Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add rules/gaebalai/cursor-def-a-rule/rule_selectorgit clone --depth 1 https://github.com/gaebalai/Cursor-DEF-A-RuleWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/rules/gaebalai/cursor-def-a-rule/rule_selector)<a href="https://agentmods.dev/rules/gaebalai/cursor-def-a-rule/rule_selector"><img src="https://agentmods.dev/badge/rules/gaebalai/cursor-def-a-rule/rule_selector.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.03282 | $0.03282 |
| Opus 5 | $0.01641 | $0.01641 |
| Sonnet 5 | $0.00656 | $0.00656 |
| Haiku 4.5 | $0.00328 | $0.00328 |
Grade A, and why
rule_selector scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 211 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Copyright (c) 2025 Jaewoo Kim
MIT License - https://opensource.org/licenses/MIT
[SELECTOR] 질문 내용에 따른 규칙 선택 시스템 (DEF-A 통합 최적화 버전)
DEF-A 질문 분석 프레임워크 (최적화 버전)
DEF-A 단계 판정 키워드 (부분 적용 대응)
🎯 Define 단계: "요구사항", "문제", "과제", "이해", "무엇을", "왜", "전략", "설계"
🔍 Explore 단계: "비교", "선택", "어느 쪽", "검토", "학습", "이해", "단계별"
✨ Formulate 단계: "설계", "구조", "통합", "최적화", "구현", "개발", "타입 안정성"
📝 Act/Apply 단계: "구현", "코딩", "작성", "실행", "긴급", "수정", "버그"
📈 Assess/Adjust 단계: "평가", "측정", "분석", "개선", "프로세스 최적화"
인지 스타일 판정 키워드 (최적화 버전)
Systems Mode: "시스템", "아키텍처", "설계", "전략", "복잡", "타입 안정성", "효율적"
Empathy Mode: "사용자", "경험", "학습", "교육", "팀", "단계별", "실용적"
기술 영역 판정 키워드 (최적화 버전)
프론트엔드: React, Vue, Angular, TypeScript, JavaScript, 컴포넌트, UI, UX
백엔드: API, 서버, 데이터베이스, Node.js, Python, 인증, 보안
테스트: 테스트, TDD, BDD, Jest, Vitest, 품질 보증, 리뷰, CI/CD
지식 관리: 암묵지, 형식지, SECI 모델, 지식 공유, 학습, 베스트 프랙티스, 규칙화
윤리적 고려: 윤리, 편향, 포용성, 공정성, 인간성, 다양성, 공감, 상호 학습
전반・설계: 설계, 아키텍처, 패턴, 최적화, 보안, 풀스택
질문 유형 판정 (최적화 버전)
구현・코딩: "구현", "코드", "작성", "개발", "에러", "버그", "수정"
설계・아키텍처: "설계", "아키텍처", "구조", "패턴", "선택", "비교"
테스트・품질: "테스트", "품질", "검증", "TDD", "BDD", "리뷰", "개선"
조사・학습: "조사", "리서치", "학습", "비교", "단계별", "이해", "실습"
긴급 대응: "긴급", "운영", "크래시", "에러", "수정", "최소화", "신속"
에러 처리: "에러", "예외", "장애", "디버깅", "로그", "모니터링"
팀 협업: "팀", "협업", "리뷰", "페어", "멘토링", "공유", "학습"
지식 관리: "암묵지", "형식지", "SECI", "지식 공유", "학습", "베스트 프랙티스", "규칙화"
윤리적 고려: "윤리", "편향", "포용성", "공정성", "인간성", "다양성", "공감", "상호 학습"
프로그래밍 윤리: "함수명", "변수명", "주석", "에러 메시지", "코드 리뷰", "네이밍 규칙", "문서화"
DEF-A 통합 규칙 선택 로직 (최적화 버전)
우선순위 기반 규칙 적용
1. DEF-A 단계 판정 → 사고 흐름 단계 결정 (완전/부분/최소 적용)
2. 인지 스타일 판정 → 응답 스타일 결정 (Systems/Empathy)
3. 기술 영역 판정 → 전문 규칙 파일 선택
4. 질문 유형 판정 → 적용 수준 결정
5. 긴급도 판정 → 복잡성 조정
6. 프로젝트 특성 → 맞춤형 적용
7. 부분 적용 전략 → 효율성 최적화
DEF-A 단계별 규칙 적용 매트릭스 (최적화 버전)
DEF-A 단계 → 적용 규칙・접근법・복잡성
├── 🎯 Define → 요구사항 분석・문제 정의・맥락 이해 (완전 적용 시에만)
├── 🔍 Explore → 다각도 분석・기술 선택・접근법 검토 (학습 지원 시 강조)
├── ✨ Formulate → 설계 통합・최적화・구현 계획 (구현 중시 시)
├── 📝 Act/Apply → 구현・코딩・통합 (긴급 대응 시 최소화)
└── 📈 Assess/Adjust → 평가・개선・최적화 (프로세스 개선 시 강조)
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago First seen · 211 lines · 3,282 tokens per session scan A 9efb7112f981
rule_selector is a cursor rule published in the GitHub repository gaebalai/Cursor-DEF-A-Rule (12 stars, last pushed 1y ago), licensed MIT. It adds 3,282 tokens to every session, about $0.0164 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other cursor rules, from other repositories
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.
cli-error-handling
CLI command error handling patterns.
control-plane-descriptors
Control plane descriptor and instance implementation patterns.
family-instance-domain-actions
Family instance domain action implementation patterns.