AI Native development methodology for building Enterprise-grade systems rapidly. Covers 10-day development patterns, document-first design, and AI collaboration workflows. Triggers: AI native, 10-day, enterprise development, 엔터프라이즈, AI協業, AI协作
/plugin marketplace add popup-studio-ai/bkit-claude-code/plugin install bkit@bkit-marketplaceThis skill is limited to using the following tools:
Build Enterprise-grade systems with AI collaboration in 10 days. Based on bkamp.ai case study: 13 microservices, 588 commits, 1 developer + Claude Code.
┌─────────────────────────────────────────────────────────────┐
│ AI Native Development 3 Principles │
├─────────────────────────────────────────────────────────────┤
│ │
│ 1. Document-First Design │
│ → AI understands structured documents │
│ → Write design docs BEFORE code │
│ │
│ 2. Monorepo Context Control │
│ → All code in one repo for AI context │
│ → CLAUDE.md per directory for instructions │
│ │
│ 3. PR-Based Collaboration │
│ → Every change through PR (even solo dev) │
│ → AI references previous PRs for context │
│ │
└─────────────────────────────────────────────────────────────┘
docs/00-requirement/, docs/01-development/01_architecture.mdservices/shared/, services/{service}/docs/
├── 00-requirement/ # Business requirements (AI context)
├── 01-development/ # Initial design (AI reference)
├── 02-scenario/ # Implementation analysis
├── 03-refactoring/ # Improvement records
└── 04-operation/ # Operation guides
Numbering System:
42-i18n-design.md = 42nd document, i18n featureproject/
├── CLAUDE.md # Project-wide context (always read)
├── services/CLAUDE.md # Backend conventions
├── frontend/CLAUDE.md # Frontend conventions
└── infra/CLAUDE.md # Infrastructure guide
Conflict Resolution: More specific CLAUDE.md takes precedence.
Traditional QA:
1. Write test cases
2. Run test scripts
3. Analyze results
4. Write bug reports
AI Native QA:
1. Human tests UI directly
2. Claude monitors Docker logs in real-time
3. Claude detects errors instantly
4. Claude fixes code immediately
Advantages:
WARNING: This methodology requires senior developer/architect skills.
┌─────────────────────────────────────────────────────────────┐
│ Prerequisites for AI Native Development │
├─────────────────────────────────────────────────────────────┤
│ │
│ 1. Verification │
│ → Can judge if AI output is correct or wrong │
│ → Code review skills, bug detection ability │
│ │
│ 2. Direction │
│ → Knows what to build clearly │
│ → Requirements analysis, architecture design │
│ │
│ 3. Quality Bar │
│ → Defines what "good code" means │
│ → Security, performance, maintainability judgment │
│ │
│ ⚠️ Without these: AI becomes "tool for fast mistakes" │
│ │
└─────────────────────────────────────────────────────────────┘
| Characteristic | Description |
|---|---|
| Broad but shallow knowledge | Knows many things, may lack depth |
| Fast execution with direction | Implements quickly with clear instructions |
| Cannot judge without context | Needs project history, business context |
| May not recognize own mistakes | Doesn't know when it's wrong |
Managing such a "junior" requires a senior.
<type>(<scope>): <subject>
Examples:
feat(i18n): implement multi-language support with gt-next
fix(auth): resolve OAuth callback redirect issue
chore: update image tags to abc123
docs: add infrastructure guide
Types: feat, fix, chore, docs, refactor, test, style
Even for solo development, use PRs for:
Branch Strategy:
main (production)
└── staging (pre-production)
└── feature/* (development)
└── fix/* (bug fixes)
└── hotfix/* (urgent fixes)
1st Priority: Codebase (always current)
2nd Priority: CLAUDE.md / Convention docs
3rd Priority: docs/ design documents (may be outdated)
❌ NEVER: Guess based on partial reading
✅ ALWAYS: Read all relevant code completely before work
enterprise - MSA/K8s/Terraform patternszero-script-qa - Log-based QA methodologydocument-standards - PDCA document writingdevelopment-pipeline - 9-phase development processThis skill should be used when the user asks to "create a slash command", "add a command", "write a custom command", "define command arguments", "use command frontmatter", "organize commands", "create command with file references", "interactive command", "use AskUserQuestion in command", or needs guidance on slash command structure, YAML frontmatter fields, dynamic arguments, bash execution in commands, user interaction patterns, or command development best practices for Claude Code.
This skill should be used when the user asks to "create an agent", "add an agent", "write a subagent", "agent frontmatter", "when to use description", "agent examples", "agent tools", "agent colors", "autonomous agent", or needs guidance on agent structure, system prompts, triggering conditions, or agent development best practices for Claude Code plugins.
This skill should be used when the user asks to "create a hook", "add a PreToolUse/PostToolUse/Stop hook", "validate tool use", "implement prompt-based hooks", "use ${CLAUDE_PLUGIN_ROOT}", "set up event-driven automation", "block dangerous commands", or mentions hook events (PreToolUse, PostToolUse, Stop, SubagentStop, SessionStart, SessionEnd, UserPromptSubmit, PreCompact, Notification). Provides comprehensive guidance for creating and implementing Claude Code plugin hooks with focus on advanced prompt-based hooks API.