AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Youtube Outliers

skill-ravsau-youtuber-skills-youtube-outliers · by ravsau

3x outlier videos in a niche, their format, and 3 adaptations. Use when picking the next video or asked what is working in a lane.

↗ Repository
— No reviews yet
0 installs
4 views
0.0% view→install

Install

$ agentstack add skill-ravsau-youtuber-skills-youtube-outliers

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • ✓ Prompt-injection patterns
  • ✓ Secret / credential exfiltration
  • ✓ Dangerous shell & filesystem operations
  • ✓ Untrusted network calls
  • ✓ Known-malicious package signatures

What it can access

  • ✓ Network access No
  • ✓ Filesystem access No
  • ✓ Shell / process execution No
  • ✓ Environment & secrets No
  • ✓ Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-ravsau-youtuber-skills-youtube-outliers)

Reliability & compatibility

✓ Security review passed
0 installs to date
— no reviews yet
● 3d ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Youtube Outliers? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

YouTube Outliers

Input: a niche, and if given, channels to study and the creator's own recent titles. Output: an outlier table, the formats behind them, and 3 adaptations ready to package.

Steps

  1. List 5-8 channels: direct niche, adjacent niche, one unrelated niche with the same audience age.
  2. For each channel, take the last 20 videos and compute the average views. Flag any video at 3x or more.
  3. Drop outliers caused by news events or collabs. Keep format-driven ones.
  4. For each kept outlier, strip the topic and write the format as one sentence with blanks.
  5. Write 3 adaptations that fill the blanks with the creator's niche, angle, and budget.
  6. Rank them by fit, then hand the top one to youtube-packaging.

Rules

  • A format is a structure, never a topic. "[Tool A] vs [Tool B], tested for [N] days" is a format.
  • Keep the tension of the original and shrink the spend to the creator's budget.
  • Every adaptation must add something the original cannot: expertise, data, or access.
  • If public view counts are unavailable, say so and ask for the creator's own numbers.
  • Never list an outlier without its multiple.

Output format

## Outliers: [niche]

| Video | Channel | Views | Channel avg | Multiple | Format |
|-------|---------|-------|-------------|----------|--------|
| ... | ... | ... | ... | 4.2x | ... |

### Adaptations
1. **[title idea]** from [format]. Angle: ... Budget: ...
2. ...
3. ...

Top pick: #N. Why: one line.

Example

Input: local AI on Mac; channels: two hardware reviewers, one ML tutorial channel

| Video | Channel | Views | Channel avg | Multiple | Format | |-------|---------|-------|-------------|----------|--------| | "I bought every M4 Mac to find the best one" | HardwareChannelA | 1.2M | 210K | 5.7x | Buy the full range, one task, one winner | | "Llama runs on a Raspberry Pi. Here is how slow" | MLChannelB | 480K | 60K | 8.0x | Big model on the smallest box, show the pain |

Adaptations

  1. Every Mac RAM tier runs the same coding agent. One wins. from "full range, one task". Angle: agent latency, not benchmarks. Budget: borrow the machines.
  2. Claude Code on an 8GB Mac with a local model from "smallest box, show the pain". Angle: what breaks first. Budget: zero.
  3. One 128GB Mac vs five cheap Minis for a team of agents from "full range, one task". Angle: cost per agent. Budget: rental.

Top pick: #2. Cheapest to film and the pain is the thumbnail.

CRITICAL: never present a guessed view count as measured. Mark estimates.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Write a review

Versions

  • v0.1.0 Imported from the upstream source.

More in Content & Media