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 instructions/julius0217/claude-code-workflow/claude-mdgit clone --depth 1 https://github.com/Julius0217/claude-code-workflowWrote 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/instructions/julius0217/claude-code-workflow/claude-md)<a href="https://agentmods.dev/instructions/julius0217/claude-code-workflow/claude-md"><img src="https://agentmods.dev/badge/instructions/julius0217/claude-code-workflow/claude-md.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 | $0.00986 | $0.00986 |
| Opus 5 | $0.00493 | $0.00493 |
| Sonnet 5 | $0.00197 | $0.00197 |
| Haiku 4.5 | $0.00099 | $0.00099 |
Grade A, and why
claude-code-workflow CLAUDE.md 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 3d 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 — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Claude Code Workflow
Claude Code 사용 원칙과 워크플로우 연구 저장소 (문서 프로젝트)
프로젝트 특성
| 항목 | 값 |
|---|---|
| 유형 | 마크다운 문서 프로젝트 (코드 없음) |
| 언어 | 한국어 (기술용어 영어 병기) |
| 대상 | Claude Code 사용자/개발자 |
구조
docs/ # 핵심 교육 문서
├── getting-started.md # 입문 가이드
├── principles.md # 7가지 핵심 원칙
└── best-practices.md # 고급 활용법 (MCP, 훅, 팀 협업)
workflows/ # 실무 워크플로우 패턴 (4-5단계 프로세스)
├── code-review.md # PR 리뷰
├── debugging.md # 버그 분석
└── refactoring.md # 코드 개선
examples/ # 실제 활용 예제
├── prompts/ # 5가지 프롬프트 카테고리
├── hooks/ # 자동화 훅 설정 (JSON)
└── mcp/ # MCP 서버 설정 (JSON)
templates/ # 복사용 템플릿 (다른 프로젝트용)
핵심 개념
| 개념 | 설명 | 참고 |
|---|---|---|
| 4대 원칙 | 컨텍스트, 점진적 작업, 검증, 프롬프트 | docs/principles.md |
| CLAUDE.md | 프로젝트 컨텍스트 파일 (전역→프로젝트→디렉토리) | templates/CLAUDE.md |
| MCP | 외부 시스템 연동 프로토콜 | examples/mcp/ |
| Hooks | 자동화 트리거 (PreToolUse, PostToolUse 등) | examples/hooks/ |
문서 컨벤션
- 구조: 제목 → 개요 → 예제 → 체크리스트
- 예제 표기: 나쁜 예 → 좋은 예 순서
- 코드 블록: 언어 태그 필수 (
bash,json,typescript등) - 비교: 복잡한 비교는 마크다운 테이블 사용
품질 검증
# 마크다운 린트 (로컬)
npx markdownlint-cli "**/*.md"
GitHub Actions 자동 실행:
markdown-lint.yml- 푸시/PR 시 스타일 검사link-check.yml- 깨진 링크 감지
작업 규칙
새 문서 추가 시
- 기존 문서 스타일 참고 (헤더 구조, 표 형식)
- 관련 문서에 링크 추가 (README.md 목차 등)
- 마크다운 린트 통과 확인
예제 추가 시
- 실제 동작 검증 후 추가
- 민감 정보는 플레이스홀더 사용 (
your-api-key,<TOKEN>) - 위험한 명령어에 경고 문구 추가
자주 요청되는 작업
| 요청 | 참고 파일 | 형식 |
|---|---|---|
| 새 워크플로우 | workflows/*.md | 4-5단계 프로세스 + 체크리스트 |
| 프롬프트 예제 | examples/prompts/*.md | 카테고리별 분류, 구체적 예시 |
| hooks 예제 | examples/hooks/*.json | JSON, 실제 동작 가능해야 함 |
| MCP 설정 | examples/mcp/*.json | JSON, 설명 주석 포함 |
변경 시 연쇄 확인
| 변경 대상 | 확인 필요 |
|---|---|
| docs/ 구조 | README.md 목차, 내부 링크 |
| 새 카테고리 | .github/labeler.yml, 이슈 템플릿 |
| 워크플로우 추가 | README.md 워크플로 테이블 |
문서 학습 경로
README.md (진입점)
↓
docs/getting-started.md (기본)
↓
docs/principles.md (심화)
↓
docs/best-practices.md (고급)
↓
workflows/ + examples/ (실무 적용)
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.
- 3d ago First seen · 114 lines · 986 tokens per session scan A c4f8da00d21d
claude-code-workflow CLAUDE.md is an instructions file published in the GitHub repository Julius0217/claude-code-workflow (3 stars, last pushed 4d ago), licensed MIT. It adds 986 tokens to every session, about $0.0049 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 2026-08-31.
Other instructions, from other repositories
cybara CLAUDE.md
Claude Code instructions for metaspartan/cybara, covering ai coding rules for cybara agent platform, no code comments (strict), runtime environment, import rules and ✅ do.
cybara AGENTS.md
AGENTS.md instructions for metaspartan/cybara, covering cybara agent rules, package manager policy (strict), command translation, typescript type safety and no comments.
awesome-ai-native CLAUDE.md
Instructions for samber/awesome-ai-native, covering claude.md — awesome-ai-native, single source of truth, scope, lint and site (site/).
awesome-copilot-mcp copilot-instructions.md
Instructions for RbBtSn0w/awesome-copilot-mcp, covering copilot / ai agent instructions — awesome-copilot-mcp, big picture, mcp compliance (required), key components & where to look and developer workflows (concrete commands).
ai-auto-work CLAUDE.md
Claude Code instructions for chaohong-ai/ai-auto-work, covering gamemaker - ai native 游戏创作平台, 项目简介, ai 工作目录约定 and 按需加载（不默认装载）.
agent-bus CLAUDE.md
Claude Code instructions for scottbaggett/agent-bus, covering agent-bus, on wake, working here, coordinating with other seats and layout.