@champpaba/claude-agent-kit 2.1.1 → 2.1.3

package/.claude/agents/02-uxui-frontend.md

@@ -18,6 +18,13 @@ color: blue
 
 Complete these steps before implementation:
 
+0. **Library Requirements Check**
+   - Review `tasks.md` for "Install X", "Configure X" patterns
+   - Review `design.md` for "D1: Use X Library" decisions
+   - Use the specified libraries (WHY: team chose them for specific reasons)
+   - Example: tasks.md says "Install shadcn/ui" → Use shadcn components, follow its docs
+
+
 1. **Context Discovery** - Load project context (STEP 0)
 2. **Design Loading** - Load style guide or design principles (STEP 0.5)
 3. **Component Search** - Check existing components (STEP 3)

package/.claude/agents/04-frontend.md

@@ -18,6 +18,13 @@ color: green
 
 Complete these steps before implementation:
 
+0. **Library Requirements Check**
+   - Review `tasks.md` for "Install X", "Configure X" patterns
+   - Review `design.md` for "D1: Use X Library" decisions
+   - Use the specified libraries (WHY: team chose them for specific reasons)
+   - Example: tasks.md says "Install react-query" → Use react-query, follow its docs
+
+
 1. **Pattern Loading** - Load error-handling, state management patterns
 2. **UI Review** - Read existing UI components (from uxui-frontend)
 3. **API Review** - Read API endpoints (from backend/integration)

package/.claude/agents/05-backend.md

@@ -7,7 +7,7 @@ color: cyan
 
 # Backend Agent
 
-> **Version:** 2.0.0 (Claude 4.5 Optimized)
+> **Version:** 2.1.2 (Instruction-based Library Detection)
 > **Role:** Create API endpoints with validation, business logic, and error handling.
 
 ---
@@ -18,6 +18,12 @@ color: cyan
 
 Complete these steps before implementation:
 
+0. **Library Requirements Check**
+   - Review `tasks.md` for "Install X", "Configure X" patterns
+   - Review `design.md` for "D1: Use X Library" decisions
+   - Use the specified libraries (WHY: team chose them for specific reasons)
+   - Example: tasks.md says "Install better-auth" → Use better-auth, follow its docs
+
 1. **Pattern Loading** - Load error-handling, logging, testing patterns
 2. **Endpoint Search** - Check for existing similar endpoints
 3. **TDD Plan** - If TDD required, plan test cases first

package/.claude/agents/06-database.md

@@ -7,7 +7,7 @@ color: pink
 
 # Database Agent
 
-> **Version:** 2.0.0 (Claude 4.5 Optimized)
+> **Version:** 2.1.2 (Instruction-based Library Detection)
 > **Role:** Design schemas, create migrations, write complex queries, optimize performance.
 
 ---
@@ -18,6 +18,12 @@ color: pink
 
 Complete these steps before implementation:
 
+0. **Library Requirements Check**
+   - Review `tasks.md` for "Install X", "Configure X" patterns
+   - Review `design.md` for "D1: Use X Library" decisions
+   - Use the specified libraries (WHY: team chose them for specific reasons)
+   - Example: design.md says "Use Drizzle ORM" → Use Drizzle, follow its docs
+
 1. **Pattern Loading** - Load ORM patterns from Context7
 2. **Schema Search** - Check existing models/schemas
 3. **Design Plan** - Plan tables, relationships, indexes

package/.claude/agents/_shared/pre-work-checklist.md

@@ -1,6 +1,6 @@
 # Pre-Work Checklist (Shared by All Agents)
 
-> **Version:** 2.0.0 (Claude 4.5 Optimized)
+> **Version:** 2.1.3 (Design Spec Injection)
 > **Purpose:** Single source of truth for pre-work validation
 
 ---
@@ -9,6 +9,48 @@
 
 Complete these steps before implementation to ensure alignment with project standards:
 
+### Step 0: Library Requirements Check
+
+**Scan for required libraries before writing code:**
+
+1. **Read `tasks.md`** - Look for these patterns:
+   - `Install X and dependencies` → Use library X
+   - `Configure X with Y adapter` → Use X with adapter Y
+   - Backtick packages like `better-auth` → Required dependency
+
+2. **Read `design.md`** - Look for these patterns:
+   - `### D1: Use X Library` → Decision to use X
+   - `**Decision:** Use X` → Chosen library
+   - Alternatives table → Shows why X was chosen over others
+
+3. **List required libraries** in your validation report
+
+4. **Use the libraries specified in tasks.md/design.md**
+   - WHY: The team chose these libraries for specific reasons (compatibility, features, constraints)
+   - Custom implementations waste effort and may miss edge cases the library handles
+
+5. **Implement according to Design Spec (not library defaults)**
+   - If design.md specifies configurations → Use those values (not library defaults)
+   - If design.md specifies custom endpoints → Implement those endpoints
+   - WHY: Design decisions were made for specific project requirements
+
+**Example:**
+```
+tasks.md says: "Install better-auth and dependencies"
+design.md says: "JWT 15min + Redis refresh token 30d"
+
+→ Use better-auth library (from tasks.md)
+→ Configure JWT with 15min expiry (from design.md)
+→ Configure Redis refresh with 30d (from design.md)
+→ Implement /api/auth/refresh endpoint (from design.md)
+
+WHY not defaults? Design spec has project-specific security requirements.
+```
+
+---
+
+### Step 1-4: Standard Checks
+
 1. **Context Discovery** - Load project context via agent-discovery.md
 2. **Pattern Loading** - Load relevant patterns for your agent type
 3. **Validation Report** - Provide pre-implementation validation report
@@ -23,6 +65,18 @@ Complete these steps before implementation to ensure alignment with project stan
 ```markdown
 Pre-Implementation Validation
 
+Library Requirements: (Step 0)
+- [ ] Scanned tasks.md for "Install X" patterns
+- [ ] Scanned design.md for design decisions
+- Required libraries: {list from tasks.md/design.md}
+
+Design Spec Implementation: (Step 5)
+- [ ] Read design.md specifications
+- [ ] Identified custom configurations (not library defaults)
+- Design requirements:
+  - {requirement 1 from design.md}
+  - {requirement 2 from design.md}
+
 Project Context:
 - Project: {name}
 - Stack: {tech-stack}
@@ -41,7 +95,8 @@ Best Practices:
 Ready to Implement:
 - [ ] Task understood
 - [ ] Dependencies identified
-- [ ] Approach planned
+- [ ] Required libraries identified
+- [ ] Approach planned (using specified libraries)
 ```
 
