codingbuddy-rules 1.3.1 → 2.0.0

package/.ai-rules/CHANGELOG.md

@@ -5,6 +5,55 @@ All notable changes to the Multi-AI Coding Assistant Common Rules System will be
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [2.0.0] - 2026-01-06
+
+### Added
+
+- **PLAN Mode Primary Agents** with intent-based resolution
+  - `solution-architect.json`: High-level system design and architecture planning
+  - `technical-planner.json`: Low-level implementation planning with TDD and bite-sized tasks
+  - Intent-based automatic selection between architects based on prompt analysis
+  - Support for Korean and English intent patterns
+
+- **AI Model Selection and Resolution**
+  - CLI init prompt for model selection (Sonnet/Opus/Haiku)
+  - `ai.defaultModel` configuration field in `codingbuddy.config.js`
+  - Agent-level model preferences with priority resolution (agent > mode > system)
+  - `resolvedModel` field in MCP tool responses
+
+- **Parallel Agent Execution Support**
+  - `get_agent_system_prompt` MCP tool for generating subagent prompts
+  - `prepare_parallel_agents` MCP tool for batch agent preparation
+  - `parallelAgentsRecommendation` in `parse_mode` response
+  - Default specialist lists per mode (PLAN/ACT/EVAL)
+
+- **Dynamic Language Configuration**
+  - `LanguageService` supporting 10 languages (ko, en, ja, zh, es, de, fr, pt, ru, hi)
+  - `languageInstruction` field in `parse_mode` response
+  - Automatic language detection from project config
+
+- **Agent Activation Transparency**
+  - `ActivationMessageBuilder` for clear agent activation reporting
+  - `activation_message` field showing active agents with tiers
+
+### Changed
+
+- **Primary Agent Resolution**
+  - Extended `PrimaryAgentSource` type to include `intent`
+  - Priority order: explicit > config > intent > context > default
+  - Centralized Primary Agent constants with `_LIST` variants
+
+- **Token Usage Optimization**
+  - Mode-based `core.md` filtering to reduce token consumption
+  - Shared `ResponseUtils` for consolidated response generation
+
+### Fixed
+
+- Canary deployment timestamp mismatch bug
+- Type assertions replaced with proper typing in keyword service
+
+---
+
 ## [1.0.0] - 2025-11-20
 
 ### Added
@@ -48,18 +97,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
-### Added
-- **Clarification Phase** for PLAN mode (`rules/clarification-guide.md`)
-  - Optional phase triggered when AI detects ambiguous requirements
-  - Sequential Q&A with progress indicator (Question N/M format)
-  - Multiple-choice questions preferred for easy response
-  - Ambiguity assessment checklist (6 categories, triggers on 2+ unclear)
-  - Question count guidelines (2-7 questions based on complexity)
-  - Korean/English output format support
-  - Updated `rules/core.md` with Clarification Phase section
-
 ### Planned
-- Real-world usage testing across all 6 AI tools
+- Real-world usage testing across all 7 AI tools (including OpenCode)
 - Performance metrics collection
 - User feedback integration
 - Advanced examples and use cases

package/.ai-rules/adapters/claude-code.md

@@ -165,3 +165,95 @@ recommend_skills({ prompt: "Build a dashboard component" })
 **Supported Languages:** English, Korean, Japanese, Chinese, Spanish
 
