Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →[DEPRECATED v2.3.0] 본 스킬은 moai-marketing:sns-content로 흡수되었습니다. 신규 호출은 /sns-content를 사용하세요. 글로벌 4채널(스레드·X·링크드인·유튜브쇼츠)과 한국 3채널(인스타·네이버 블로그·카카오)을 모두 sns-content에서 지원합니다. 본 스텁은 v2.5.0까지 외부 참조 호환성 보존용으로 유지됩니다.
.claude/skills/modu-ai-social-media/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✗→✓ | ▲ Improved | -32% | 0% |
| case-01 | ✗→✓ | ▲ Improved | -17% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 11% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 17% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -26% | 0% |
본 스킬은 v2.3.0부터 moai-marketing:sns-content로 흡수되었습니다.
| 기존 호출 | 신규 호출 | |----------|----------| | /social-media | /sns-content |
sns-content는 다음 두 가지 모드를 모두 지원합니다.
moai-content:social-media와 moai-marketing:sns-content가 SNS 콘텐츠 영역에서 책임 경계가 모호하여 셀러·마케터가 어디로 호출할지 혼동되는 문제를 해결하기 위해 단일 스킬로 통합했습니다 (SPEC-CATALOG-CLEANUP-007 REQ-CLEANUP-003).
본 스텁은 cowork-plugins 정책상 최소 2 minor 버전 (v2.5.0)까지 유지됩니다. 외부에서 /social-media를 직접 참조하는 코드가 있다면 v2.5.0 이전에 /sns-content로 마이그레이션해주세요.
Version: 2.3.0 Classification: DEPRECATED → moai-marketing:sns-content Relocated: v2.3.0 (SPEC-CATALOG-CLEANUP-007)
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-08 | fail→pass | 11,004 | 4,917 | -55% | 1 | 1 | 0% | 1,713 | 1,172 | -32% | 0 | 0 | — |
case-01 | fail→pass | 18,131 | 13,471 | -26% | 1 | 1 | 0% | 2,885 | 2,394 | -17% | 0 | 0 | — |
case-02 | fail→pass | 19,090 | 18,546 | -3% | 1 | 1 | 0% | 2,978 | 3,301 | +11% | 0 | 0 | — |
case-03 | fail→pass | 17,381 | 16,575 | -5% | 1 | 1 | 0% | 2,687 | 3,131 | +17% | 0 | 0 | — |
case-04 | fail→pass | 7,856 | 2,922 | -63% | 1 | 1 | 0% | 1,238 | 921 | -26% | 0 | 0 | — |
case-05 | fail→pass | 20,847 | 13,623 | -35% | 1 | 1 | 0% | 3,208 | 2,736 | -15% | 0 | 0 | — |
case-06 | fail→pass | 10,679 | 3,553 | -67% | 1 | 1 | 0% | 1,725 | 1,021 | -41% | 0 | 0 | — |
case-07 | fail→pass | 16,187 | 12,155 | -25% | 1 | 1 | 0% | 2,764 | 2,660 | -4% | 0 | 0 | — |
case-09 | fail→pass | 10,485 | 2,566 | -76% | 1 | 1 | 0% | 1,837 | 872 | -53% | 0 | 0 | — |
case-10 | pass→pass | 15,251 | 11,093 | -27% | 1 | 1 | 0% | 2,182 | 2,084 | -4% | 0 | 0 | — |
case-11 | pass→pass | 15,910 | 5,327 | -67% | 1 | 1 | 0% | 2,818 | 1,289 | -54% | 0 | 0 | — |
case-12 | fail→pass | 16,657 | 10,414 | -37% | 1 | 1 | 0% | 2,433 | 2,334 | -4% | 0 | 0 | — |
case-13 | fail→pass | 4,523 | 2,082 | -54% | 1 | 1 | 0% | 675 | 719 | +7% | 0 | 0 | — |
case-14 | fail→pass | 17,316 | 13,971 | -19% | 1 | 1 | 0% | 2,783 | 2,806 | +1% | 0 | 0 | — |
case-15 | fail→pass | 8,442 | 2,672 | -68% | 1 | 1 | 0% | 1,225 | 825 | -33% | 0 | 0 | — |
case-16 | pass→pass | 12,203 | 5,254 | -57% | 1 | 1 | 0% | 1,846 | 1,269 | -31% | 0 | 0 | — |
case-17 | pass→pass | 9,710 | 1,791 | -82% | 1 | 1 | 0% | 1,504 | 690 | -54% | 0 | 0 | — |
case-18 | pass→pass | 17,972 | 9,611 | -47% | 1 | 1 | 0% | 2,966 | 1,894 | -36% | 0 | 0 | — |
case-19 | pass→pass | 21,186 | 17,903 | -15% | 1 | 1 | 0% | 3,401 | 3,341 | -2% | 0 | 0 | — |
case-20 | pass→pass | 15,828 | 15,482 | -2% | 1 | 1 | 0% | 2,909 | 3,690 | +27% | 0 | 0 | — |
case-21 | pass→pass | 16,112 | 13,561 | -16% | 1 | 1 | 0% | 2,626 | 2,696 | +3% | 0 | 0 | — |
case-22 | fail→pass | 12,581 | 8,385 | -33% | 1 | 1 | 0% | 2,235 | 2,024 | -9% | 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 +64 percentage points is the difference between those two pass rates over the 22 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.