YouTube Insights
Extract transcripts and deliver structured insight summaries from YouTube videos.
When to Use
- User shares a YouTube URL and wants insights, summary, or transcript
- User says "get me the insight of this video", "resume cette video", "what's in this video"
- User shares multiple YouTube URLs to analyze
- Any request involving YouTube video content extraction
Prerequisites
yt-dlp must be installed and in PATH (check with where yt-dlp or which yt-dlp)- Python 3.x available
Workflow
Step 1: Extract Transcripts
For EACH YouTube URL provided, run the extraction script in parallel (one agent per video if multiple):
python "C:\Users\momomuchu\.claude\skills\youtube-insights\scripts\get-transcript.py" "<youtube-url>"
The script returns JSON with: video_id, title, channel, duration, upload_date, view_count, description, transcript.
If the script fails or transcript is empty, try specifying languages:
python "C:\Users\momomuchu\.claude\skills\youtube-insights\scripts\get-transcript.py" "<youtube-url>" "fr,en"
Step 2: Analyze & Summarize
For each video, produce a structured insight summary using this format:
## [Video Title] — [Channel Name]
**Duration:** X min | **Views:** X | **Date:** YYYY-MM-DD
### Key Insights
- [Insight 1 — the most important, actionable takeaway]
- [Insight 2]
- [Insight 3]
- [...]
### What to Take Away
[2-3 sentences: what's personally actionable for the user based on their context/projects]
### Notable Quotes
> "[Direct quote from transcript that captures a key point]"
Rules
- Parallel extraction — if multiple URLs, launch all transcript downloads simultaneously
- Language detection — the script auto-tries en, fr, es, de, pt, ja, ko, zh. French videos will get French transcripts
- Insight focus — don't just summarize chronologically. Extract the NON-OBVIOUS insights, frameworks, numbers, and actionable advice
- User context — if memory has info about the user's projects/goals, connect insights to what matters to them
- Be concise — max 10 bullet points per video. Quality over quantity
- Numbers matter — always extract specific metrics, revenue numbers, conversion rates, timeframes
- Frameworks — if the speaker shares a step-by-step framework, extract it as a numbered list
- Warnings — if the speaker warns against something, flag it clearly
- Skip filler — ignore sponsor segments, self-promotion, "like and subscribe" content
1---2name: youtube-insights3description: Extract transcripts and key insights from YouTube videos. Handles single or multiple URLs. Uses yt-dlp for reliable transcript extraction, supports all languages. Actions: get transcript, summarize video, extract insights, analyze video. Triggers on: YouTube URL, 'insight video', 'resume video', 'transcript', 'what does this video say'.4---56# YouTube Insights78Extract transcripts and deliver structured insight summaries from YouTube videos.910## When to Use1112- User shares a YouTube URL and wants insights, summary, or transcript13- User says "get me the insight of this video", "resume cette video", "what's in this video"14- User shares multiple YouTube URLs to analyze15- Any request involving YouTube video content extraction1617## Prerequisites1819- `yt-dlp` must be installed and in PATH (check with `where yt-dlp` or `which yt-dlp`)20- Python 3.x available2122## Workflow2324### Step 1: Extract Transcripts2526For EACH YouTube URL provided, run the extraction script **in parallel** (one agent per video if multiple):2728```bash29python "C:\Users\momomuchu\.claude\skills\youtube-insights\scripts\get-transcript.py" "<youtube-url>"30```3132The script returns JSON with: `video_id`, `title`, `channel`, `duration`, `upload_date`, `view_count`, `description`, `transcript`.3334If the script fails or transcript is empty, try specifying languages:35```bash36python "C:\Users\momomuchu\.claude\skills\youtube-insights\scripts\get-transcript.py" "<youtube-url>" "fr,en"37```3839### Step 2: Analyze & Summarize4041For each video, produce a structured insight summary using this format:4243```markdown44## [Video Title] — [Channel Name]4546**Duration:** X min | **Views:** X | **Date:** YYYY-MM-DD4748### Key Insights49- [Insight 1 — the most important, actionable takeaway]50- [Insight 2]51- [Insight 3]52- [...]5354### What to Take Away55[2-3 sentences: what's personally actionable for the user based on their context/projects]5657### Notable Quotes58> "[Direct quote from transcript that captures a key point]"59```6061### Rules62631. **Parallel extraction** — if multiple URLs, launch all transcript downloads simultaneously642. **Language detection** — the script auto-tries en, fr, es, de, pt, ja, ko, zh. French videos will get French transcripts653. **Insight focus** — don't just summarize chronologically. Extract the NON-OBVIOUS insights, frameworks, numbers, and actionable advice664. **User context** — if memory has info about the user's projects/goals, connect insights to what matters to them675. **Be concise** — max 10 bullet points per video. Quality over quantity686. **Numbers matter** — always extract specific metrics, revenue numbers, conversion rates, timeframes697. **Frameworks** — if the speaker shares a step-by-step framework, extract it as a numbered list708. **Warnings** — if the speaker warns against something, flag it clearly719. **Skip filler** — ignore sponsor segments, self-promotion, "like and subscribe" content
momomuchu