 The tool returns skill recommendations with confidence levels (high/medium) and matched patterns for transparency.
+
+## Agent Hierarchy
+
+CodingBuddy uses a layered agent hierarchy for different types of tasks:
+
+### Tier 1: Primary Agents (Mode-specific)
+
+| Mode | Agents | Description |
+|------|--------|-------------|
+| **PLAN** | solution-architect, technical-planner | Design and planning tasks |
+| **ACT** | frontend-developer, backend-developer, devops-engineer, agent-architect | Implementation tasks |
+| **EVAL** | code-reviewer | Code review and evaluation |
+
+### Tier 2: Specialist Agents
+
+Specialist agents can be invoked by any Primary Agent as needed:
+
+- security-specialist
+- accessibility-specialist
+- performance-specialist
+- test-strategy-specialist
+- documentation-specialist
+- architecture-specialist
+- code-quality-specialist
+- seo-specialist
+- design-system-specialist
+
+### Agent Resolution
+
+1. **PLAN mode**: Always uses `solution-architect` or `technical-planner` based on prompt analysis
+2. **ACT mode**: Uses recommended agent from PLAN, or falls back to AI analysis
+3. **EVAL mode**: Always uses `code-reviewer`
+
+## Activation Messages
+
+When agents or skills are activated, CodingBuddy displays activation messages for transparency:
+
+### Output Format
+
+```
+🤖 solution-architect [Primary Agent]
+👤 security-specialist [Specialist] (by solution-architect)
+⚡ brainstorming [Specialist] (by technical-planner)
+```
+
+### Icons
+
+| Icon | Meaning |
+|------|---------|
+| 🤖 | Primary Agent |
+| 👤 | Specialist Agent |
+| ⚡ | Skill |
+
+### ParseMode Response Fields
+
+The `parse_mode` MCP tool returns these agent-related fields:
+
+```json
+{
+  "mode": "PLAN",
+  "delegates_to": "solution-architect",
+  "primary_agent_source": "intent",
+  "activation_message": {
+    "formatted": "🤖 solution-architect [Primary Agent]",
+    "activations": [
+      {
+        "type": "agent",
+        "name": "solution-architect",
+        "tier": "primary",
+        "timestamp": "2024-01-06T12:00:00Z"
+      }
+    ]
+  },
+  "recommended_act_agent": {
+    "agentName": "backend-developer",
+    "reason": "API implementation task detected",
+    "confidence": 0.9
+  }
+}
+```
+
+### Displaying Activation Messages
+
+AI assistants should display the `activation_message.formatted` field at the start of their response:
+
+```
+🤖 solution-architect [Primary Agent]
+
+# Mode: PLAN
+
+...
+```

package/.ai-rules/adapters/opencode.md

@@ -182,11 +182,40 @@ Once connected, you can use:
 - `search_rules`: Query AI rules and guidelines
 - `get_agent_details`: Get specialist agent information
 - `recommend_skills`: Get skill recommendations based on prompt
-- `parse_mode`: Parse PLAN/ACT/EVAL workflow mode (now returns Mode Agent information)
+- `parse_mode`: Parse PLAN/ACT/EVAL workflow mode (includes dynamic language instructions)
+
+#### Dynamic Language Configuration
+
+OpenCode agents get language instructions dynamically from the MCP server:
+
+1. **Set language in codingbuddy.config.js:**
+   ```javascript
+   module.exports = {
+     language: 'ko',  // or 'en', 'ja', 'zh', 'es', etc.
+     // ... other config
+   };
+   ```
+
+2. **Call parse_mode to get dynamic language instruction:**
+   ```bash
+   # AI should call parse_mode when user starts with PLAN/ACT/EVAL
+   # Returns languageInstruction field automatically
+   ```
+
+3. **Remove hardcoded language from agent prompts:**
+   ```json
+   {
+     "agent": {
+       "plan-mode": {
+         "prompt": "{file:...plan-mode.json}\n\n[OpenCode Override]\nMode: PLAN only. Use languageInstruction from parse_mode response.",
+       }
+     }
+   }
+   ```
 
 #### Enhanced parse_mode Response
 
-The `parse_mode` tool now returns additional Mode Agent information:
+The `parse_mode` tool now returns additional Mode Agent information and dynamic language instructions:
 
 ```json
 {
@@ -194,6 +223,8 @@ The `parse_mode` tool now returns additional Mode Agent information:
   "originalPrompt": "새로운 사용자 등록 기능을 만들어줘",
   "instructions": "설계 우선 접근. TDD 관점에서...",
   "rules": [...],
+  "language": "ko",
+  "languageInstruction": "Always respond in Korean (한국어).",
   "agent": "plan-mode",
   "delegates_to": "frontend-developer", 
   "delegate_agent_info": {
@@ -205,6 +236,8 @@ The `parse_mode` tool now returns additional Mode Agent information:
 ```
 
 **New Fields:**
+- `language`: Language code from codingbuddy.config.js
+- `languageInstruction`: Formatted instruction text for AI assistants (🆕)
 - `agent`: Mode Agent name (plan-mode, act-mode, eval-mode)
 - `delegates_to`: Which specialist agent the Mode Agent delegates to
 - `delegate_agent_info`: Detailed information about the delegate agent (optional)

package/.ai-rules/agents/README.md

@@ -22,39 +22,109 @@ AI Agent definitions for specialized development roles.
 