 ---
@@ -49,9 +104,11 @@ Ready to Implement:
 ## Why This Matters
 
 Pre-work validation prevents:
+- **Wrong library choices** - Using custom code when a library was already chosen by the team
+- **Library defaults instead of spec** - Using default configs when design.md specifies custom values
 - Misaligned implementations (wrong patterns, wrong style)
 - Duplicate components (not searching existing code first)
 - Wrong package manager usage (CI/CD breaks)
 - Missing best practices (outdated patterns)
 
-→ 80% of implementation issues caught at validation stage
+→ 80% of implementation issues are caught at validation stage

package/.claude/commands/cdev.md

@@ -94,7 +94,7 @@ See: `.claude/lib/agent-executor.md` for full implementation
 ```typescript
 // Step 4: Invoke Agent with Retry & Validation
 
-// 4.1: Build agent prompt (with design reference + best practices - v1.8.0)
+// 4.1: Build agent prompt (with design reference + best practices - v2.1.2)
 function buildAgentPrompt(phase, changeContext) {
   let prompt = `
 # Phase ${phase.phase_number}: ${phase.name}
@@ -105,7 +105,62 @@ ${changeContext.proposal}
 ## Tasks
 ${changeContext.tasks}
 
+## Design Decisions (from design.md)
+${changeContext.design || '(no design.md found)'}
+
 ${phase.instructions}
+
+---
+
+## 📦 Library Requirements
+
+**Scan for required libraries before implementation:**
+
+1. **Review tasks.md above** for these patterns:
+   - "Install X and dependencies" → Use library X
+   - "Configure X with Y adapter" → Use X with adapter Y
+   - Backtick packages like \`better-auth\` → Required dependency
+
+2. **Review "Design Decisions" section above** for:
+   - "### D1: Use X Library" → Decision to use X
+   - "**Decision:** Use X" → Chosen library
+   - Specific configurations (e.g., "JWT 15min", "Redis refresh token 30d")
+   - Architecture decisions (e.g., "Hybrid Session Strategy")
+
+3. **Use the specified libraries**
+   - WHY: The team chose these for compatibility, features, or constraints
+   - Custom implementations duplicate effort and miss edge cases the library handles
+
+4. **Implement according to Design Spec (not library defaults)**
+   - If design.md specifies "JWT 15min + Redis refresh 30d" → Configure library with these values
+   - If design.md specifies custom endpoints → Implement those endpoints
+   - Library defaults are starting points, not final implementation
+   - WHY: Design decisions were made for specific project requirements
+
+**Example:**
+\`\`\`
+design.md says:
+  "Hybrid Session Strategy: JWT 15min + Redis refresh 30d"
+
+→ Use better-auth library (from tasks.md)
+→ Configure JWT plugin with 15min expiry
+→ Configure refresh token with 30d in Redis
+→ Implement /api/auth/refresh endpoint
+
+WHY not defaults? Design spec has project-specific security requirements
+that differ from library defaults.
+\`\`\`
+
+**Report in validation:**
+\`\`\`
+Libraries: better-auth, @better-auth/drizzle, ioredis
+Design Spec Implementation:
+- [x] JWT access token: 15min (design.md D1)
+- [x] Redis refresh token: 30d (design.md D2)
+- [x] /api/auth/refresh endpoint (design.md D3)
+\`\`\`
+
+---
 `
 
   // Add best-practices reference for ALL agents

package/README.md

@@ -27,6 +27,12 @@ cak init
 | frontend | 3 | Connect UI to API |
 | test-debug | 4 | Tests & fixes |
 
+## Features
+
+- **Instruction-based Library Detection** - Agents scan `tasks.md`/`design.md` for required libraries
+- **Cross-session Context** - `PROJECT_STATUS.yml` maintains state across sessions
+- **Design System v2.0** - Interactive setup with theme selection
+
 ## Commands
 
 **CLI:** `cak init` | `cak update`

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@champpaba/claude-agent-kit",
-  "version": "2.1.1",
+  "version": "2.1.3",
   "description": "Universal multi-agent template for Claude Code - AI-assisted development with specialized agents",
   "main": "bin/cli.js",
   "bin": {