Org OS의 실무 사례는 학부 제품으로 불리는 소설 회사입니다. Orgs 작업장을위한 Claude 코드에 대한 비교.

이것은 무엇인가?

실제 제품 팀의 공유 워크플로 레이어가 어떻게 보일까요? 폴더를 검색하여 네 개의 버킷을 볼 수 있습니다. —Context, Actions, Behavior, Connections — 연습에서 보여줍니다.

모든 하위 폴더에는 자체가 있습니다 CLAUDE.md - 그 폴더의 탐색지도. Root는 Claude를 알려줍니다. 배열 된 파일은 Claude에게 내부입니다.

그것을 사용하는 방법

첫째로 이동합니다. Org OS가 유선 된 방법을 볼 수 있도록이를 엽니 다 :

1. CLAUDE.md — 야윈 루트 파일 (팀 roster + Slack 채널과 함께 한 페이지 아래)
2. customers/CLAUDE.md — 네비게이션 맵으로 배열된 CLAUDE.md의 예
3. customers/account-summaries/acme-q1-2026.md - 캡처 된 고객 지식은 다음과 같습니다.
4. .claude/skills/exec-preso-prep/SKILL.md - 기술 파일이 같은 것
5. .claude/skills/share-check/SKILL.md - 5개의 공유성 규칙을 실행하는 문 기술
6. .claude/skills/metrics-check/SKILL.md - 메트릭 테이블에서 X를 측정하는 방법
7. `.claude/skills/org-os-updates/SKILL.

Faculty Products — Org OS

A worked example of a Org OS for a fictional company called Faculty Products. Companion to the Claude Code for Orgs workshop.

What this is

A populated example of how a real product team’s shared workflow layer might look. Browse the folders to see how the four buckets — Context, Actions, Behavior, Connections — show up in practice.

.
├── CLAUDE.md                  # the lean root — the index, not the content
├── .mcp.json                  # CONNECTIONS: project-level MCP config
├── .env.example               # env vars the MCP config expects
├── customers/                 # CONTEXT: customer calls, summaries, playbooks
├── analytics/                 # CONTEXT: metric definitions, queries, playbooks
│   └── metrics.md             # the canonical metrics table (read by /metrics-check)
├── templates/                 # CONTEXT: PRD, status update, RFC structures
└── .claude/
    ├── CLAUDE.md              # what's in this folder
    ├── connections.md         # CONNECTIONS: human-readable list of plug-ins
    ├── skills/                # ACTIONS: shared workflows (slash commands)
    └── rules/                 # BEHAVIOR: file naming, prioritization, etc.

Every subfolder also has its own CLAUDE.md — the navigation map for that folder. Root tells Claude where to look; nested files tell Claude what’s inside.

How to use it

Browse first. Open these in order to see how a Org OS is wired:

1. CLAUDE.md — the lean root file (under one page, with team roster + Slack channels)
2. customers/CLAUDE.md — example of a nested CLAUDE.md as a navigation map
3. customers/account-summaries/acme-q1-2026.md — what captured customer knowledge looks like
4. .claude/skills/exec-preso-prep/SKILL.md — what a skill file looks like
5. .claude/skills/share-check/SKILL.md — the gate skill that runs the 5 shareability rules
6. .claude/skills/metrics-check/SKILL.md — answers “how do we measure X?” from the metrics table
7. .claude/skills/org-os-updates/SKILL.md — weekly visibility post for #org-os
8. .claude/connections.md — human-readable list of what this OS plugs into
9. analytics/metrics.md — the canonical metrics table
10. .claude/rules/prioritization.md — what a team rule looks like

Then clone it and adapt. Replace the FP-specific content with yours. The structure is the load-bearing thing.

git clone https://github.com/carlvellotti/faculty-products-org-os.git my-org-os
cd my-org-os
# Edit CLAUDE.md with your team's roster + channels
# Replace the customer summaries with your accounts
# Replace the metric definitions with your metrics
# Keep the skills + rules structure, add your own

Where the workshop fits

This repo is the example. The /orgos page walks through it — what each piece is, why it’s structured this way, and how the four buckets compose into a flow when you actually use it.