Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Get a local business found on Google -- audit and optimize the Google Business Profile, local pack ranking factors, review velocity, NAP consistency across directories, and location pages. For businesses whose customers search "near me," this outranks everything else in marketing.
.claude/skills/onewave-ai-local-seo-optimizer/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | -24% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 25% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -16% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 31% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -8% | 0% |
For a local business, the Google Business Profile is the real homepage and the local pack (map results) is the real search ranking. This skill audits the three local ranking pillars -- relevance, distance, prominence -- and produces the fix list. Complements seo-optimizer (content/on-page SEO); this is the maps-and-profile layer.
review-response-writer for the response side.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-04 | fail→fail | 10,242 | 11,043 | +8% | 1 | 1 | 0% | 1,606 | 2,482 | +55% | 0 | 0 | — |
case-05 | fail→pass | 23,436 | 15,518 | -34% | 1 | 1 | 0% | 3,795 | 2,877 | -24% | 0 | 0 | — |
case-01 | fail→fail | 21,381 | 23,202 | +9% | 1 | 1 | 0% | 3,113 | 4,177 | +34% | 0 | 0 | — |
case-02 | fail→pass | 16,444 | 16,923 | +3% | 1 | 1 | 0% | 2,604 | 3,268 | +25% | 0 | 0 | — |
case-03 | fail→fail | 23,225 | 17,069 | -27% | 1 | 1 | 0% | 3,777 | 3,532 | -6% | 0 | 0 | — |
case-23 | fail→fail | 16,527 | 17,428 | +5% | 1 | 1 | 0% | 2,378 | 3,785 | +59% | 0 | 0 | — |
case-06 | fail→fail | 15,299 | 15,832 | +3% | 1 | 1 | 0% | 2,857 | 3,408 | +19% | 0 | 0 | — |
case-07 | pass→pass | 8,858 | 7,016 | -21% | 1 | 1 | 0% | 1,592 | 1,846 | +16% | 0 | 0 | — |
case-08 | fail→pass | 15,305 | 9,856 | -36% | 1 | 1 | 0% | 2,658 | 2,241 | -16% | 0 | 0 | — |
case-09 | pass→pass | 11,389 | 8,305 | -27% | 1 | 1 | 0% | 1,987 | 2,061 | +4% | 0 | 0 | — |
case-10 | pass→pass | 14,331 | 10,942 | -24% | 1 | 1 | 0% | 2,103 | 2,273 | +8% | 0 | 0 | — |
case-11 | pass→pass | 15,634 | 12,169 | -22% | 1 | 1 | 0% | 2,276 | 2,472 | +9% | 0 | 0 | — |
case-12 | fail→pass | 9,888 | 8,459 | -14% | 1 | 1 | 0% | 1,483 | 1,940 | +31% | 0 | 0 | — |
case-13 | pass→pass | 16,039 | 12,366 | -23% | 1 | 1 | 0% | 2,346 | 2,583 | +10% | 0 | 0 | — |
case-14 | pass→fail | 12,771 | 10,148 | -21% | 1 | 1 | 0% | 2,195 | 2,324 | +6% | 0 | 0 | — |
case-15 | fail→pass | 14,114 | 9,428 | -33% | 1 | 1 | 0% | 2,335 | 2,154 | -8% | 0 | 0 | — |
case-16 | fail→fail | 12,130 | 11,287 | -7% | 1 | 1 | 0% | 2,140 | 2,649 | +24% | 0 | 0 | — |
case-17 | pass→pass | 14,135 | 11,322 | -20% | 1 | 1 | 0% | 2,918 | 3,016 | +3% | 0 | 0 | — |
case-18 | pass→pass | 12,537 | 9,505 | -24% | 1 | 1 | 0% | 2,471 | 2,471 | 0% | 0 | 0 | — |
case-19 | pass→pass | 10,114 | 6,189 | -39% | 1 | 1 | 0% | 1,745 | 1,698 | -3% | 0 | 0 | — |
case-20 | fail→fail | 10,289 | 9,243 | -10% | 1 | 1 | 0% | 1,786 | 2,135 | +20% | 0 | 0 | — |
case-21 | fail→pass | 11,833 | 9,354 | -21% | 1 | 1 | 0% | 2,055 | 2,287 | +11% | 0 | 0 | — |
case-22 | fail→pass | 12,814 | 10,049 | -22% | 1 | 1 | 0% | 2,167 | 2,283 | +5% | 0 | 0 | — |
case-24 | fail→fail | 8,248 | 13,076 | +59% | 1 | 1 | 0% | 1,263 | 2,491 | +97% | 0 | 0 | — |
case-25 | fail→pass | 16,493 | 12,169 | -26% | 1 | 1 | 0% | 2,587 | 2,453 | -5% | 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. 25 cases were attempted. The headline lift of +28 percentage points is the difference between those two pass rates over the 25 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.