Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Downloads videos from YouTube and other platforms for offline viewing, editing, or archival. Handles various formats and quality options.
.claude/skills/davila7-video-downloader/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 73% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 49% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 162% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 28% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 44% | 0% |
This skill downloads videos from YouTube and other platforms directly to your computer.
Download this YouTube video: https://youtube.com/watch?v=...Download this video in 1080p qualityDownload the audio from this YouTube video as MP3Download all videos from this YouTube playlist: [URL]Download these 5 YouTube videos:
1. [URL]
2. [URL]
...User: "Download this YouTube video: https://youtube.com/watch?v=abc123"
Output:
Downloading from YouTube...
Video: "How to Build Products Users Love"
Channel: Lenny's Podcast
Duration: 45:32
Quality: 1080p
Progress: ████████████████████ 100%
✓ Downloaded: how-to-build-products-users-love.mp4
✓ Saved thumbnail: how-to-build-products-users-love.jpg
✓ Size: 342 MB
Saved to: ~/Downloads/Inspired by: Lenny's workflow from his newsletter
⚠️ Copyright & Fair Use
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 4,260 | 3,632 | -15% | 1 | 1 | 0% | 725 | 1,252 | +73% | 0 | 0 | — |
case-02 | fail→pass | 4,776 | 3,421 | -28% | 1 | 1 | 0% | 857 | 1,273 | +49% | 0 | 0 | — |
case-03 | fail→pass | 4,680 | 7,868 | +68% | 1 | 1 | 0% | 810 | 2,124 | +162% | 0 | 0 | — |
case-04 | pass→pass | 6,411 | 7,051 | +10% | 1 | 1 | 0% | 1,118 | 1,864 | +67% | 0 | 0 | — |
case-05 | pass→pass | 6,473 | 6,466 | -0% | 1 | 1 | 0% | 1,175 | 1,771 | +51% | 0 | 0 | — |
case-06 | fail→fail | 5,059 | 6,289 | +24% | 1 | 1 | 0% | 785 | 1,644 | +109% | 0 | 0 | — |
case-07 | pass→pass | 3,731 | 7,300 | +96% | 1 | 1 | 0% | 586 | 1,750 | +199% | 0 | 0 | — |
case-08 | fail→pass | 5,845 | 3,861 | -34% | 1 | 1 | 0% | 959 | 1,228 | +28% | 0 | 0 | — |
case-09 | fail→pass | 5,407 | 4,475 | -17% | 1 | 1 | 0% | 1,029 | 1,486 | +44% | 0 | 0 | — |
case-10 | pass→pass | 6,284 | 8,853 | +41% | 1 | 1 | 0% | 1,016 | 2,126 | +109% | 0 | 0 | — |
case-11 | fail→pass | 6,697 | 3,404 | -49% | 1 | 1 | 0% | 1,155 | 1,167 | +1% | 0 | 0 | — |
case-12 | pass→pass | 5,336 | 5,628 | +5% | 1 | 1 | 0% | 947 | 1,641 | +73% | 0 | 0 | — |
case-13 | pass→pass | 9,200 | 2,507 | -73% | 1 | 1 | 0% | 1,068 | 1,046 | -2% | 0 | 0 | — |
case-14 | pass→pass | 13,167 | 7,602 | -42% | 1 | 1 | 0% | 2,147 | 1,874 | -13% | 0 | 0 | — |
case-15 | pass→pass | 7,699 | 4,517 | -41% | 1 | 1 | 0% | 1,461 | 1,418 | -3% | 0 | 0 | — |
case-16 | pass→pass | 3,431 | 2,635 | -23% | 1 | 1 | 0% | 615 | 1,187 | +93% | 0 | 0 | — |
case-17 | fail→pass | 3,510 | 3,577 | +2% | 1 | 1 | 0% | 657 | 1,240 | +89% | 0 | 0 | — |
case-18 | pass→pass | 3,831 | 4,815 | +26% | 1 | 1 | 0% | 685 | 1,502 | +119% | 0 | 0 | — |
case-19 | fail→pass | 4,187 | 7,344 | +75% | 1 | 1 | 0% | 696 | 1,919 | +176% | 0 | 0 | — |
case-20 | fail→pass | 5,428 | 4,009 | -26% | 1 | 1 | 0% | 1,010 | 1,309 | +30% | 0 | 0 | — |
case-21 | fail→pass | 6,014 | 5,167 | -14% | 1 | 1 | 0% | 1,098 | 1,594 | +45% | 0 | 0 | — |
case-22 | fail→pass | 3,124 | 3,373 | +8% | 1 | 1 | 0% | 476 | 1,239 | +160% | 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. 22 cases were attempted. The headline lift of +50 percentage points is the difference between those two pass rates over the 22 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.