taipmtaipm/setup-c2/CLAUDE.md

setup-c2

CLAUDE.md1 starsChanged 7 months agoMITView on GitHub

setup-c2 is a CLI tool that scaffolds Claude Code configuration for any project. Two scenarios: [1] new projects (from scratch), [2] existing projects (non-destructive, never overwrites). Bundled plugin with 4 skills: Install: claude plugin install /path/to/setup-c2/plugins/microai Autonomous development team — 5 agents, 3 skills: Command: /dev-team [task | #issue | --from-roadmap] Install: claude plugin install /path/to/setup-c2/plugins/dev-team Codebase analysis team — 3 agents, 3 skills: Command: `/pr

In this repository4 other files it gives its agents.
- templates/CLAUDE.mdCLAUDE.md
- deep-questionsSkill
- project-discoverySkill
- understand-questionsSkill

The file

Create a free account to mount

Files

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Language

Document: Vietnamese
Code: English

Project Overview

setup-c2 is a CLI tool that scaffolds Claude Code configuration for any project.

npx setup-c2 init              # Interactive mode
npx setup-c2 init --yes        # Non-interactive with defaults
npx setup-c2 init --targets skills,hooks   # Selective targets

Two scenarios: [1] new projects (from scratch), [2] existing projects (non-destructive, never overwrites).

Architecture

bin/setup-c2.js          # CLI entry point (Commander.js)
src/
├── init.js              # Main flow: wizard → confirm → generate
├── wizard.js            # Interactive prompts (@inquirer/prompts)
├── generator.js         # Template engine: load template → replace vars → write if not exists
├── detector.js          # Language detection from marker files (10 rules)
└── plugins.js           # Plugin discovery + installation
templates/               # 14 templates (6 CLAUDE.md + 4 skills + 4 scaffolds)
plugins/                 # 3 bundled plugins (microai, dev-team, discovery-team)

Development

node bin/setup-c2.js init          # Run locally
node bin/setup-c2.js --version     # Check version
npm test                           # Run tests (node --test)
npm publish --access public        # Publish to npm

Constraints

CLAUDE.md hard limit: 200 lines — generator truncates at 200 lines, enforced in both create and merge modes

Setup Targets

Target	Generated file
`claude-md`	`CLAUDE.md`
`claude-dir`	`.claude/settings.local.json`
`skills`	`.claude/skills/{project-discovery,understand-questions,deep-questions,example}/SKILL.md`
`commands`	`.claude/commands/example.md`
`hooks`	`.claude/hooks/README.md`
`agents`	`.claude/agents/example-agent.md`

Plugin: microai (v0.1.1-alpha)

Bundled plugin with 4 skills:

Skill	Command	Role
`what-next`	`/what-next [step]`	Plan & execute: assess → ideas → scope → milestones → issues → PRs
`health-check`	`/health-check [dimension]`	Verify & correct: consistency, completeness, quality, artifacts, alignment, risks
`optimize-claude-md`	`/optimize-claude-md`	Intelligent CLAUDE.md condensing within 200-line limit
`follow-up`	`/follow-up <owner/repo>`	Deep-analyze & track GitHub projects: architecture, patterns, techniques

Install: claude plugin install /path/to/setup-c2/plugins/microai

Plugin: dev-team (v0.1.0-alpha)

Autonomous development team — 5 agents, 3 skills:

Agent	Role
`architect`	Design architecture, break down tasks
`developer`	Implement code per design doc
`reviewer`	Review: correctness, security, performance, concurrency
`tester`	Write and run tests
`devops`	CI/CD, branch, commit, PR

Skill	Role
`task-runner`	Pipeline orchestrator (architect → developer → reviewer → tester → devops)
`code-standards`	Conventions + concurrency best practices for Rust, Go, Python
`git-workflow`	Deep Gitea/GitHub integration: issues, labels, milestones, PRs, releases

Command: /dev-team [task | #issue | --from-roadmap]

Install: claude plugin install /path/to/setup-c2/plugins/dev-team

Plugin: discovery-team (v0.1.0-alpha)

Codebase analysis team — 3 agents, 3 skills:

Agent	Role
`discovery-architect`	Architecture analysis, dependency graph, design patterns
`discovery-strategist`	Analysis strategy, scope identification
`discovery-auditor`	Quality audit, risks, gaps

Skill	Role
`project-discovery`	Deep dive codebase → `discovery.md`
`understand-questions`	Onboarding Q&A → `understand-questions.md`
`deep-questions`	Production readiness audit → `deep-questions.md`

Command: /project-discovery, /understand-questions, /deep-questions

Install: claude plugin install /path/to/setup-c2/plugins/discovery-team

Note: Plugins có versioning riêng biệt với CLI (CLI: v0.1.1-alpha).

Versioning

Semantic versioning with release lifecycle:

Mỗi lần release tăng patch: 0.0.1
Lifecycle: alpha → beta → rc → release (không suffix)
Ví dụ: 0.1.0 → 0.1.1-alpha → 0.1.1-beta → 0.1.1-rc → 0.1.1
Chỉ tăng minor/major khi có chỉ thị rõ ràng từ user

Repository

npm: https://www.npmjs.com/package/setup-c2
GitHub: https://github.com/taipm/setup-c2
Gitea: https://git.microai.club/taipm/setup-c2

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

It worked
Partly
It didn't

What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free