 ## Quick Reference: Which Agent?
 
-| 작업 유형 | 추천 에이전트 | 파일 |
-|----------|-------------|------|
-| **React/Next.js 개발** | Frontend Developer | `frontend-developer.json` |
-| **백엔드 API 개발** | Backend Developer | `backend-developer.json` |
-| **코드 리뷰 (EVAL)** | Code Reviewer | `code-reviewer.json` |
-| **아키텍처 설계** | Architecture Specialist | `architecture-specialist.json` |
-| **테스트 전략** | Test Strategy Specialist | `test-strategy-specialist.json` |
-| **성능 최적화** | Performance Specialist | `performance-specialist.json` |
-| **보안 검토** | Security Specialist | `security-specialist.json` |
-| **접근성 검토** | Accessibility Specialist | `accessibility-specialist.json` |
-| **SEO 최적화** | SEO Specialist | `seo-specialist.json` |
-| **UI/UX 디자인** | UI/UX Designer | `ui-ux-designer.json` |
-| **문서화** | Documentation Specialist | `documentation-specialist.json` |
-| **코드 품질** | Code Quality Specialist | `code-quality-specialist.json` |
-| **인프라/배포** | DevOps Engineer | `devops-engineer.json` |
+| Task Type | Recommended Agent | File |
+|-----------|-------------------|------|
+| **High-level Architecture Design** | Solution Architect | `solution-architect.json` |
+| **Implementation Planning** | Technical Planner | `technical-planner.json` |
+| **React/Next.js Development** | Frontend Developer | `frontend-developer.json` |
+| **Backend API Development** | Backend Developer | `backend-developer.json` |
+| **Code Review (EVAL)** | Code Reviewer | `code-reviewer.json` |
+| **Architecture Design** | Architecture Specialist | `architecture-specialist.json` |
+| **Test Strategy** | Test Strategy Specialist | `test-strategy-specialist.json` |
+| **Performance Optimization** | Performance Specialist | `performance-specialist.json` |
+| **Security Review** | Security Specialist | `security-specialist.json` |
+| **Accessibility Review** | Accessibility Specialist | `accessibility-specialist.json` |
+| **SEO Optimization** | SEO Specialist | `seo-specialist.json` |
+| **UI/UX Design** | UI/UX Designer | `ui-ux-designer.json` |
+| **Documentation** | Documentation Specialist | `documentation-specialist.json` |
+| **Code Quality** | Code Quality Specialist | `code-quality-specialist.json` |
+| **Infrastructure/Deployment** | DevOps Engineer | `devops-engineer.json` |
+| **Agent Management** | Agent Architect | `agent-architect.json` |
 
