darnoux

Claude Code For Teams — Development skill for Claude Code

Development community

The starter kit for running Claude Code across a team of 2 to 30 users.

How to install Claude Code For Teams

This entry records only its repository, not the path inside it, so there is no exact command to give. Open darnoux/claude-code-for-teams and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Claude Code For Teams does

The starter kit for running Claude Code across a team of 2 to 30 users. Companion to the playbook at docs.heyarnoux.com/teams.

Alternatives in Development

Skill Creator — Create new skills, modify existing skills, and measure skill performance 94.1k ★
Claude SEO — Multi-Platform Agent Instructions — For Cursor, Cursor Cloud Agents, Google Antigravity, and Gemini CLI 5k ★
Oh My Agent — Portable multi-agent harness for .agents-based skills, workflows, and standards-aware agent teams across Antig 786 ★

README

Claude Code for Teams

The starter kit for running Claude Code across a team of 2 to 30 users. Devs, marketers, sales, ops.

[**Read the full playbook (~30 pages) →**](https://docs.heyarnoux.com/teams)

This repo is the executable companion. Clone it, copy `starter/` into your project, customize, ship.

Pick your path

Most teams of 2 to 30 run one of three setups. Pick yours first.

Path	For	Start here
A. Cowork-only	Marketing, sales, ops, finance, legal teams. No terminal users.	`/cowork`
B. Mixed	Devs on Claude Code + everyone else on Cowork. Most teams of 2-30 in 2026.	`/mixed`
C. Claude Code only	Engineering-led teams. Repo-first.	`/starter` (this is the default)

The big asymmetry: **Cowork is excluded from Anthropic's Audit Logs, Compliance API, Data Exports, and BAA** as of May 2026. If you're in a regulated industry, Cowork is disqualified for sensitive workloads. Put that work in Claude Code (which IS covered).

What's in here

.
├── starter/                     ← Path C, copy this into your team repo (Claude Code)
│   ├── .claude/
│   │   ├── settings.json        ← deny rules, env vars, hook wiring
│   │   ├── hooks/               ← 4 baseline hooks (block-destructive, require-approval, scan-secrets, audit-log)
│   │   ├── skills/research/     ← sample team-shared skill
│   │   ├── agents/              ← sample subagent (code-reviewer)
│   │   └── rules/               ← path-scoped rule examples
│   ├── .github/
│   │   ├── CODEOWNERS           ← route AI-edited paths to a human
│   │   ├── workflows/ai-safety.yml  ← secrets + PII guard on every PR
│   │   └── pull_request_template.md
│   ├── .mcp.json.example        ← read-only MCP scaffold (pinned versions)
│   └── CLAUDE.md.example        ← team rules template (< 200 lines)
├── cowork/                      ← Path A, Cowork-only governance kit
│   ├── README.md
│   ├── a