Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →When the user wants to run YouTube ads, set up TrueView or Bumper campaigns, or optimize video ad creative. Also use when the user mentions "YouTube ads," "TrueView," "Bumper ads," "YouTube Discovery," "video ads," "YouTube campaign," or "in-feed ads." For organic YouTube, use youtube-seo.
.claude/skills/kostja94-youtube-ads/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -3% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 72% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 7% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -20% | 0% |
| case-20 | ✗→✓ | ▲ Improved | 22% | 0% |
Guides YouTube advertising: TrueView, Bumper, and Discovery (in-feed) formats. Use this skill when planning or optimizing YouTube ad campaigns. For Google Search/Display/PMax, see google-ads. For organic YouTube optimization, see youtube-seo.
When invoking: On first use, if helpful, open with 1–2 sentences on what this skill covers and why it matters, then provide the main output. On subsequent use or when the user asks to skip, go directly to the main output.
Check for project context first: If .claude/project-context.md or .cursor/project-context.md exists, read Sections 3 (Value Proposition), 4 (Audience).
Identify:
| Format | Length | Skippable | Best For | |--------|--------|-----------|----------| | TrueView Skippable | 12 sec+ | After 5 sec | Reach; large audiences | | TrueView Non-skippable | 15–20 sec | No | Full message; guaranteed view | | TrueView for Action | — | — | Conversions; CTA-focused | | Bumper | 6 sec or less | No | Brand awareness; memorable; lower cost | | Discovery (In-Feed) | Thumbnail + text | — | Search, related videos, homepage; interest targeting |
| Goal | Format | |------|--------| | Brand awareness | Bumper | | Website traffic | TrueView | | Conversions | TrueView for Action | | Interested audiences | Discovery |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 16,371 | 12,741 | -22% | 1 | 1 | 0% | 2,765 | 2,684 | -3% | 0 | 0 | — |
case-17 | pass→pass | 9,733 | 8,452 | -13% | 1 | 1 | 0% | 1,774 | 1,965 | +11% | 0 | 0 | — |
case-02 | fail→fail | 15,411 | 10,831 | -30% | 1 | 1 | 0% | 2,793 | 2,431 | -13% | 0 | 0 | — |
case-03 | pass→pass | 13,500 | 9,909 | -27% | 1 | 1 | 0% | 2,475 | 2,441 | -1% | 0 | 0 | — |
case-04 | pass→pass | 14,142 | 8,108 | -43% | 1 | 1 | 0% | 2,643 | 1,898 | -28% | 0 | 0 | — |
case-05 | pass→fail | 10,937 | 11,818 | +8% | 1 | 1 | 0% | 1,935 | 2,504 | +29% | 0 | 0 | — |
case-06 | pass→pass | 13,111 | 9,794 | -25% | 1 | 1 | 0% | 2,222 | 2,101 | -5% | 0 | 0 | — |
case-07 | fail→pass | 7,264 | 9,427 | +30% | 1 | 1 | 0% | 1,284 | 2,209 | +72% | 0 | 0 | — |
case-08 | pass→pass | 6,041 | 7,638 | +26% | 1 | 1 | 0% | 1,013 | 2,069 | +104% | 0 | 0 | — |
case-09 | pass→pass | 4,590 | 6,904 | +50% | 1 | 1 | 0% | 879 | 1,863 | +112% | 0 | 0 | — |
case-10 | fail→pass | 13,268 | 10,651 | -20% | 1 | 1 | 0% | 2,332 | 2,503 | +7% | 0 | 0 | — |
case-11 | fail→fail | 10,346 | 8,889 | -14% | 1 | 1 | 0% | 1,627 | 2,157 | +33% | 0 | 0 | — |
case-12 | pass→pass | 3,617 | 4,792 | +32% | 1 | 1 | 0% | 619 | 1,356 | +119% | 0 | 0 | — |
case-13 | pass→pass | 4,798 | 5,110 | +7% | 1 | 1 | 0% | 897 | 1,524 | +70% | 0 | 0 | — |
case-14 | pass→pass | 4,017 | 5,060 | +26% | 1 | 1 | 0% | 723 | 1,457 | +102% | 0 | 0 | — |
case-15 | fail→pass | 17,633 | 10,184 | -42% | 1 | 1 | 0% | 2,959 | 2,379 | -20% | 0 | 0 | — |
case-16 | fail→fail | 12,768 | 9,558 | -25% | 1 | 1 | 0% | 2,148 | 2,128 | -1% | 0 | 0 | — |
case-18 | pass→pass | 6,074 | 7,712 | +27% | 1 | 1 | 0% | 1,010 | 1,961 | +94% | 0 | 0 | — |
case-19 | pass→pass | 5,104 | 4,601 | -10% | 1 | 1 | 0% | 946 | 1,323 | +40% | 0 | 0 | — |
case-20 | fail→pass | 8,968 | 7,330 | -18% | 1 | 1 | 0% | 1,553 | 1,897 | +22% | 0 | 0 | — |
case-21 | pass→pass | 10,030 | 6,228 | -38% | 1 | 1 | 0% | 1,818 | 1,675 | -8% | 0 | 0 | — |
case-22 | pass→pass | 6,930 | 5,052 | -27% | 1 | 1 | 0% | 1,195 | 1,481 | +24% | 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 +18 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
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.