anchi-toolkit 1.0.0

package/docs/FAQ.md

@@ -0,0 +1,174 @@
+# ❓ FAQ - Câu Hỏi Thường Gặp
+
+---
+
+## 🔧 Commands & Cursor
+
+### Q: Commands không hoạt động?
+**A:** 
+1. Kiểm tra đang dùng **Chat Panel** (`Ctrl+L`), không phải Terminal
+2. Restart Cursor
+3. Kiểm tra folder `.cursor/commands/` có files
+
+### Q: AI không hiểu context dự án?
+**A:** Tag files vào câu hỏi:
+```
+@.cursorrules @docs/PROJECT.md
+
+[Câu hỏi của bạn]
+```
+
+### Q: Command nào dùng trước?
+**A:** Workflow chuẩn:
+```
+/plan → /cook → /test → /review
+```
+
+### Q: Sự khác biệt /plan:fast và /plan:hard?
+**A:**
+- `/plan:fast` - Task đơn giản, nhanh
+- `/plan:hard` - Task phức tạp, cần phân tích kỹ
+
+---
+
+## 📦 Presets
+
+### Q: Chọn preset nào?
+**A:**
+| Timeline | Team Size | Chọn |
+|----------|-----------|------|
+| < 2 tuần | 1-2 người | `rapid` |
+| 1-6 tháng | 3-10 người | `professional` |
+| > 6 tháng | 10+ người | `enterprise` |
+
+### Q: Có thể đổi preset giữa chừng?
+**A:** Có, xem `presets/MIGRATION_GUIDE.md`
+
+### Q: Setup script bị lỗi?
+**A:**
+1. Kiểm tra đã cài Node.js 18+
+2. Chạy với quyền admin (Windows) hoặc sudo (Linux)
+3. Xóa `node_modules/` và chạy lại
+
+---
+
+## 🔀 Git & GitHub
+
+### Q: Branch naming?
+**A:**
+```
+feature/[name]   # New feature
+fix/[name]       # Bug fix
+hotfix/[name]    # Critical fix
+```
+
+### Q: Commit message format?
+**A:** Conventional Commits:
+```
+feat(scope): description
+fix(scope): description
+docs(scope): description
+```
+
+### Q: CI failed, làm gì?
+**A:**
+1. Check GitHub Actions logs
+2. Fix lỗi locally
+3. Push lại
+
+### Q: PR cần bao nhiêu approval?
+**A:** Tối thiểu 1 approval (configurable)
+
+---
+
+## 🗄️ Database
+
+### Q: Làm sao reset database?
+**A:**
+```bash
+npx prisma db push --force-reset
+```
+
+### Q: Migration lỗi?
+**A:**
+```bash
+npx prisma migrate reset
+npx prisma generate
+```
+
+### Q: Rapid dùng SQLite, muốn đổi sang PostgreSQL?
+**A:** 
+1. Update `DATABASE_URL` trong `.env`
+2. Chạy `npx prisma migrate dev`
+
+---
+
+## 🎨 UI/Styling
+
+### Q: Thêm component shadcn/ui?
+**A:**
+```bash
+npx shadcn-ui@latest add [component-name]
+```
+
+### Q: Customize theme?
+**A:** Edit `tailwind.config.ts` và `globals.css`
+
+---
+
+## 🧪 Testing
+
+### Q: Chạy tests?
+**A:**
+```bash
+npm test              # Unit tests
+npm run test:e2e      # E2E tests (nếu có)
+```
+
+### Q: Test coverage?
+**A:**
+```bash
+npm run test:coverage
+```
+
+---
+
+## 🚀 Deployment
+
+### Q: Deploy lên Vercel?
+**A:**
+1. Connect GitHub repo
+2. Set environment variables
+3. Deploy
+
+### Q: Environment variables?
+**A:** Không commit `.env`, dùng `.env.example` làm template
+
+---
+
+## 💡 Tips
+
+### Q: Làm sao code nhanh hơn?
+**A:**
+1. Luôn `/plan` trước khi code
+2. Dùng `/scout` khi chưa quen codebase
+3. Chia task nhỏ
+4. Commit thường xuyên
+
+### Q: AI response quá dài?
+**A:** Thêm vào prompt: "Trả lời ngắn gọn"
+
+### Q: Muốn AI follow style cụ thể?
