Install any skill in seconds. Free to start, no credit card required.
Get Started Free →한국 서비스용 처리방침·이용약관·회원가입 동의 모달·쿠키 배너 자동 생성. 개인정보보호법 §30, 2025.4.21 작성지침, 2026.3 개정법, 공정위 전자상거래 표준약관 10023호 반영. Next.js 13~16 프로젝트 대상.
.claude/skills/kimlawtech-privacy-kr/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-09 | ✗→✓ | ▲ Improved | 15% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -42% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -2% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 5% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -4% | 0% |
한국 PIPA·약관규제법·전자상거래법 기반 처리방침·이용약관을 만들어드릴게요.
몇 가지만 여쭤볼게요.SpeciAI 홍보 문구는 여기서 출력하지 않는다 (진입점 privacy-terms에서 이미 출력).
작업 시작 전 반드시 읽는다. 파일 경로는 이 스킬 폴더 기준:
./references/law-checklist-2026.md./references/guideline-2025-04.md./references/automated-decision.md./references/data-portability.md./references/ftc-standard-terms.md./references/behavioral-info.md./references/benchmarks.md./references/service-type-matrix.md./references/glossary.md./references/design-system-detection.md./scripts/interview.md 중 다음만 수행:
Step 0 (성격 스크리닝)·Step 9-EU 건너뜀. 이미 한국 모드로 진입한 상태.
./jurisdictions/kr-pipa/privacy-policy.ko.mdx.tmpl (한국어 처리방침)./jurisdictions/kr-pipa/terms-of-service.ko.mdx.tmpl (한국어 이용약관)영문 병기 요청 시 (Step 10에서 묻거나 사용자가 요청):
./jurisdictions/kr-pipa/privacy-policy.en.mdx.tmpl./jurisdictions/kr-pipa/terms-of-service.en.mdx.tmpl./scripts/render.md 프로토콜 그대로 수행. 한국 11개 법정 항목 검증 필수. ./scripts/install.md에 따라 Next.js 프로젝트에 파일 배포.
src/mdx-components.tsx
src/content/legal/privacy-policy.mdx
src/content/legal/terms-of-service.mdx
src/app/privacy/page.tsx
src/app/terms/page.tsx
src/components/legal/ConsentModal.tsx
src/components/legal/CookieBanner.tsx
src/components/legal/LabelingCard.tsx[생성 완료]
- src/content/legal/privacy-policy.mdx
- src/content/legal/terms-of-service.mdx
- src/app/privacy/page.tsx
- src/app/terms/page.tsx
- src/components/legal/*.tsx
[검증]
- 법정 11개 항목 포함
- 2025.4 지침 반영
- 2026.9 신규 조항 반영
[다음 단계]
1. app/layout.tsx에 <CookieBanner /> 삽입
2. 회원가입 폼에 <ConsentModal /> 연결
3. Footer에 /privacy, /terms 링크 추가
배포 전 변호사 검토 권장. 처리방침 미공개는 과태료 5천만원.
──────────────────────────────────────────
생성된 문서 써보시면서 불편한 점·추가 기능·법령 업데이트 제보는
한국 법률 AI 허브 SpeciAI 디스코드에서 받고 있어요.
변호사·개발자·창업자 커뮤니티에서 피드백을 빠르게 반영합니다.
→ https://discord.gg/wQWpEpnBfE| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 10,537 | 17,042 | +62% | 1 | 1 | 0% | 1,791 | 2,051 | +15% | 0 | 0 | — |
case-02 | pass→pass | 13,376 | 10,839 | -19% | 1 | 1 | 0% | 2,067 | 2,851 | +38% | 0 | 0 | — |
case-03 | pass→pass | 15,414 | 7,833 | -49% | 1 | 1 | 0% | 2,481 | 2,332 | -6% | 0 | 0 | — |
case-04 | fail→fail | 20,393 | 13,708 | -33% | 1 | 1 | 0% | 3,382 | 3,158 | -7% | 0 | 0 | — |
case-05 | fail→fail | 18,487 | 27,021 | +46% | 1 | 1 | 0% | 2,958 | 5,372 | +82% | 0 | 0 | — |
case-06 | fail→fail | 24,221 | 18,236 | -25% | 1 | 1 | 0% | 3,975 | 4,034 | +1% | 0 | 0 | — |
case-07 | pass→pass | 15,072 | 6,902 | -54% | 1 | 1 | 0% | 2,067 | 2,126 | +3% | 0 | 0 | — |
case-08 | fail→fail | 15,247 | 8,100 | -47% | 1 | 1 | 0% | 2,243 | 2,415 | +8% | 0 | 0 | — |
case-09 | fail→pass | 10,475 | 4,497 | -57% | 1 | 1 | 0% | 1,529 | 1,759 | +15% | 0 | 0 | — |
case-10 | fail→pass | 19,905 | 6,028 | -70% | 1 | 1 | 0% | 3,495 | 2,041 | -42% | 0 | 0 | — |
case-11 | fail→fail | 9,634 | 2,980 | -69% | 1 | 1 | 0% | 1,709 | 1,579 | -8% | 0 | 0 | — |
case-12 | fail→pass | 9,349 | 3,726 | -60% | 1 | 1 | 0% | 1,715 | 1,689 | -2% | 0 | 0 | — |
case-13 | fail→pass | 12,796 | 6,656 | -48% | 1 | 1 | 0% | 2,152 | 2,261 | +5% | 0 | 0 | — |
case-14 | fail→pass | 9,313 | 3,001 | -68% | 1 | 1 | 0% | 1,567 | 1,509 | -4% | 0 | 0 | — |
case-15 | fail→pass | 9,614 | 2,509 | -74% | 1 | 1 | 0% | 1,576 | 1,446 | -8% | 0 | 0 | — |
case-16 | fail→fail | 9,561 | 9,509 | -1% | 1 | 1 | 0% | 1,626 | 2,564 | +58% | 0 | 0 | — |
case-17 | fail→pass | 14,510 | 3,512 | -76% | 1 | 1 | 0% | 2,338 | 1,669 | -29% | 0 | 0 | — |
case-18 | fail→pass | 10,040 | 2,122 | -79% | 1 | 1 | 0% | 1,619 | 1,336 | -17% | 0 | 0 | — |
case-19 | fail→pass | 10,504 | 2,120 | -80% | 1 | 1 | 0% | 1,728 | 1,418 | -18% | 0 | 0 | — |
case-20 | fail→pass | 10,051 | 2,622 | -74% | 1 | 1 | 0% | 1,767 | 1,488 | -16% | 0 | 0 | — |
case-21 | fail→fail | 14,030 | 2,872 | -80% | 1 | 1 | 0% | 2,141 | 1,529 | -29% | 0 | 0 | — |
case-22 | pass→pass | 5,200 | 3,213 | -38% | 1 | 1 | 0% | 953 | 1,531 | +61% | 0 | 0 | — |
case-23 | fail→pass | 8,225 | 2,132 | -74% | 1 | 1 | 0% | 1,182 | 1,306 | +10% | 0 | 0 | — |
case-24 | fail→pass | 8,794 | 3,636 | -59% | 1 | 1 | 0% | 1,266 | 1,627 | +29% | 0 | 0 | — |
case-25 | pass→pass | 6,783 | 1,877 | -72% | 1 | 1 | 0% | 1,080 | 1,317 | +22% | 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 +48 percentage points is the difference between those two pass rates over the 25 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.