Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Techniques for expanding seed keywords and clustering by topic and intent. Use when building keyword lists, planning content calendars, or identifying topic clusters for pillar content strategy.
.claude/skills/nicepkg-keyword-cluster-builder/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 24% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -13% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -26% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -18% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -20% | 0% |
PILLAR: "content marketing" (highest volume)
+-- CLUSTER: "content marketing strategy" (commercial)
| +-- content marketing plan template
| +-- content marketing framework
| +-- how to create content marketing strategy
+-- CLUSTER: "content marketing examples" (informational)
| +-- B2B content marketing examples
| +-- content marketing case studies
| +-- content marketing success stories
+-- CLUSTER: "content marketing tools" (commercial)
+-- best content marketing tools
+-- content marketing software
+-- content marketing platforms| Signal | Intent | |--------|--------| | "what is", "how to", "guide" | Informational | | "best", "vs", "review", "compare" | Commercial | | "buy", "price", "discount", brand | Transactional | | Brand name, specific product | Navigational |
When generating keyword clusters, use this format:
markdown## Keyword Cluster Report **Seed Keyword**: {seed} **Total Keywords**: {count} **Clusters**: {cluster_count} ### Cluster 1: {cluster_name} **Intent**: {intent} **Funnel Stage**: {stage} **Keywords**: 1. {keyword1} - {estimated_volume} 2. {keyword2} - {estimated_volume} ... ### Cluster 2: {cluster_name} ...
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-22 | pass→fail | 16,335 | 14,795 | -9% | 1 | 1 | 0% | 2,746 | 2,951 | +7% | 0 | 0 | — |
case-20 | pass→fail | 18,044 | 19,476 | +8% | 1 | 1 | 0% | 2,764 | 3,614 | +31% | 0 | 0 | — |
case-01 | fail→pass | 16,655 | 36,453 | +119% | 1 | 1 | 0% | 2,882 | 3,575 | +24% | 0 | 0 | — |
case-02 | fail→pass | 19,139 | 11,437 | -40% | 1 | 1 | 0% | 3,404 | 2,950 | -13% | 0 | 0 | — |
case-03 | fail→fail | 22,062 | 12,277 | -44% | 1 | 1 | 0% | 4,052 | 3,092 | -24% | 0 | 0 | — |
case-04 | fail→fail | 15,922 | 9,203 | -42% | 1 | 1 | 0% | 2,854 | 2,588 | -9% | 0 | 0 | — |
case-05 | fail→pass | 14,813 | 6,403 | -57% | 1 | 1 | 0% | 2,485 | 1,836 | -26% | 0 | 0 | — |
case-06 | fail→pass | 11,463 | 5,181 | -55% | 1 | 1 | 0% | 1,852 | 1,525 | -18% | 0 | 0 | — |
case-07 | pass→pass | 11,041 | 4,575 | -59% | 1 | 1 | 0% | 1,718 | 1,361 | -21% | 0 | 0 | — |
case-21 | pass→pass | 7,341 | 9,180 | +25% | 1 | 1 | 0% | 1,390 | 2,345 | +69% | 0 | 0 | — |
case-08 | fail→pass | 15,937 | 7,698 | -52% | 1 | 1 | 0% | 2,696 | 2,164 | -20% | 0 | 0 | — |
case-09 | pass→pass | 16,455 | 8,286 | -50% | 1 | 1 | 0% | 2,827 | 2,312 | -18% | 0 | 0 | — |
case-10 | pass→pass | 12,925 | 8,737 | -32% | 1 | 1 | 0% | 2,421 | 2,293 | -5% | 0 | 0 | — |
case-11 | pass→pass | 18,395 | 8,815 | -52% | 1 | 1 | 0% | 2,836 | 2,333 | -18% | 0 | 0 | — |
case-12 | pass→pass | 8,192 | 4,238 | -48% | 1 | 1 | 0% | 1,234 | 1,267 | +3% | 0 | 0 | — |
case-13 | pass→fail | 17,022 | 13,190 | -23% | 1 | 1 | 0% | 2,879 | 3,203 | +11% | 0 | 0 | — |
case-14 | pass→pass | 17,585 | 9,789 | -44% | 1 | 1 | 0% | 3,082 | 2,477 | -20% | 0 | 0 | — |
case-15 | pass→pass | 20,456 | 9,575 | -53% | 1 | 1 | 0% | 3,821 | 2,526 | -34% | 0 | 0 | — |
case-16 | fail→pass | 12,586 | 7,586 | -40% | 1 | 1 | 0% | 2,129 | 2,034 | -4% | 0 | 0 | — |
case-17 | pass→pass | 17,967 | 10,238 | -43% | 1 | 1 | 0% | 3,193 | 2,661 | -17% | 0 | 0 | — |
case-18 | pass→pass | 14,926 | 9,153 | -39% | 1 | 1 | 0% | 2,502 | 2,245 | -10% | 0 | 0 | — |
case-19 | pass→pass | 13,982 | 7,933 | -43% | 1 | 1 | 0% | 2,424 | 2,089 | -14% | 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 +14 percentage points is the difference between those two pass rates over the 22 comparable cases. 3 cases got worse with the skill loaded, and they are 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.