Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Input your best customers and find 100+ companies that match the profile. Uses firmographic data, tech stack, growth signals, and similarity scoring to identify ideal prospects. Use when building target account lists or expanding to new markets.
.claude/skills/onewave-ai-lookalike-customer-finder/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -33% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 126% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 161% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 157% | 0% |
| case-20 | ✓→✓ | ＝ Same ✓ | 15% | 0% |
Analyze a company's best customers and find similar companies that match the same profile, producing a high-quality, ranked target account list.
references/scoring-model.md - Profile dimensions, weighted scoring model, and score bands.references/output-template.md - Full Markdown report structure (ICP, ranked lookalikes, market insights, targeting strategy, action plan).references/data-sources.md - Recommended enrichment tools and data points to gather.references/examples.md - Best practices, trigger phrases, and an example request.references/scoring-model.md for the five profile dimensions.references/data-sources.md.references/scoring-model.md.references/output-template.md, including market insights, a tiered targeting strategy, and a quick-start action plan.references/examples.md throughout: favor quality over quantity, weight growth signals, and enrich contacts before recommending outreach.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-09 | fail→fail | 15,792 | 36,223 | +129% | 1 | 1 | 0% | 2,725 | 6,979 | +156% | 0 | 0 | — |
case-10 | fail→fail | 3,321 | 8,544 | +157% | 1 | 1 | 0% | 543 | 1,889 | +248% | 0 | 0 | — |
case-01 | fail→fail | 48,722 | 53,457 | +10% | 1 | 1 | 0% | 8,280 | 8,653 | +5% | 0 | 0 | — |
case-02 | fail→fail | 44,662 | 46,923 | +5% | 1 | 1 | 0% | 7,499 | 8,405 | +12% | 0 | 0 | — |
case-03 | fail→fail | 47,098 | 43,891 | -7% | 1 | 1 | 0% | 7,666 | 8,647 | +13% | 0 | 0 | — |
case-04 | fail→pass | 9,578 | 3,817 | -60% | 1 | 1 | 0% | 1,535 | 1,030 | -33% | 0 | 0 | — |
case-20 | pass→pass | 16,040 | 16,580 | +3% | 1 | 1 | 0% | 2,543 | 2,916 | +15% | 0 | 0 | — |
case-05 | fail→fail | 22,925 | 43,759 | +91% | 1 | 1 | 0% | 3,866 | 8,258 | +114% | 0 | 0 | — |
case-06 | fail→fail | 11,542 | 9,120 | -21% | 1 | 1 | 0% | 1,861 | 1,959 | +5% | 0 | 0 | — |
case-07 | fail→pass | 22,460 | 44,192 | +97% | 1 | 1 | 0% | 3,799 | 8,595 | +126% | 0 | 0 | — |
case-08 | fail→fail | 13,826 | 42,144 | +205% | 1 | 1 | 0% | 2,328 | 8,611 | +270% | 0 | 0 | — |
case-11 | fail→pass | 12,912 | 30,131 | +133% | 1 | 1 | 0% | 2,117 | 5,524 | +161% | 0 | 0 | — |
case-12 | fail→fail | 15,034 | 40,245 | +168% | 1 | 1 | 0% | 2,796 | 7,718 | +176% | 0 | 0 | — |
case-13 | fail→fail | 6,235 | 8,861 | +42% | 1 | 1 | 0% | 957 | 1,809 | +89% | 0 | 0 | — |
case-14 | fail→fail | 19,258 | 45,149 | +134% | 1 | 1 | 0% | 2,905 | 8,601 | +196% | 0 | 0 | — |
case-15 | fail→pass | 16,335 | 37,164 | +128% | 1 | 1 | 0% | 2,640 | 6,793 | +157% | 0 | 0 | — |
case-16 | fail→fail | 8,747 | 25,224 | +188% | 1 | 1 | 0% | 1,605 | 4,983 | +210% | 0 | 0 | — |
case-17 | fail→fail | 17,145 | 31,039 | +81% | 1 | 1 | 0% | 3,016 | 5,824 | +93% | 0 | 0 | — |
case-18 | fail→fail | 24,730 | 44,659 | +81% | 1 | 1 | 0% | 4,280 | 8,598 | +101% | 0 | 0 | — |
case-19 | fail→fail | 18,615 | 40,074 | +115% | 1 | 1 | 0% | 3,080 | 7,342 | +138% | 0 | 0 | — |
case-21 | pass→pass | 20,819 | 19,394 | -7% | 1 | 1 | 0% | 3,725 | 3,994 | +7% | 0 | 0 | — |
case-22 | pass→pass | 18,929 | 31,599 | +67% | 1 | 1 | 0% | 3,001 | 4,746 | +58% | 0 | 0 | — |
case-23 | pass→pass | 14,999 | 17,471 | +16% | 1 | 1 | 0% | 2,724 | 3,613 | +33% | 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 +17 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.