youtube-research_skill

This skill discovers trending YouTube videos in a niche, analyzes top performers, and generates actionable hook formulas and reports.
  • Python

16

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill bradautomates/head-of-content --skill youtube-research

  • SKILL.md6.6 KB

Overview

This skill researches high-performing YouTube outlier videos in a niche using TubeLab's outlier detection API and AI-driven video analysis. It identifies outlier videos, selects up to three that are most relevant to your channel, and produces a report with replicable hook formulas and actionable content recommendations. Use it to discover what’s working now and turn viral patterns into repeatable ideas.

How this skill works

The skill runs keyword searches for direct and adjacent audience topics, detects outlier videos by z-score and recency boost, and downloads thumbnails and transcripts. It filters the outliers to a maximum of three videos most relevant to the user’s niche, then analyzes each video with an AI content analyzer to extract hook techniques, content structure, retention tactics, and CTAs. Finally it compiles a structured report with top-performing hooks, content structure patterns, CTA strategies, and clear takeaways.

When to use it

  • When you need trending video ideas tailored to your channel niche
  • To research competitor content and reverse-engineer successful formats
  • When looking for viral patterns and repeatable hook formulas
  • To generate specific content concepts based on high-performing videos
  • When running periodic YouTube trend audits for growth strategy

Best practices

  • Provide 4 direct keywords and 4 adjacent keywords to broaden discovery
  • Limit analysis to max 3 most relevant outliers to keep insights focused
  • Prefer videos that match your content style (tutorials vs entertainment)
  • Use the extracted replicable hook formulas directly in scripts and titles
  • Prioritize suggestions with both title and transcript relevance

Example use cases

  • A cooking channel finds three outlier recipe videos and adapts the top hook formula for a new series
  • A tech reviewer reverse-engineers competitors’ retention techniques and CTA placements
  • A creator tests three AI-suggested title formulas derived from top-performing hooks
  • A channel audit identifies adjacent-audience topics to expand content verticals
  • A growth team compiles weekly reports to track shifting hook performance

FAQ

The pipeline filters outliers and analyzes up to 3 videos that are most relevant to your niche.

What makes a video an outlier?

Outliers are ranked by a zScore that measures performance above the channel average, multiplied by a recency boost that favors recent growth.

Do I need API keys?

Yes. The skill requires a TubeLab API key for outlier detection and a GenAI key for the video analysis step.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational