@champpaba/claude-agent-kit 1.8.0 → 2.0.1

package/.claude/lib/detailed-guides/agent-system.md

@@ -2,7 +2,7 @@
 
 > **Detailed guide to the multi-agent architecture**
 > **Source:** Extracted from CLAUDE.md (Navigation Hub)
-> **Version:** 1.7.0 (Opus 4.5)
+> **Version:** 2.0.0 (Claude 4.5 Optimized)
 
 ---
 
@@ -58,77 +58,38 @@ Main Claude: *Executes uxui-frontend agent directly*
 
 ---
 
-## 🔒 Main Claude Self-Check Protocol (MANDATORY)
+## 🔒 Main Claude Self-Check Protocol
 
-**⚠️ CRITICAL: Main Claude MUST complete this checklist BEFORE doing ANY work**
+**Before starting work, Main Claude routes tasks appropriately.**
 
-See: `@/.claude/lib/agent-router.md` for complete routing protocol
+→ See: `@/.claude/lib/agent-router.md` for complete routing protocol
 
-**Pre-Work Checklist (Run for EVERY user request):**
+**Quick Routing Table:**
 
-```markdown
-## ✅ Pre-Work Self-Check
-
-[ ] 1. Read user request carefully
-       - What are they asking for?
-       - What is the end goal?
-
-[ ] 2. Detect work type
-       - Is this implementation work? (writing code, creating files)
-       - Is this planning/analysis? (reading, explaining, breaking down)
-
-[ ] 3. If IMPLEMENTATION work:
-       - Read: @/.claude/contexts/patterns/task-classification.md
-       - Which agent should handle this?
-         • UI components → uxui-frontend
-         • API endpoints → backend
-         • Database schemas → database
-         • API integration → frontend
-         • Tests/bugs → test-debug
-         • Contracts → integration
-
-[ ] 4. Can Main Claude do this?
-       ✅ YES for: Planning, reading files, explaining, orchestrating workflows
-       ❌ NO for: Writing components, creating endpoints, designing schemas
-
-[ ] 5. If MUST delegate:
-       - Use Task tool with selected agent
-       - Include all necessary context
-       - Wait for agent response
-       - Update flags.json after completion (if using /cdev)
-
-[ ] 6. Report decision to user
-       ```
-       🔍 Task Analysis:
-       - Work type: [type]
-       - Requires: [agent] agent
-       - Reason: [explanation]
-
-       🚀 Invoking [agent] agent...
-       ```
-```
+| Task Type | Route To |
+|-----------|----------|
+| UI components | uxui-frontend |
+| API endpoints | backend |
+| Database schemas | database |
+| API integration | frontend |
+| Tests/bugs | test-debug |
+| Contracts | integration |
+| Planning, reading, explaining | Main Claude (direct) |
 
 **Main Claude's Role:**
-- ✅ Orchestrator (plan, coordinate, report)
-- ✅ Progress tracker (update flags.json)
-- ✅ Analyst (read files, explain code)
-- ❌ NOT implementer (no writing code directly)
+- Orchestrator (plan, coordinate, report)
+- Progress tracker (update flags.json)
+- Analyst (read files, explain code)
 
-**If Main Claude skips this self-check for implementation work, it violates system protocol.**
+WHY routing matters: Specialist agents have domain-specific validation (design tokens, TDD patterns, error handling) that ensures higher quality output.
 
 ---
 
-## ⚠️ Agent Pre-Work Requirements
+## 📋 Agent Pre-Work Requirements
 
-**STEP 0 (ALL agents):** Every agent must discover project context first
+→ See: `.claude/agents/_shared/pre-work-checklist.md` for detailed checklist
 
-**STEP 1-5 (uxui-frontend only):** Design fundamentals checklist
-
----
-
-### STEP 0: Project Discovery (ALL Agents)
-
-**Every agent MUST complete this before ANY work:**
+### STEP 0: Project Discovery (All Agents)
 
 ```
 1. Read: domain/index.md → Get current project name
@@ -138,80 +99,41 @@ See: `@/.claude/lib/agent-router.md` for complete routing protocol
 5. Report: "✅ Project Context Loaded"
 ```
 
-**STEP 0.5 (uxui-frontend ONLY):**
+### STEP 0.5: Design Context (uxui-frontend only)
 
 ```
 6. Check: design-system/STYLE_GUIDE.md exists?
-   - If YES → Read STYLE_GUIDE.md (Priority #1 - project-specific)
-   - If NO → Read .claude/contexts/design/*.md (Fallback - general principles)
-7. Report: "✅ Style Guide Loaded" or "⚠️ No style guide - using general principles"
+   - If YES → Read STYLE_GUIDE.md (project-specific)
+   - If NO → Read .claude/contexts/design/*.md (fallback)
+7. Report: "✅ Style Guide Loaded"
 ```
 
-**Why this matters:**
-- STYLE_GUIDE.md = project-specific design system (colors, spacing, components)
-- design/*.md = universal design principles (box thinking, color theory)
-- Priority: STYLE_GUIDE.md > design/*.md
+WHY: STYLE_GUIDE.md has project-specific tokens. design/*.md has universal principles.
 
-**Fallback:** If discovery fails, warn user to run `/agentsetup` or `/designsetup`
+**Fallback:** If discovery fails, suggest `/agentsetup` or `/designsetup`
 
 ---
 
 ### STEP 1-5: Design Fundamentals (uxui-frontend only)
 
-**When invoking uxui-frontend agent, Main Claude MUST include these requirements in the Task prompt:**
+→ See: `.claude/agents/02-uxui-frontend.md` for complete checklist
 
-```
-MANDATORY PRE-WORK CHECKLIST (after STEP 0):
-
-Before writing ANY code, you MUST:
-
-1. **Read ALL design contexts:**
-   - @/.claude/contexts/design/index.md
-   - @/.claude/contexts/design/box-thinking.md
-   - @/.claude/contexts/design/color-theory.md
-   - @/.claude/contexts/design/spacing.md
-   - @/.claude/contexts/patterns/ui-component-consistency.md
-   - @/.claude/contexts/patterns/frontend-component-strategy.md
-
-2. **Do Box Thinking Analysis:**
-   - Identify all boxes (parent, children, siblings)
-   - Document relationships (container, adjacent, nested)
-   - Plan space flow using spacing scale (8, 16, 24, 32, 40, 48px)
-   - Plan responsive behavior (stack/merge/compress)
-
-3. **Search for Existing Components:**
-   - Glob: "**/*{Keyword}*.{tsx,jsx,vue}"
-   - Grep: "[similar-pattern]"
-   - Decision: Reuse > Compose > Extend > Create New
-   - If creating new: Extract design tokens from most similar component
-
-4. **Extract Design Tokens from Reference Component:**
-   ```typescript
-   const DESIGN_TOKENS = {
-     spacing: { padding: '[from reference]', gap: '[from reference]' },
-     colors: { bg: '[theme token]', text: '[theme token]', border: '[theme token]' },
-     shadows: '[from reference - e.g., shadow-sm]',
-     borderRadius: '[from reference - e.g., rounded-md]'
-   }
-   ```
-
-5. **Report Pre-Implementation Analysis:**
-   You MUST provide a detailed report covering steps 1-4 BEFORE writing any code.
-
-CRITICAL RULES:
-- ❌ NO hardcoded colors (text-gray-500) → ✅ Use theme tokens (text-foreground/70)
-- ❌ NO arbitrary spacing (p-5) → ✅ Use spacing scale (p-4, p-6)
-- ❌ NO inconsistent icons (h-5 w-5, opacity-50) → ✅ Match reference (h-4 w-4, text-foreground/70)
-- ❌ NO creating duplicate components → ✅ Search and reuse first
-
-If you skip these steps, your work will be rejected.
-```
+**Summary:**
+1. Read design contexts (box-thinking, color-theory, spacing)
+2. Do Box Thinking Analysis (identify boxes, relationships, spacing)
+3. Search for existing components (Reuse > Compose > Extend > Create)
+4. Extract design tokens from reference component
+5. Report pre-implementation analysis
+
+**Style Guidelines:**
+
+| Instead of | Use | WHY |
+|------------|-----|-----|
+| text-gray-500 | text-foreground/70 | Theme-aware |
+| p-5 | p-4 or p-6 | Spacing scale |
+| h-5 w-5, opacity-50 | h-4 w-4, text-foreground/70 | Consistency |
 
-**Why this enforcement matters:**
-- Prevents visual inconsistency (mismatched colors, spacing, shadows)
-- Ensures component reuse (avoids duplicates)
-- Maintains design system integrity
-- Saves implementation time
+WHY these steps matter: Prevents visual inconsistency, ensures component reuse, maintains design system integrity.
 
 ---
 

package/.claude/lib/detailed-guides/taskmaster-analysis.md

@@ -166,7 +166,7 @@ Priority Distribution:
 - 🟢 LOW: 1
 
 Risk Assessment:
-- 🚨 HIGH: 2 tasks → TDD required
+- 🔴 HIGH: 2 tasks → TDD required
 - ⚠️ MEDIUM: 3 tasks
 - ✅ LOW: 3 tasks
 

package/.claude/lib/document-loader.md

@@ -1,7 +1,7 @@
 # Document Loader - Unified Loading Pattern
 
 > **Token-efficient, consistent document loading for all commands and agents**
-> **Version:** 1.0.0 (Context Optimization)
+> **Version:** 2.0.0 (Claude 4.5 Optimized)
 
 ---
 
@@ -120,9 +120,9 @@ function buildDesignReference(projectName: string): string {
 
   // Don't load content! Just send paths + minimal summary
   return `
-## 🎨 Design System (Required Reading)
+## 🎨 Design System Reference
 
-**MANDATORY for uxui-frontend agent (STEP 0.5):**
+**For uxui-frontend agent (STEP 0.5):**
 
 1. Read: ${designContextPath} (~1K tokens)
    → Project design summary, file paths
@@ -133,15 +133,15 @@ function buildDesignReference(projectName: string): string {
 3. Optional: ${styleGuidePath} (selective sections ~2K tokens)
    → Full guide - load Component Styles, Layout Patterns if needed
 
-**Critical Rules:**
-- ❌ NO hardcoded colors (text-gray-500)
-- ✅ USE theme tokens (text-foreground/70)
-- ❌ NO arbitrary spacing (p-5)
-- ✅ USE spacing scale (p-4, p-6)
+**Style Guidelines:**
+| Instead of | Use | WHY |
+|------------|-----|-----|
+| text-gray-500 | text-foreground/70 | Theme-aware |
+| p-5 | p-4 or p-6 | Spacing scale |
 
-**You MUST report:**
-✅ "Design Context Loaded: design-context.md + STYLE_TOKENS.json"
-✅ "Design Tokens Extracted: [list key tokens]"
+**Report format:**
+"Design Context Loaded: design-context.md + STYLE_TOKENS.json"
+"Design Tokens Extracted: [list key tokens]"
   `
 
   // Total sent: ~200 tokens (reference only, not content!)
@@ -229,20 +229,17 @@ ${report.join('\n')}
 
 ---
 
-## 🚨 Critical Rules
-
-### DO:
-- ✅ Always load design-context.md first (if exists)
-- ✅ Load STYLE_TOKENS.json for UI work
-- ✅ Load STYLE_GUIDE.md selectively (specific sections only)
-- ✅ Validate files exist before loading
-- ✅ Report what was loaded (transparency)
-
-### DON'T:
-- ❌ Load full STYLE_GUIDE.md unless absolutely needed
-- ❌ Skip design-context.md (it's the entry point!)
-- ❌ Load design files for non-UI work (backend, database)
-- ❌ Hardcode paths (use project name variable)
+## 📋 Best Practices
+
+| Practice | WHY |
+|----------|-----|
+| Load design-context.md first | Entry point with file paths |
+| Load STYLE_TOKENS.json for UI work | Lightweight token reference |
+| Load STYLE_GUIDE.md selectively | Save tokens (5K → 2K) |
+| Validate files exist before loading | Prevent errors |
+| Report what was loaded | Transparency for debugging |
+| Skip design files for backend/database | Not needed, saves tokens |
+| Use project name variable for paths | Portability |
 
 ---
 

package/.claude/lib/flags-updater.md

@@ -2,18 +2,16 @@
 
 > **WHO:** Main Claude (orchestrator)
 > **WHEN:** Immediately after sub-agent completes phase
-> **HOW:** Follow this exact flow
-> **PURPOSE:** Ensure real-time progress tracking for users
+> **PURPOSE:** Real-time progress tracking for users
+> **Version:** 2.0.0 (Claude 4.5 Optimized)
 
 ---
 
 ## 🎯 Core Principle
 
-**Main Claude is ALWAYS responsible for updating flags.json**
+**Main Claude updates flags.json after each phase.**
 
-- ❌ NOT sub-agents (they don't have direct access)
-- ❌ NOT user (they shouldn't have to)
-- ✅ Main Claude after EVERY phase completion
+WHY: Sub-agents don't have direct access to flags.json. Users shouldn't have to manually track progress. Immediate updates ensure accurate /cstatus and /cview output.
 
 ---
 
@@ -24,14 +22,14 @@ Step 1: Sub-agent responds with completion message
    ↓
 Step 2: Main Claude validates response quality
    ↓
-Step 3: Main Claude updates flags.json (MANDATORY - THIS STEP!)
+Step 3: Main Claude updates flags.json
    ↓
 Step 4: Main Claude reports progress to user
    ↓
 Step 5: Main Claude asks to continue (or auto-continue)
 ```
 
-**⚠️ CRITICAL:** Step 3 CANNOT be skipped. Ever.
+WHY Step 3 is important: Without immediate update, /cstatus shows stale data and users can't track real progress.
 
 ---
 
@@ -279,34 +277,28 @@ function calculateTimeRemaining(flags: Flags): number {
 
 ---
 
-## ⚠️ CRITICAL RULES
+## 📋 Best Practices
 
-### Rule 1: NEVER Skip This Step
+### Update Timing
 
-```markdown
-❌ WRONG:
-Sub-agent completes → Main Claude asks user to continue
-(flags.json not updated)
+| When | Action |
+|------|--------|
+| Sub-agent returns success | Update flags.json immediately |
+| Before asking user to continue | Flags should already be updated |
+| After each phase | Update individually (not batched) |
 
-✅ CORRECT:
-Sub-agent completes → Main Claude updates flags.json → Main Claude reports progress → Main Claude asks user to continue
-```
+WHY: Immediate updates ensure /cstatus always shows accurate progress.
 
-### Rule 2: Update IMMEDIATELY After Sub-Agent Responds
+### Correct Flow
 
-```markdown
-Don't wait for:
-- ❌ User confirmation
-- ❌ Next phase to start
-- ❌ Batch multiple updates
-
-Update:
-- ✅ As soon as sub-agent returns success
-- ✅ Before asking user anything
-- ✅ After EACH individual phase
+```
+Sub-agent completes
+   → Main Claude updates flags.json
+   → Main Claude reports progress
+   → Main Claude asks user to continue
 ```
 
-### Rule 3: Validate Before Update
+### Validate Before Update
 
 ```typescript
 // Check 1: flags.json exists
@@ -329,10 +321,10 @@ if (!agentResponse.includes('✅') && !agentResponse.includes('Complete')) {
 }
 ```
 
-### Rule 4: Always Report to User
+### Report to User
 
 ```markdown
-After updating flags.json, ALWAYS show:
+After updating flags.json, show:
 
 📊 Progress Updated:
    ✅ Phase "frontend-mockup" marked complete
@@ -466,4 +458,4 @@ if (flags.ready_to_archive) {
 
 ---
 
-**💡 Remember:** Main Claude updates flags.json. Always. No exceptions.
+**💡 Summary:** Main Claude updates flags.json immediately after each phase completes. This ensures accurate progress tracking.

package/.claude/templates/STYLE_GUIDE.template.md

@@ -815,7 +815,7 @@ Before submitting a new component, verify:
 
 ---
 
-## 🚨 Troubleshooting
+## 🔧 Troubleshooting
 
 ### Colors look different than expected
 

package/.claude/templates/design-context-template.md

@@ -109,7 +109,7 @@ Buttons (hover): shadow-sm
 
 ---
 
-## 🚨 Critical Design Rules
+## ⚠️ Design Rules
 
 ### Colors
 - ❌ **NO** hardcoded colors: `text-gray-500`, `#64748b`

package/.claude/templates/phases-sections/frontend-mockup.md

@@ -70,11 +70,14 @@ Create UI components with mockup data. Focus on visual design, layout, and user
 - Error states: user-friendly messages (mockup errors)
 - Responsive design: test 3 breakpoints minimum
 
-**CRITICAL RULES:**
-- ❌ NO hardcoded colors → ✅ Use theme tokens
-- ❌ NO arbitrary spacing → ✅ Use spacing scale
-- ❌ NO inconsistent icons → ✅ Match reference components
-- ❌ NO creating duplicate components → ✅ Search and reuse first
+**Design Consistency Rules:**
+
+| Instead of | Use | WHY |
+|------------|-----|-----|
+| Hardcoded colors | Theme tokens | Maintains dark/light mode support |
+| Arbitrary spacing | Spacing scale (8, 16, 24...) | Consistent visual rhythm |
+| Random icons | Reference component icons | Visual consistency |
+| Creating new components | Search and reuse existing | Prevents duplicates, faster development |
 
 ## ✅ Success Criteria
 

package/README.md

@@ -903,60 +903,98 @@ Built with:
 
 ---
 
-## 🆕 What's New in v1.8.0
+## 🆕 What's New in v2.0.0
 
-**Major: Token Optimization & Streamlined Workflow** 🚀
+**Two Major Improvements:** Claude 4.5 Optimization + Design System v2.0 🚀🎨
 
-### Key Changes
+### Part 1: Claude 4.5 Optimization
 
-1. **Removed `/psetup` and `/agentsetup`** - Best practices now auto-generated by `/csetup`
-2. **Removed documentation/report phases** - No more `PHASE_*.txt`, `*_REPORT.md` clutter
-3. **Verbose terminal output** - Results shown in terminal, not saved to files
-4. **Auto-cleanup temp files** - Any forbidden files auto-deleted after phases
-5. **Agents inject best-practices** - `/cdev` now injects relevant best-practices paths into agent prompts
+1. **Claude 4.5 Best Practices** - All files refactored for better AI comprehension
+2. **Professional Tone** - Replaced aggressive language with respectful instructions
+3. **Positive Instructions** - "Use X" instead of "Don't do Y" with WHY explanations
+4. **Token Reduction** - Agent files ~65% smaller + 30 supporting files optimized
+5. **Shared Components** - New `.claude/agents/_shared/` folder prevents duplication
 
-### Streamlined Workflow
+### Tone Calibration
 
-**Before (v1.7.x):**
-```bash
-/psetup                    # Setup project
-/csetup landing-page       # Setup change
-/cdev landing-page         # Execute
-→ Creates: PHASE_11_DELIVERY.txt, FINAL_REPORT.md, etc. 😞
-```
+| Before | After | WHY |
+|--------|-------|-----|
+| "MUST", "WILL BE REJECTED" | Professional tone | Claude 4.5 works better with respectful instructions |
+| "Don't do X", "Never Y" | "Use X instead" | Positive instructions are clearer |
+| "🚨 CRITICAL!", "🚫 Forbidden" | "📋 Guidelines", "⚠️ Note" | Less aggressive, same clarity |
+| Rules without context | Rules with WHY | Claude applies rules more intelligently |
 
-**After (v1.8.0):**
-```bash
-/csetup landing-page       # Setup + auto-detect stack + generate best-practices
-/cdev landing-page         # Execute
-→ Verbose summary in terminal ✅
-→ No temp files created ✅
+### New Shared Components
+
+```
+.claude/agents/_shared/
+├── pre-work-checklist.md     # Common validation steps
+├── package-manager.md        # Package manager protocol
+├── documentation-policy.md   # What files to create
+├── agent-boundaries.md       # When to use which agent
+└── README.md                 # Overview
 ```
 
-### Token Savings
+### Token Savings (Agents)
+
+| Agent | Before | After | Reduction |
+|-------|--------|-------|-----------|
+| uxui-frontend | ~1037 | ~375 | 64% |
+| integration | ~600 | ~210 | 65% |
+| backend | ~700 | ~244 | 65% |
+| database | ~680 | ~273 | 60% |
+| frontend | ~650 | ~296 | 54% |
+| test-debug | ~580 | ~252 | 57% |
+| **Total** | **~4247** | **~1650** | **61%** |
+
+Plus ~500 tokens in shared files = **~2150 total** (was ~4247)
+
+### Part 2: Design System v2.0
+
+**Complete overhaul of design system with interactive setup, theme selection, and smart page planning.**
+
+### New Design System Features
+
+- 🎯 **`/extract`** - Multi-URL design extraction with style detection
+- 🎭 **`/designsetup`** - Interactive 3-round loop with theme recommendations
+- 📦 **`tokens.json`** - Enhanced design tokens with style/theme/animations (~800 tokens)
+- 📝 **`patterns/*.md`** - Selective code patterns (buttons, cards, forms, scroll-animations, decorations)
+- 🔍 **Auto Page Type Detection** - `/pageplan` auto-detects landing/dashboard/auth
 
