Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Meta Ads deep analysis covering Facebook and Instagram advertising. Evaluates 46 checks across Pixel/CAPI health, creative diversity and fatigue, account structure, and audience targeting. Includes Advantage+ assessment. Triggers on: "Meta Ads", "Facebook Ads", "Instagram Ads", "Advantage+", "Meta campaign", "Meta audit", "FB ads"
.claude/skills/miosa-osa-ads-meta/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-17 | ✗→✓ | ▲ Improved | 55% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 13% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -1% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -4% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -23% | 0% |
> 46-check audit of Meta (Facebook/Instagram) advertising accounts.
bash/ads meta
| EMQ Score | Status | Action | |-----------|--------|--------| | 8.0-10.0 | Excellent | Maintain current setup | | 6.0-7.9 | Good | Add more customer_information parameters | | 4.0-5.9 | Fair | Implement CAPI, improve data quality | | <4.0 | Poor | Critical: CAPI + Enhanced Matching required |
| Metric | Pass | Warning | Fail | |--------|------|---------|------| | EMQ (Purchase) | >=8.0 | 6.0-7.9 | <6.0 | | Dedup rate | >=90% | 70-90% | <70% | | CTR | >=1.0% | 0.5-1.0% | <0.5% | | Creative formats | >=3 | 2 | 1 | | Creatives per ad set | >=5 | 3-4 | <3 | | Learning Limited | <30% | 30-50% | >50% | | Budget per ad set | >=5x CPA | 2-5x CPA | <2x CPA |
META-ADS-REPORT.md — Full 46-check findings with pass/warning/fail| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-17 | fail→pass | 15,480 | 14,532 | -6% | 1 | 1 | 0% | 2,182 | 3,385 | +55% | 0 | 0 | — |
case-01 | fail→pass | 48,569 | 44,864 | -8% | 1 | 1 | 0% | 8,284 | 9,347 | +13% | 0 | 0 | — |
case-02 | fail→pass | 45,539 | 49,567 | +9% | 1 | 1 | 0% | 8,260 | 8,155 | -1% | 0 | 0 | — |
case-03 | fail→pass | 46,213 | 38,581 | -17% | 1 | 1 | 0% | 8,261 | 7,940 | -4% | 0 | 0 | — |
case-04 | fail→pass | 22,994 | 10,598 | -54% | 1 | 1 | 0% | 3,584 | 2,767 | -23% | 0 | 0 | — |
case-05 | fail→fail | 29,044 | 31,478 | +8% | 1 | 1 | 0% | 3,828 | 6,252 | +63% | 0 | 0 | — |
case-06 | fail→fail | 46,512 | 33,781 | -27% | 1 | 1 | 0% | 8,232 | 7,303 | -11% | 0 | 0 | — |
case-07 | pass→pass | 12,793 | 10,504 | -18% | 1 | 1 | 0% | 1,989 | 2,470 | +24% | 0 | 0 | — |
case-16 | pass→pass | 7,673 | 67,367 | +778% | 1 | 1 | 0% | 1,439 | 2,158 | +50% | 0 | 0 | — |
case-08 | pass→pass | 13,521 | 6,733 | -50% | 1 | 1 | 0% | 1,825 | 2,143 | +17% | 0 | 0 | — |
case-09 | fail→pass | 18,917 | 18,834 | -0% | 1 | 1 | 0% | 2,275 | 2,898 | +27% | 0 | 0 | — |
case-10 | pass→pass | 9,152 | 7,111 | -22% | 1 | 1 | 0% | 1,612 | 1,906 | +18% | 0 | 0 | — |
case-11 | pass→pass | 12,652 | 9,246 | -27% | 1 | 1 | 0% | 2,032 | 2,488 | +22% | 0 | 0 | — |
case-12 | pass→pass | 10,063 | 7,026 | -30% | 1 | 1 | 0% | 2,045 | 2,090 | +2% | 0 | 0 | — |
case-13 | fail→pass | 9,905 | 6,962 | -30% | 1 | 1 | 0% | 1,627 | 2,196 | +35% | 0 | 0 | — |
case-14 | pass→pass | 17,087 | 7,176 | -58% | 1 | 1 | 0% | 1,913 | 2,221 | +16% | 0 | 0 | — |
case-15 | pass→pass | 15,305 | 11,965 | -22% | 1 | 1 | 0% | 2,529 | 3,488 | +38% | 0 | 0 | — |
case-18 | fail→fail | 16,379 | 31,489 | +92% | 1 | 1 | 0% | 2,394 | 3,199 | +34% | 0 | 0 | — |
case-19 | pass→pass | 14,376 | 13,778 | -4% | 1 | 1 | 0% | 1,874 | 2,817 | +50% | 0 | 0 | — |
case-20 | pass→pass | 9,718 | 4,918 | -49% | 1 | 1 | 0% | 1,694 | 1,907 | +13% | 0 | 0 | — |
case-21 | fail→pass | 7,033 | 2,770 | -61% | 1 | 1 | 0% | 830 | 1,473 | +77% | 0 | 0 | — |
case-22 | fail→pass | 10,179 | 4,314 | -58% | 1 | 1 | 0% | 1,973 | 1,912 | -3% | 0 | 0 | — |
case-23 | fail→pass | 10,827 | 6,205 | -43% | 1 | 1 | 0% | 1,564 | 1,995 | +28% | 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 +43 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.