youtube-research
Skill0 starsChanged 35 days agoMITView on GitHub
Research high-performing YouTube videos in a niche WITHOUT TubeLab: yt-dlp scraping + local z-score outlier math (free). Identifies outlier videos, analyzes top 3 relevant videos with AI, and generates reports with actionable hook formulas.
- Reads credentials
16 other projects use a version of this, 1 of them edited.
Adopted by
Most recently taken up by
- aya-taleb/video-boost-skillstheir own version, first seen Nov 15, 2026
- marciob/youtube-researchtheir own version, first seen Nov 8, 2026
- PostPlusAI/postplus-skillstheir own version, first seen Nov 5, 2026
- jindratilk/youtube-research-mcptheir own version, first seen Nov 1, 2026
- matthewharrilal/claude-skillstheir own version, first seen Oct 13, 2026
- laodao-ai/laodao-skillstheir own version, first seen Oct 13, 2026
Other versions16 projects use a version of this.
Word for word · 0
None found.
Edited · 1
- bradautomates/head-of-content★ 22796% same
Wrote their own version · 15
- wnstify/tubeflow★ 36same name
- PostPlusAI/postplus-skills★ 27same name
- Samin12/youtube-research-skill★ 9same name
- Samin12/skills★ 8same name
- YepAPI/skills★ 8same name
- TimBroddin/skills★ 6same name
- gfreeau/claude-skills★ 2same name
- aya-taleb/video-boost-skills★ 1same name
- marciob/youtube-research★ 1same name
- MKirovBG/scribefy-mcp★ 1same name
5 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository5 other files it gives its agents.
- content-plannerSkill
- instagram-researchSkill
- tiktok-researchSkill
- video-content-analyzerSkill
- x-researchSkill
The file
We split this skill into the rules it gives an agent
8 rules. 0 are common, 4 are distinctive, and 4 are very specific to this project.
0
Common
Most of the 13 edited versions we read keep these.
4
Distinctive
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 26
No API key needed: yt-dlp scrapes channels, outlier z-scores computed locally
- Very specific: line 40
Read .claude/context/youtube-channel.md to get the channel ID.
- Distinctive: line 62
Run the outlier search with both keyword sets:
- Very specific: line 84
Read outliers.json and the user's niche from .claude/context/youtube-channel.md.
- Distinctive: line 88
Transcript relevance: If transcript exists, check it mentions relevant topics
- Distinctive: line 89
Direct niche priority: Prefer videos from direct keyword search over adjacent
- Very specific: line 93
Write the filtered videos to {RUN_FOLDER}/filtered-outliers.json:
- Very specific: line 121
Read {RUN_FOLDER}/outliers.json and {RUN_FOLDER}/video-analysis.json, then generate {RUN_FOLDER}/report.md.
“Common” is measured: kept by at least 6 in 10 of the 13 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.
Files
YouTube Research
Research high-performing YouTube outlier videos, analyze top content with AI, and generate actionable reports.
Prerequisites
- No API key needed: yt-dlp scrapes channels, outlier z-scores computed locally
GEMINI_API_KEYenvironment variable (for video analysis)google-genai,yt-dlp,youtube-transcript-apiPython packages
Workflow
Step 1: Create Run Folder
mkdir -p youtube-research/$(date +%Y-%m-%d_%H%M%S)Step 2: Get Channel ID
Read .claude/context/youtube-channel.md to get the channel ID.
Step 3: Fetch Channel Videos
python scripts/get_channel_videos.py CHANNEL_ID --format summaryThis returns JSON with the channel's video titles and view counts.
Step 4: Analyze Channel
Analyze the channel data to extract:
- keywords: 4 search terms for the channel's direct niche
- adjacent-keywords: 4 search terms for topics the same audience watches
- audience: 2-3 profiles with objections, transformations, stakes
- formulas: Reusable title templates
See references/channel-analysis-schema.md for the full schema and example output.
Step 5: Search for Outliers
Run the outlier search with both keyword sets:
python .claude/skills/youtube-research/scripts/find_outliers.py \
--keywords "keyword1" "keyword2" "keyword3" "keyword4" \
--adjacent-keywords "adjacent1" "adjacent2" "adjacent3" "adjacent4" \
--output-dir youtube-research/{run-folder} \
--top 5This runs two searches:
- Direct niche: keywords with 5K+ views threshold
- Adjacent audience: adjacent-keywords with 10K+ views threshold
Output files:
outliers.json- All outliers normalized for video analysisreport.md- Basic markdown reportthumbnails/*.jpg- Video thumbnailstranscripts/*.txt- Video transcripts
Step 6: Filter Relevant Videos for Analysis
Read outliers.json and the user's niche from .claude/context/youtube-channel.md.