+**A:** Cập nhật `.cursorrules` hoặc `docs/code-standards.md`
+
+---
+
+## 🆘 Vẫn Stuck?
+
+1. Search trong `docs/`
+2. Hỏi trong team chat
+3. Hỏi Team Lead
+
+---
+
+> 💡 Có câu hỏi mới? Thêm vào file này để giúp team!

package/docs/ONBOARDING.md

@@ -0,0 +1,111 @@
+# 📋 Onboarding Checklist
+
+Checklist cho member mới vào team. Đánh dấu `[x]` khi hoàn thành.
+
+---
+
+## 🗓️ Day 1: Setup Environment
+
+### Software Installation
+- [ ] Node.js 18+ installed
+- [ ] Git installed
+- [ ] Cursor IDE installed
+- [ ] Đăng nhập Cursor (Google/GitHub)
+
+### Project Setup
+- [ ] Cài anchi-toolkit: `npx anchi-toolkit init` (chọn profile phù hợp)
+- [ ] Mở project: `cursor .`
+- [ ] Chạy `/start` và chọn option
+- [ ] Chọn preset: `/use-preset [name]`
+
+### Verification
+- [ ] Chạy `npm run dev` thành công
+- [ ] Truy cập localhost:3000
+
+---
+
+## 📖 Day 1: Read Documentation
+
+### Required Reading
+- [ ] `.cursorrules` - Biết preset và conventions
+- [ ] `docs/PROJECT.md` - Hiểu dự án
+- [ ] `docs/WORKFLOW.md` - Workflow chuẩn
+- [ ] `BEGINNER_GUIDE_VI.md` - Hướng dẫn sử dụng
+
+### Optional Reading
+- [ ] `docs/code-standards.md` - Chi tiết coding standards
+- [ ] `docs/system-architecture.md` - Kiến trúc hệ thống
+- [ ] `docs/WORKFLOW.md` - Git workflow (đã merged)
+
+---
+
+## 🔧 Day 1: Practice Commands
+
+### Essential Commands
+- [ ] Thử `/scout "Tổng quan project"`
+- [ ] Thử `/plan "Simple test task"`
+- [ ] Thử `/cook "Phase 1"` (cancel nếu không cần)
+
+### Verify Understanding
+- [ ] Biết cách mở Chat Panel (`Ctrl+L`)
+- [ ] Biết cách tag files (`@filename`)
+- [ ] Biết phím tắt cơ bản
+
+---
+
+## ✅ Day 1: First Task
+
+### Complete First Task
+- [ ] Nhận task từ Team Lead
+- [ ] `/plan "task description"`
+- [ ] Review plan
+- [ ] `/cook "phase 1"`
+- [ ] `/test` hoặc manual test
+- [ ] `/review`
+- [ ] Git commit và push
+- [ ] Create Pull Request
+
+---
+
+## 📝 Day 2-3: Deep Dive
+
+### Codebase Understanding
+- [ ] Dùng `/scout` để tìm hiểu modules chính
+- [ ] Đọc `docs/CODEBASE.md`
+- [ ] Hiểu folder structure
+
+### Team Integration
+- [ ] Join team channels (Slack/Discord)
+- [ ] Attend standup meetings
+- [ ] Ask questions khi stuck
+
+---
+
+## 🎯 Week 1: Milestones
+
+- [ ] Hoàn thành 3+ tasks
+- [ ] Có ít nhất 1 PR merged
+- [ ] Comfortable với workflow
+- [ ] Có thể giúp đỡ member mới khác
+
+---
+
+## 📞 Contacts
+
+| Role | Person | Contact |
+|------|--------|---------|
+| Team Lead | [Name] | [Email/Slack] |
+| Tech Lead | [Name] | [Email/Slack] |
+| Buddy | [Name] | [Email/Slack] |
+
+---
+
+## 🆘 Stuck?
+
+1. Xem `docs/FAQ.md`
+2. Hỏi trong Chat Panel: `@.cursorrules explain this project`
+3. Hỏi Team Lead/Buddy
+
+---
+
+> ✅ Khi hoàn thành tất cả, báo Team Lead để nhận badge "Onboarded"!

