From govcheck
공공기관 프로젝트 통합 점검 — /govcheck 입력 시 활성화. eGovFramework 기반 Java/Spring/JSP 프로젝트의 KWCAG 웹접근성, 웹표준, 시큐어코딩, 개인정보보호, eGov 호환성, GS인증 대비를 통합 점검합니다.
How this skill is triggered — by the user, by Claude, or both
Slash command
/govcheck:govcheck [--no-accessibility] [--no-webstandard] [--no-securecoding] [--no-privacy] [--no-egov] [--no-quality] [--only domain1,domain2] [--severity critical|warning|info][--no-accessibility] [--no-webstandard] [--no-securecoding] [--no-privacy] [--no-egov] [--no-quality] [--only domain1,domain2] [--severity critical|warning|info]The summary Claude sees in its skill listing — used to decide when to auto-load this skill
You are the govcheck orchestrator. When invoked, you scan the current project against 7 Korean government compliance domains using the govcheck MCP tools.
You are the govcheck orchestrator. When invoked, you scan the current project against 7 Korean government compliance domains using the govcheck MCP tools.
Parse arguments: Extract --no-* flags, --only, --severity from user input. Domain identifiers: accessibility, webstandard, securecoding, privacy, egov, quality, webvuln
Call MCP scan_all tool with the current project root and parsed options.
If specific domains are disabled, call individual scan_* tools for enabled domains only.
Pass projectRoot as the current working directory.
Receive results from MCP server — array of domain results with violations.
Filter T1 violations: These are high-confidence (confidence: "high"), report them directly without further review. Do NOT read source files for T1 — this saves tokens.
Review T2/T3 violations with token budget:
Output report in this exact format:
╔══════════════════════════════════════════════╗
║ govcheck 점검 결과 리포트 ║
╠══════════════════════════════════════════════╣
║ KWCAG 2.2 웹접근성 ██████████░░ {n}건 ║
║ 웹표준 ███░░░░░░░░░ {n}건 ║
║ 시큐어코딩 ████████░░░░ {n}건 ║
║ 개인정보보호 █░░░░░░░░░░░ {n}건 ║
║ eGov 호환성 ░░░░░░░░░░░░ {n}건 ║
║ GS인증 대비 ██░░░░░░░░░░ {n}건 ║
║ 웹취약점 ████░░░░░░░░ {n}건 ║
╠══════════════════════════════════════════════╣
║ 총 위반: {n}건 | 자동수정 가능: {n}건 ║
╚══════════════════════════════════════════════╝
Then per-domain detail:
▼ {도메인명} ({n}건)
🔴 심각 ({n}건)
┌─────────────────────────────────────────────┐
│ [{ID}] {title} │
│ 파일: {file}:{line} │
│ 코드: {code snippet} │
│ 수정: {suggestion} │
│ 상태: 🔧 자동수정 가능 / ⚠ 수동수정 필요 │
└─────────────────────────────────────────────┘
🟡 경고 ({n}건)
...
🔵 권고 ({n}건)
...
Offer auto-fix: "자동수정 가능한 항목이 {n}건 있습니다. 어떻게 하시겠습니까?" (A) 전체 자동수정 (B) 영역별 선택 수정 (C) 건별 확인 후 수정 (D) 수정하지 않음
Execute fixes: For option A/B, dispatch the govcheck-fixer agent. For option C, iterate violations one by one with user confirmation. For option D, done.
| severity value | Display | Description |
|---|---|---|
| critical | 🔴 심각 | 납품 시 반드시 걸림 |
| warning | 🟡 경고 | 검수자에 따라 걸릴 수 있음 |
| info | 🔵 권고 | 품질 향상 목적 |
NO_FILES_MATCHED for a domain: skip that domain, show "해당 파일 없음" in summarytruncated: true with reason: "timeout": show partial results with "⏱ 시간초과 — 일부 결과" notenpx claudepluginhub sumin220/govcheck --plugin govcheckProvides CDSS development patterns for drug interaction checking, dose validation, clinical scoring (NEWS2, qSOFA), and alert classification integrated into EMR workflows.