CRITICAL: Select MAX 3 videos that are most relevant to the user's niche. Filter by:
- Title relevance: Title contains keywords related to user's niche/topics
- Transcript relevance: If transcript exists, check it mentions relevant topics
- Direct niche priority: Prefer videos from direct keyword search over adjacent
Skip videos that are clearly outside the user's content style (e.g., entertainment/vlogs when user does tutorials).
Write the filtered videos to {RUN_FOLDER}/filtered-outliers.json:
{
"outliers": [/* max 3 relevant videos */],
"filter_reason": "Selected based on relevance to [user's niche]"
}Step 7: Analyze Top Videos with AI
python3 .claude/skills/video-content-analyzer/scripts/analyze_videos.py \
--input {RUN_FOLDER}/filtered-outliers.json \
--output {RUN_FOLDER}/video-analysis.json \
--platform youtube \
--max-videos 3Extracts from each video:
- Hook technique and replicable formula
- Content structure and sections
- Retention techniques
- CTA strategy
See the video-content-analyzer skill for full output schema and hook/format types.
Step 8: Generate Final Report
Read {RUN_FOLDER}/outliers.json and {RUN_FOLDER}/video-analysis.json, then generate {RUN_FOLDER}/report.md.
Report Structure:
# YouTube Research Report
Generated: {date}
## Top Performing Hooks
Ranked by engagement. Use these formulas for your content.
### Hook 1: {technique} - {channelTitle}
- **Video**: "{title}"
- **Opening**: "{opening_line}"
- **Why it works**: {attention_grab}
- **Replicable Formula**: {replicable_formula}
- **Views**: {viewCount} | **zScore**: {zScore}
- [Watch Video]({url})
[Repeat for each analyzed video]
## Content Structure Patterns
| Video | Format | Pacing | Key Retention Techniques |
|-------|--------|--------|--------------------------|
| {title} | {format} | {pacing} | {techniques} |
## CTA Strategies
| Video | CTA Type | CTA Text | Placement |
|-------|----------|----------|-----------|
| {title} | {type} | "{cta_text}" | {placement} |
## All Outliers
### Direct Niche
| Rank | Channel | Title | Views | zScore |
|------|---------|-------|-------|--------|
[List direct niche outliers]
### Adjacent Audience
| Rank | Channel | Title | Views | zScore |
|------|---------|-------|-------|--------|
[List adjacent outliers]
## Actionable Takeaways
[Synthesize patterns into 4-6 specific recommendations based on video analysis]Focus on actionable insights. The "Top Performing Hooks" section with replicable formulas should be prominent.
Quick Reference
Full pipeline:
RUN_FOLDER="youtube-research/$(date +%Y-%m-%d_%H%M%S)" && mkdir -p "$RUN_FOLDER" && \
python .claude/skills/youtube-research/scripts/find_outliers.py \
--keywords "k1" "k2" "k3" "k4" \
--adjacent-keywords "a1" "a2" "a3" "a4" \
--output-dir "$RUN_FOLDER" --top 5Then filter outliers for niche relevance (max 3), run video analysis, and generate the report.
Script Reference
get_channel_videos.py
python .claude/skills/youtube-research/scripts/get_channel_videos.py CHANNEL_ID [--format json|summary]| Arg | Description |
CHANNEL_ID | YouTube channel ID (24 chars) |
--format | json (full data) or summary (for analysis) |
find_outliers.py
python .claude/skills/youtube-research/scripts/find_outliers.py --keywords K1 K2 K3 K4 --adjacent-keywords A1 A2 A3 A4 --output-dir DIR [options]| Arg | Description |
--keywords | Direct niche keywords (4 recommended) |
--adjacent-keywords | Adjacent topic keywords (4 recommended) |
--output-dir | Output directory (required) |
--top | Videos per category (default: 5) |
--days | Days back to search (default: 30) |
--json | Also save raw JSON data |
Output: outliers.json, report.md, thumbnails/, transcripts/
Scoring Algorithm
Videos ranked by: zScore × recency_boost
- zScore: How much video outperforms its channel average
- recency_boost: 1.0 for today, decays 5%/day (min 0.3×)
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
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.

