Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Connect NotFair paid-ad accounts and establish reusable campaign context. Use when setting up paid media in a new workspace, connecting Google or Meta, refreshing brand context, or preparing an agent to work on ads safely.
.claude/skills/nowork-studio-paid-ads-setup/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -40% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -39% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -45% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -49% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -8% | 0% |
Read ../../paid-ads/paid-ads-setup/SKILL.md completely, then follow it as the active workflow. Normalize that path from the directory containing this wrapper: the canonical file is <plugin-root>/paid-ads/paid-ads-setup/SKILL.md, not <plugin-root>/skills/paid-ads-setup/paid-ads-setup/SKILL.md. Resolve every relative reference from the canonical file against <plugin-root>/paid-ads/paid-ads-setup/. If the canonical file cannot be read, stop and report the packaging error; never substitute a similarly named skill from another plugin.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 21,989 | 34,793 | +58% | 1 | 1 | 0% | 3,133 | 5,755 | +84% | 0 | 0 | — |
case-02 | fail→pass | 24,009 | 11,247 | -53% | 1 | 1 | 0% | 3,267 | 1,952 | -40% | 0 | 0 | — |
case-03 | fail→fail | 35,523 | 6,797 | -81% | 1 | 1 | 0% | 5,075 | 498 | -90% | 0 | 0 | — |
case-04 | fail→pass | 28,311 | 17,466 | -38% | 1 | 1 | 0% | 3,388 | 2,083 | -39% | 0 | 0 | — |
case-05 | fail→fail | 32,251 | 7,609 | -76% | 1 | 1 | 0% | 4,267 | 509 | -88% | 0 | 0 | — |
case-06 | fail→pass | 22,211 | 20,359 | -8% | 1 | 1 | 0% | 3,145 | 1,719 | -45% | 0 | 0 | — |
case-07 | fail→fail | 56,788 | 7,659 | -87% | 1 | 1 | 0% | 8,220 | 464 | -94% | 0 | 0 | — |
case-08 | fail→fail | 36,515 | 36,590 | +0% | 1 | 1 | 0% | 5,524 | 418 | -92% | 0 | 0 | — |
case-09 | fail→fail | 41,535 | 6,390 | -85% | 1 | 1 | 0% | 3,427 | 423 | -88% | 0 | 0 | — |
case-10 | fail→fail | 60,176 | 40,845 | -32% | 1 | 1 | 0% | 4,811 | 5,688 | +18% | 0 | 0 | — |
case-11 | fail→pass | 24,265 | 11,931 | -51% | 1 | 1 | 0% | 3,892 | 1,974 | -49% | 0 | 0 | — |
case-12 | fail→fail | 20,454 | 34,543 | +69% | 1 | 1 | 0% | 2,968 | 5,507 | +86% | 0 | 0 | — |
case-13 | fail→fail | 25,641 | 6,095 | -76% | 1 | 1 | 0% | 3,649 | 494 | -86% | 0 | 0 | — |
case-14 | fail→fail | 54,691 | 40,141 | -27% | 1 | 1 | 0% | 7,971 | 5,926 | -26% | 0 | 0 | — |
case-15 | fail→fail | 36,188 | 42,150 | +16% | 1 | 1 | 0% | 5,528 | 7,122 | +29% | 0 | 0 | — |
case-16 | fail→pass | 26,301 | 56,510 | +115% | 1 | 1 | 0% | 3,974 | 3,649 | -8% | 0 | 0 | — |
case-17 | fail→fail | 26,177 | 7,228 | -72% | 1 | 1 | 0% | 3,608 | 379 | -89% | 0 | 0 | — |
case-18 | fail→fail | 28,589 | 76,013 | +166% | 1 | 1 | 0% | 4,085 | 8,310 | +103% | 0 | 0 | — |
case-19 | fail→fail | 25,038 | 6,729 | -73% | 1 | 1 | 0% | 3,685 | 543 | -85% | 0 | 0 | — |
case-20 | pass→pass | 10,908 | 7,456 | -32% | 1 | 1 | 0% | 1,590 | 1,055 | -34% | 0 | 0 | — |
case-21 | pass→pass | 8,566 | 9,101 | +6% | 1 | 1 | 0% | 1,364 | 1,694 | +24% | 0 | 0 | — |
case-22 | pass→pass | 17,972 | 19,981 | +11% | 1 | 1 | 0% | 2,454 | 2,652 | +8% | 0 | 0 | — |
case-23 | pass→pass | 15,006 | 17,373 | +16% | 1 | 1 | 0% | 2,218 | 2,500 | +13% | 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, and 15 counted toward the lift figure. The other 8 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +22 percentage points is the difference between those two pass rates over the 15 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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 9/9/2026 | +36% |
| gemini-3.6-flash | verified | 8/29/2026 | +13% |
Other measured skills in the registry, with their headline benchmark lift.