Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Discover, analyze, and prioritize keywords for SEO and GEO content strategies from a seed keyword or niche. Identifies high-value opportunities based on search volume, competition, intent, and business relevance. Generates topic clusters. For building a dated editorial calendar from your own Search Console data, use `content-planner` instead — this skill is seed-driven keyword discovery, not calendar scheduling. Use when asked to "find keywords", "keyword research", "keyword analysis", "search v
.claude/skills/nowork-studio-keyword-research/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | -9% | 0% |
| case-05 | ✓→✓ | ＝ Same ✓ | 31% | 0% |
| case-10 | ✓→✓ | ＝ Same ✓ | 9% | 0% |
| case-13 | ✓→✓ | ＝ Same ✓ | 7% | 0% |
| case-16 | ✓→✓ | ＝ Same ✓ | 5% | 0% |
Read ../../seo/keyword-research/SKILL.md completely, then follow it as the active workflow. Resolve every relative reference from that file against ../../seo/keyword-research/.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 20,638 | 21,254 | +3% | 1 | 1 | 0% | 3,708 | 3,940 | +6% | 0 | 0 | — |
case-02 | fail→fail | 16,558 | 25,558 | +54% | 1 | 1 | 0% | 2,960 | 4,427 | +50% | 0 | 0 | — |
case-03 | fail→fail | 26,885 | 34,140 | +27% | 1 | 1 | 0% | 4,297 | 5,907 | +37% | 0 | 0 | — |
case-04 | fail→fail | 14,431 | 18,658 | +29% | 1 | 1 | 0% | 2,405 | 2,291 | -5% | 0 | 0 | — |
case-05 | pass→pass | 14,051 | 18,618 | +33% | 1 | 1 | 0% | 2,272 | 2,977 | +31% | 0 | 0 | — |
case-06 | fail→pass | 24,954 | 25,512 | +2% | 1 | 1 | 0% | 2,369 | 2,163 | -9% | 0 | 0 | — |
case-07 | fail→fail | 14,179 | 18,316 | +29% | 1 | 1 | 0% | 2,250 | 3,084 | +37% | 0 | 0 | — |
case-08 | fail→fail | 13,141 | 18,722 | +42% | 1 | 1 | 0% | 2,130 | 3,090 | +45% | 0 | 0 | — |
case-09 | fail→fail | 18,734 | 22,095 | +18% | 1 | 1 | 0% | 2,838 | 3,536 | +25% | 0 | 0 | — |
case-10 | pass→pass | 12,676 | 14,022 | +11% | 1 | 1 | 0% | 2,183 | 2,383 | +9% | 0 | 0 | — |
case-11 | fail→fail | 19,401 | 17,457 | -10% | 1 | 1 | 0% | 2,854 | 2,840 | -0% | 0 | 0 | — |
case-12 | fail→fail | 13,337 | 13,959 | +5% | 1 | 1 | 0% | 1,877 | 1,960 | +4% | 0 | 0 | — |
case-13 | pass→pass | 16,576 | 16,334 | -1% | 1 | 1 | 0% | 2,391 | 2,566 | +7% | 0 | 0 | — |
case-14 | fail→fail | 16,351 | 13,768 | -16% | 1 | 1 | 0% | 2,398 | 1,949 | -19% | 0 | 0 | — |
case-15 | fail→fail | 12,719 | 29,004 | +128% | 1 | 1 | 0% | 2,105 | 5,810 | +176% | 0 | 0 | — |
case-16 | pass→pass | 14,004 | 15,990 | +14% | 1 | 1 | 0% | 2,102 | 2,217 | +5% | 0 | 0 | — |
case-17 | fail→fail | 18,614 | 17,434 | -6% | 1 | 1 | 0% | 2,667 | 2,560 | -4% | 0 | 0 | — |
case-18 | fail→fail | 19,361 | 13,371 | -31% | 1 | 1 | 0% | 2,863 | 2,180 | -24% | 0 | 0 | — |
case-19 | pass→pass | 11,232 | 11,851 | +6% | 1 | 1 | 0% | 1,688 | 1,910 | +13% | 0 | 0 | — |
case-20 | pass→pass | 18,521 | 21,613 | +17% | 1 | 1 | 0% | 3,039 | 3,396 | +12% | 0 | 0 | — |
case-21 | pass→pass | 9,832 | 9,185 | -7% | 1 | 1 | 0% | 1,866 | 1,853 | -1% | 0 | 0 | — |
case-22 | pass→pass | 11,328 | 7,936 | -30% | 1 | 1 | 0% | 1,689 | 1,251 | -26% | 0 | 0 | — |
case-23 | pass→pass | 18,268 | 43,701 | +139% | 1 | 1 | 0% | 2,880 | 3,599 | +25% | 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 +4 percentage points is the difference between those two pass rates over the 23 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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 8/12/2026 | +28% |
Other measured skills in the registry, with their headline benchmark lift.