사용 방법
- 설치 명령을 복사하고 자신의 에이전트 환경에 설치하기 전에 소스 패키지를 확인하세요.
- 필요한 도구, 인증 정보, 지원 환경은 원본 문서에서 확인하세요.
- 입력과 원하는 산출물을 설명하고 출처와 결과를 대조하세요.
npx skills add https://github.com/majesticlabs-dev/majestic-marketplace --skill data-integrity-reviewer시작 전 확인
- 설치에는 모델 구독, API 한도, 파일 접근 권한이 포함되지 않습니다.
- 별점 수와 설치 수는 활동 지표이며 독립적인 품질·보안 검증이 아닙니다.
원본 Skill 문서
원본 저장소의 제목, 예시, 코드, 표, 링크, 이미지를 유지해 표시합니다.
Data Integrity Reviewer
Audience: Reviewers auditing Rails changes for data correctness risks. Goal: Route to specialized skills based on changed files and aggregate findings.
Routing Table
| Concern | Apply | Triggers |
|---|---|---|
| Migrations | rails-refactorer (Migration Safety section) | db/migrate/*, schema changes, column ops |
| Constraints | constraints-reviewer | Validations, FKs, unique indexes, orphan risks |
| Transactions | business-logic-coder (Transaction Boundaries section) | Multi-step ops, locks, isolation levels |
| Privacy | privacy-reviewer | PII fields, encryption, GDPR/CCPA |
Workflow
CHANGED_FILES = git diff --name-only
SKILLS = []
If CHANGED_FILES matches db/migrate/*:
SKILLS.append(rails-refactorer) # Migration Safety section
If CHANGED_FILES matches app/models/* OR validations changed:
SKILLS.append(constraints-reviewer)
If CHANGED_FILES matches service objects OR transaction blocks:
SKILLS.append(business-logic-coder) # Transaction Boundaries section
If CHANGED_FILES matches PII models OR encryption config:
SKILLS.append(privacy-reviewer)
For each S in SKILLS:
Read skills/<S>/SKILL.md and apply criteria to CHANGED_FILES → FINDINGS[S]
AGGREGATE FINDINGS into unified reportFull Audit Mode
When user requests comprehensive data integrity audit, apply all 4 skills regardless of file diff.
Input Schema
yaml
scope:
changed_files: [string] # optional; default = git diff --name-only
mode: routed | full-audit # default routedOutput Schema
yaml
overall_status: PASS | WARN | FAIL
areas:
migrations: { status: PASS | WARN | FAIL, critical_count: int }
constraints: { status: PASS | WARN | FAIL, critical_count: int }
transactions: { status: PASS | WARN | FAIL, critical_count: int }
privacy: { status: PASS | WARN | FAIL, critical_count: int }
critical_risks:
- area: migrations | constraints | transactions | privacy
file: string
line: int
description: string
recommendations:
- severity: P1 | P2 | P3
action: stringError Handling
| Condition | Action |
|---|---|
| No matching skills triggered | Return overall_status: PASS, empty findings, note "no data-integrity changes detected" |
| Skill file missing | Skip that area, mark status WARN, note in output |
| Git diff empty | Return overall_status: PASS, "no changes to review" |