package/docs/ORCHESTRATION.md

@@ -0,0 +1,345 @@
+# Agent Orchestration System
+
+> Meta-layer điều phối AI: Tự nhận thức, tự kiểm soát, tự giải thích
+
+---
+
+## 📜 Mục Lục
+
+1. [Agent Capability](#1-agent-capability--boundary)
+2. [Skill Activation](#2-skill-activation-rules)
+3. [Multi-Agent Chain](#3-multi-agent-orchestration)
+4. [Confidence/Risk](#4-confidence--risk-aware)
+5. [Explainability](#5-explainability-layer)
+6. [Failure Mode](#6-failure-mode--self-restraint)
+7. [Agent-Skill Index](#7-agent-skill-index)
+
+---
+
+## 1️⃣ Agent Capability & Boundary
+
+Mỗi agent tự khai báo năng lực và giới hạn:
+
+```yaml
+# Trong mỗi agent .md file
+---
+name: fullstack-developer
+capabilities:
+  - frontend_development
+  - backend_development
+  - api_design
+  - database_queries
+  
+boundaries:
+  - NOT: security_audit (escalate to: security-auditor)
+  - NOT: infrastructure (escalate to: devops-engineer)
+  - NOT: ui_design (escalate to: ui-ux-designer)
+
+confidence_levels:
+  frontend: 0.9
+  backend: 0.85
+  devops: 0.3
+
+escalation:
+  security_concern: security-auditor
+  performance_issue: debugger
+  architecture_decision: ddd-architect
+---
+```
+
+### Behavior
+
+```
+User: /do "Add payment with security"
+
+AI: 📊 Agent Analysis
+    ├── Primary: fullstack-developer (payment: 0.85)
+    ├── Required: security-auditor (security: 0.95)
+    └── Chain: fullstack → security-auditor → review
+```
+
+---
+
+## 2️⃣ Skill Activation Rules
+
+Thay vì bật thủ công, skill tự kích hoạt:
+
+```yaml
+# Trong mỗi skill folder
+---
+name: payment-integration
+activation:
+  triggers:
+    - keyword: ["payment", "stripe", "checkout", "billing"]
+    - file_exists: ["package.json contains stripe"]
+    - command: ["/do", "/integrate"]
+  
+  avoid_when:
+    - preset: "rapid" AND complexity > 0.7
+    - existing_payment_system: true
+  
+dependencies:
+  required: [backend-development, databases]
+  optional: [security-best-practices]
+---
+```
+
+### Behavior
+
+```
+User: /do "Add Stripe checkout"
+
+AI: 🔍 Skill Auto-Detection
+    ├── Triggered: payment-integration (keyword: "stripe")
+    ├── Dependencies: backend-development ✓, databases ✓
+    └── Activation: APPROVED
+```
+
+---
+
+## 3️⃣ Multi-Agent Orchestration
+
+Chain agents theo workflow:
+
+```yaml
+# .cursor/orchestration.yaml
+chains:
+  feature_development:
+    - agent: planner
+      output: plan.md
+    - agent: fullstack-developer
+      input: plan.md
+      output: code
+    - agent: tester
+      input: code
+      output: test_results
+    - agent: code-reviewer
+      input: [code, test_results]
+      output: review
+      
+  security_audit:
+    - agent: security-auditor
+      parallel: true
+    - agent: code-reviewer
+      
+risk_based_chains:
+  low_risk:
+    agents: [fullstack-developer]
+    auto_approve: true
+  medium_risk:
+    agents: [fullstack-developer, code-reviewer]
+    auto_approve: false
+  high_risk:
+    agents: [planner, fullstack-developer, security-auditor, code-reviewer]
+    require_lead_approval: true
+```
+
+### Behavior
+
+```
+User: /do "Add payment" (detected: high_risk)
+
+AI: 🔗 Multi-Agent Chain
+    [1] planner → Create plan
+    [2] fullstack-developer → Implement
+    [3] security-auditor → Security review
+    [4] code-reviewer → Final review
+    
+    ⚠️ Requires lead approval before commit
+```
+
+---
+
+## 4️⃣ Confidence & Risk-aware
+
+AI tự đánh giá sau mỗi action:
+
+```yaml
+# Auto-generated after each command
+assessment:
+  command: "/do 'Add payment'"
+  confidence: 0.75
+  risk_level: high
+  
+  factors:
+    - payment_handling: +risk
+    - existing_patterns: -risk
+    - test_coverage: unknown
+    
+  recommendations:
+    - require_security_review: true
+    - require_test_before_commit: true
+    - suggest_manual_verification: true
+```
+
+### Behavior
+
+```
+User: /do "Add login"
+
+AI: ✅ Implemented login feature