-| Template | Before | After | Saved |
-|----------|--------|-------|-------|
-| full-stack | 19 phases | 17 phases | ~25 min |
-| frontend-only | 11 phases | 9 phases | ~25 min |
-| backend-only | 10 phases | 8 phases | ~25 min |
-| bug-fix | 5 phases | 4 phases | ~10 min |
+### Page Type Handling
 
-### Forbidden Files (Auto-Deleted)
+| Page Type | Decorations | Scroll Anims | Buyer Avatar | Patterns Loaded |
+|-----------|-------------|--------------|--------------|-----------------|
+| Landing/Marketing | ✅ Full | ✅ Enabled | ✅ Enabled | buttons, cards, scroll-anims, decorations |
+| Dashboard/Admin | ❌ Minimal | ❌ Disabled | ❌ Skipped | buttons, cards, forms |
+| Auth (Login/Register) | ❌ None | ❌ Disabled | ❌ Skipped | buttons, forms |
 
-Agents should only create **actual code files**. These are now forbidden:
-- Report/summary files (`*_REPORT.md`, `*_SUMMARY.txt`)
-- Phase output files (`PHASE_*.txt`, `*_DELIVERY.txt`)
-- Temp files (ALL_CAPS filenames)
+### Design System Token Savings
 
-**Rule of thumb:** If it wouldn't be committed to git, don't create it.
+| Approach | Tokens | Improvement |
+|----------|--------|-------------|
+| Old: Full STYLE_GUIDE.md | ~5000 | - |
+| New: tokens.json + selective patterns | ~800-1200 | **84%** |
 
-### Best Practices Now Enforced
+### Files Refactored (36 total)
 
-- `/csetup` auto-detects stack from: `package.json` → `design.md` → `proposal.md`
-- Auto-generates best practices from Context7 MCP
-- `/cdev` injects relevant files into agent prompts
-- Agents MUST report "Best Practices Loaded" (validated)
+- **6 agent files** - Professional tone, positive instructions
+- **6 lib files** - WHY explanations, table formats
+- **4 command files** - Softer language, guidelines
+- **11 pattern files** - "⚠️ Common Mistakes" instead of "🚨 Critical"
+- **3 design files** - "should check" instead of "MUST check"
+- **3 template files** - Table with WHY
+- **3 other lib files** - Consistent formatting
+
+### Best Practices Applied (Claude 4 Guidelines)
+
+1. **Tone Calibration** - Professional, direct (not aggressive)
+2. **Action Orientation** - Explicit "Write code" vs "Consider"
+3. **Prevent Overengineering** - Clear boundaries
+4. **Encourage Exploration** - Read before implementing
+5. **Rich Output When Needed** - Specify requirements
+6. **Context for Rules** - Explain WHY
+7. **Positive Instructions** - "Use X" not "Don't Y"
 
 ---
 
@@ -1265,6 +1303,27 @@ All your customizations in `.claude/contexts/domain/` are preserved!
 
 ## 📜 Changelog
 
+### v2.0.0 (2025-11-30)
+**Major: Claude 4.5 Optimization - Full Template Refactor**
+
+**Added:**
+- Shared components folder `.claude/agents/_shared/`
+- WHY explanations for all rules
+
+**Changed:**
+- All 6 agent files refactored (~65% smaller)
+- 30 supporting files updated for Claude 4.5 best practices
+- "MUST", "CRITICAL" → Professional tone
+- "Don't do X" → "Use Y instead" with WHY
+- "🚨" → "⚠️" throughout
+
+**Benefits:**
+- Better AI comprehension with professional tone
+- 61% token reduction in agent files
+- No duplication (shared components)
+
+---
+
 ### v1.8.0 (2025-11-26)
 **Major: Token Optimization & Streamlined Workflow**
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@champpaba/claude-agent-kit",
-  "version": "1.8.0",
+  "version": "2.0.1",
   "description": "Universal multi-agent template for Claude Code - AI-assisted development with specialized agents",
   "main": "bin/cli.js",
   "bin": {