language"}.
  • If no subtitles exist, the script exits with code 1.

2. Read transcript and metadata

Read /tmp/yt_meta.json for the title and language, then read /tmp/yt_transcript.txt for the full text.

3. Summarize

Produce a structured summary in the user's language. Use the following structure as a guide (adapt headings to fit the content):

  • Title / topic (one line)
  • Core concept (1-2 paragraphs explaining the main idea)
  • Key sections (3-5 sections with subheadings, covering the major points)
  • Conclusion / takeaway

Keep the summary concise but comprehensive. Preserve technical terms and proper nouns from the original.

4. Clean up

Remove temporary files after summarization:

rm -f /tmp/yt_transcript.txt /tmp/yt_meta.json

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free