+
+    📊 Self-Assessment:
+    ├── Confidence: 0.85 (high)
+    ├── Risk: medium
+    └── Recommendation: Run /test before commit
+
+    Auto-enabled: Code review required
+```
+
+---
+
+## 5️⃣ Explainability Layer
+
+Command `/why` giải thích quyết định:
+
+```
+/why
+
+═══════════ DECISION EXPLANATION ═══════════
+
+Last Command: /do "Add user authentication"
+
+🤖 Agent Selection:
+   ├── fullstack-developer (primary)
+   │   └── WHY: Task involves both frontend + backend
+   ├── database-admin (secondary)
+   │   └── WHY: User table schema changes
+   └── NOT security-auditor
+       └── WHY: Basic auth, no sensitive data handling
+
+📚 Skill Activation:
+   ├── better-auth/
+   │   └── WHY: Detected NextAuth in package.json
+   ├── databases/
+   │   └── WHY: Schema modification required
+   └── NOT payment-integration/
+       └── WHY: No payment keywords detected
+
+📜 Rules Applied:
+   ├── P2: "Use service layer" → Created AuthService
+   ├── S1: "No direct DB in routes" → Used repository
+   └── C1: "Match existing patterns" → Followed project style
+
+📊 Risk Assessment:
+   └── Medium risk (auth = sensitive)
+       → Auto-enabled code review
+```
+
+---
+
+## 6️⃣ Failure Mode & Self-Restraint
+
+AI biết khi nào phải dừng:
+
+```yaml
+# .cursor/failure-modes.yaml
+stop_conditions:
+  - condition: "confidence < 0.5"
+    action: "ask_user"
+    message: "I'm not confident about this approach..."
+    
+  - condition: "missing_context"
+    action: "request_info"
+    message: "I need more information about..."
+    
+  - condition: "architecture_violation"
+    action: "stop"
+    message: "This would violate the project architecture..."
+    
+  - condition: "security_concern"
+    action: "escalate"
+    escalate_to: "security-auditor"
+    
+  - condition: "outside_expertise"
+    action: "escalate"
+    message: "This requires expertise I don't have..."
+
+restraint_rules:
+  - "Never guess database credentials"
+  - "Never modify .env without confirmation"
+  - "Never deploy without explicit approval"
+  - "Never skip security review for auth/payment"
+```
+
+### Behavior
+
+```
+User: /do "Add admin bypass for auth"
+
+AI: 🛑 STOPPING - Self-Restraint Triggered
+
+    Condition: security_concern
+    Reason: Adding auth bypass could create security vulnerability
+    
+    Options:
+    1. Explain your use case (testing, development?)
+    2. Add with proper safeguards
+    3. Cancel
+```
+
+---
+
+## 7️⃣ Agent-Skill Index
+
+Bảng chỉ mục tham chiếu:
+
+```yaml
+# .cursor/agent-skill-index.yaml
+index:
+  fullstack-developer:
+    primary_skills:
+      - frontend-development
+      - backend-development
+      - web-frameworks
+    secondary_skills:
+      - databases
+      - testing
+    escalation_partners:
+      - security-auditor (for: security)
+      - ddd-architect (for: architecture)
+      - devops-engineer (for: deployment)
+    typical_commands:
+      - /do
+      - /generate
+      
+  security-auditor:
+    primary_skills:
+      - security-best-practices
+      - authentication
+    never_combine:
+      - shopify (out of scope)
+    escalation_from:
+      - fullstack-developer
+      - database-admin
+      
+  ui-ux-designer:
+    primary_skills:
+      - ui-styling
+      - frontend-development
+    never_skills:
+      - databases
+      - devops
+```
+
+---
+
+## 📊 Summary
+
+| Before | After |
+|--------|-------|
+| Static mapping | Dynamic, conditional |
+| Agent = role | Agent = role + boundaries |
+| Skill = knowledge | Skill + activation rules |
+| AI passive | AI self-aware |
+| Trust: medium | Trust: high |
+| Team scale: limited | Team scale: good |
+
+---
+
+## 🔗 Related Files
+
+- `docs/AI_CONTRACT.md` - Constitutional rules
+- `.cursor/orchestration.yaml` - Agent chains
+- `.cursor/failure-modes.yaml` - Stop conditions
+- `.cursor/agent-skill-index.yaml` - Reference index