Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Analyze YouTube channels for outlier videos and packaging patterns. Identifies what's working (2x+ average views) across any set of channels. Use when asked for YouTube competitive analysis, viral video patterns, or packaging/title inspiration.
.claude/skills/evolution-foundation-social-yt-competitive/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 35% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 102% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -62% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -57% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -33% | 0% |
Outlier detection and packaging pattern extraction for YouTube channels.
$YOUTUBE_API_KEYbash# Analyze specific channels python3 analyze.py "$YOUTUBE_API_KEY" --channels "@handle1,@handle2" --days 30 # Use predefined sets python3 analyze.py "$YOUTUBE_API_KEY" --set ai python3 analyze.py "$YOUTUBE_API_KEY" --set business python3 analyze.py "$YOUTUBE_API_KEY" --set both # Export formats python3 analyze.py "$YOUTUBE_API_KEY" --set both --output json python3 analyze.py "$YOUTUBE_API_KEY" --set both --output console
AI Creators: Jeff Su, Alex Finn, Riley Brown, Dan Martell, Matt Wolfe, Nate Herk, Grace Leung, Matt Berman
Business Creators: Alex Hormozi, Gary Vaynerchuk, Patrick Bet-David, Codie Sanchez, Leila Hormozi, Iman Gadzhi, My First Million
Long-form:
Shorts:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-02 | fail→pass | 18,893 | 20,050 | +6% | 1 | 1 | 0% | 3,268 | 4,406 | +35% | 0 | 0 | — |
case-03 | fail→fail | 15,248 | 2,898 | -81% | 1 | 1 | 0% | 2,661 | 824 | -69% | 0 | 0 | — |
case-01 | fail→pass | 18,949 | 31,775 | +68% | 1 | 1 | 0% | 3,312 | 6,684 | +102% | 0 | 0 | — |
case-04 | fail→pass | 12,857 | 2,365 | -82% | 1 | 1 | 0% | 1,990 | 755 | -62% | 0 | 0 | — |
case-05 | pass→pass | 15,261 | 10,076 | -34% | 1 | 1 | 0% | 2,459 | 2,090 | -15% | 0 | 0 | — |
case-06 | fail→pass | 11,613 | 2,049 | -82% | 1 | 1 | 0% | 1,649 | 708 | -57% | 0 | 0 | — |
case-07 | pass→pass | 7,965 | 1,754 | -78% | 1 | 1 | 0% | 1,194 | 715 | -40% | 0 | 0 | — |
case-22 | pass→pass | 15,087 | 16,802 | +11% | 1 | 1 | 0% | 3,017 | 3,900 | +29% | 0 | 0 | — |
case-08 | fail→pass | 12,379 | 5,901 | -52% | 1 | 1 | 0% | 2,083 | 1,403 | -33% | 0 | 0 | — |
case-09 | fail→pass | 12,317 | 2,769 | -78% | 1 | 1 | 0% | 1,992 | 915 | -54% | 0 | 0 | — |
case-10 | fail→pass | 11,566 | 9,987 | -14% | 1 | 1 | 0% | 1,736 | 1,998 | +15% | 0 | 0 | — |
case-11 | fail→pass | 10,954 | 3,987 | -64% | 1 | 1 | 0% | 1,657 | 1,143 | -31% | 0 | 0 | — |
case-12 | fail→pass | 12,728 | 6,714 | -47% | 1 | 1 | 0% | 1,929 | 1,459 | -24% | 0 | 0 | — |
case-13 | fail→pass | 10,576 | 3,058 | -71% | 1 | 1 | 0% | 1,802 | 1,022 | -43% | 0 | 0 | — |
case-14 | fail→pass | 10,228 | 2,290 | -78% | 1 | 1 | 0% | 1,553 | 839 | -46% | 0 | 0 | — |
case-15 | fail→pass | 10,818 | 1,591 | -85% | 1 | 1 | 0% | 1,667 | 717 | -57% | 0 | 0 | — |
case-16 | fail→pass | 17,543 | 2,716 | -85% | 1 | 1 | 0% | 1,272 | 912 | -28% | 0 | 0 | — |
case-17 | fail→pass | 4,752 | 2,089 | -56% | 1 | 1 | 0% | 754 | 795 | +5% | 0 | 0 | — |
case-18 | pass→pass | 9,792 | 2,578 | -74% | 1 | 1 | 0% | 1,505 | 817 | -46% | 0 | 0 | — |
case-19 | pass→pass | 13,256 | 11,245 | -15% | 1 | 1 | 0% | 1,930 | 2,235 | +16% | 0 | 0 | — |
case-20 | pass→pass | 4,036 | 1,518 | -62% | 1 | 1 | 0% | 614 | 718 | +17% | 0 | 0 | — |
case-21 | pass→pass | 10,469 | 8,970 | -14% | 1 | 1 | 0% | 2,162 | 2,541 | +18% | 0 | 0 | — |
case-23 | pass→pass | 4,369 | 3,496 | -20% | 1 | 1 | 0% | 720 | 997 | +38% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted. The headline lift of +61 percentage points is the difference between those two pass rates over the 23 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.