# Youtube Packaging

> Analyze YouTube channels for normalized outliers, title and thumbnail patterns, and ownable content gaps. Use when selecting video topics, improving packaging, or testing whether a competitor pattern is repeatable rather than a one-off spike.

- Skill: `majesticlabs-dev/youtube-packaging` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add majesticlabs-dev/youtube-packaging`
- Raw SKILL.md: https://api.skillmd.com/api/skills/majesticlabs-dev/youtube-packaging/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Marketing & Growth
- Author: majesticlabs-dev (https://skillmd.com/u/majesticlabs-dev)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/majesticlabs-dev/youtube-packaging

---


# YouTube Packaging

## Boundary

Analyze packaging and topic signals, not copy competitors or guarantee views. Public data is incomplete and must be labeled accordingly.

## Required Inputs

- Target channel, audience, and business goal
- Competitor channels or discovery queries
- Time window and video format
- Available public metrics and the user’s credible proof or production assets

## Workflow

1. Build a relevant channel set and record the selection method.
2. Estimate each channel’s recent baseline by comparable format and age.
3. Flag normalized outliers and note data limitations.
4. Classify promise, tension, specificity, authority, format, title, and thumbnail patterns.
5. Separate repeated patterns from one-off anomalies.
6. Find gaps the user can credibly own with distinct evidence or perspective.
7. Draft title and thumbnail concepts as hypotheses, then define tests and production needs.

## Output

1. **Channel set and confidence**
2. **Normalized outlier table**
3. **Repeated pattern and anomaly analysis**
4. **Ownable content gaps**
5. **Packaging concepts and production recommendations**

## Quality Gate

- Raw views are not compared across unlike channels.
- No title or thumbnail is copied.
- The user can substantiate every proposed promise.
- Recommendations include counterevidence and production cost.

## Reference

Use [data-currentness.md](references/data-currentness.md) to check what public data supports before analysis.