 ### Agent Summary
 
-| Agent | 한 줄 설명 |
-|-------|-----------|
-| Frontend Developer | React/Next.js TDD 기반 프론트엔드 개발 |
-| Backend Developer | 멀티스택 지원 백엔드 API 개발 (Node, Python, Go, Java, Rust) |
-| Code Reviewer | EVAL 모드 자동 활성화, 다차원 코드 품질 평가 |
-| Architecture Specialist | 레이어 경계, 의존성 방향, Clean Architecture |
-| Test Strategy Specialist | TDD 전략, 테스트 커버리지, 테스트 품질 |
-| Performance Specialist | Core Web Vitals, 번들 최적화, 렌더링 성능 |
-| Security Specialist | OWASP, 인증/인가, XSS/CSRF 방어 |
-| Accessibility Specialist | WCAG 2.1 AA, 시맨틱 HTML, 스크린 리더 |
-| SEO Specialist | 메타데이터, JSON-LD, Open Graph |
-| UI/UX Designer | 비주얼 계층, UX 법칙, 인터랙션 패턴 |
-| Documentation Specialist | 코드 주석, JSDoc, 문서 품질 평가 |
-| Code Quality Specialist | SOLID, DRY, 복잡도 분석 |
-| DevOps Engineer | Docker, 모니터링, 배포 최적화 |
+| Agent | Description |
+|-------|-------------|
+| Solution Architect | High-level system design and architecture planning |
+| Technical Planner | Low-level implementation planning with TDD and bite-sized tasks |
+| Frontend Developer | TDD-based frontend development with React/Next.js |
+| Backend Developer | Multi-stack backend API development (Node, Python, Go, Java, Rust) |
+| Code Reviewer | Auto-activated in EVAL mode, multi-dimensional code quality assessment |
+| Architecture Specialist | Layer boundaries, dependency direction, Clean Architecture |
+| Test Strategy Specialist | TDD strategy, test coverage, test quality |
+| Performance Specialist | Core Web Vitals, bundle optimization, rendering performance |
+| Security Specialist | OWASP, authentication/authorization, XSS/CSRF defense |
+| Accessibility Specialist | WCAG 2.1 AA, semantic HTML, screen reader support |
+| SEO Specialist | Metadata, JSON-LD, Open Graph |
+| UI/UX Designer | Visual hierarchy, UX laws, interaction patterns |
+| Documentation Specialist | Code comments, JSDoc, documentation quality assessment |
+| Code Quality Specialist | SOLID, DRY, complexity analysis |
+| DevOps Engineer | Docker, monitoring, deployment optimization |
+| Agent Architect | AI agent design, validation, checklist auditing |
+
+---
+
+## Primary Agent System
+
+**Primary Agents** are core agents that receive delegation from Mode Agents (PLAN/ACT/EVAL) to perform actual work.
+
+### Dynamic Primary Agent Resolution
+
+Primary Agent is dynamically determined based on the following priority:
+
+| Priority | Source | Description |
+|----------|--------|-------------|
+| 1 | **explicit** | Explicit request in prompt (e.g., "use backend-developer agent") |
+| 2 | **config** | Project configuration's `primaryAgent` setting |
+| 3 | **context** | Inference based on file path (e.g., `.go` → backend-developer) |
+| 4 | **default** | Default value (frontend-developer) |
+
+### Primary Agent Request Patterns
+
+**Korean:**
+```
+backend-developer로 작업해      # "Work with backend-developer" (~로 작업해 = "work with ~")
+agent-architect으로 해줘        # "Do it with agent-architect" (~으로 해줘 = "do with ~")
+devops-engineer로 개발해        # "Develop with devops-engineer" (~로 개발해 = "develop with ~")
+```
+
+**English:**
+```
+use backend-developer agent
+using frontend-developer create this
+as agent-architect, design new agent
+```
+
+### Available Primary Agents
+
+**PLAN Mode Primary Agents:**
+
+| Agent | role.type | Activation Condition |
+|-------|-----------|---------------------|
+| Solution Architect | `primary` | Architecture design, system design, technology selection |
+| Technical Planner | `primary` | Implementation planning, task breakdown, TDD planning |
+
+**ACT Mode Primary Agents:**
+
+| Agent | role.type | Activation Condition |
+|-------|-----------|---------------------|
+| Frontend Developer | `primary` | Default for ACT mode, React/Next.js projects |
+| Backend Developer | `primary` | Backend file context (.go, .py, .java, .rs) |
+| Agent Architect | `primary` | Agent-related work requests |
+| DevOps Engineer | `primary` | Dockerfile, docker-compose context |
+
+### EVAL Mode
+
+EVAL mode always uses `code-reviewer` (regardless of Primary Agent settings).
+
+### Intent-Based Resolution (PLAN Mode)
+
+PLAN mode uses intent-based resolution to automatically select between Solution Architect and Technical Planner:
+
+| Intent Pattern | Selected Agent |
+|----------------|----------------|
+| Architecture, system design, technology selection | Solution Architect |
+| Implementation plan, task breakdown, TDD | Technical Planner |
 
 ---
 
@@ -68,14 +138,15 @@ Mode Agents are workflow orchestrators that provide seamless integration with Op
 
 ```
 Mode Agents (Workflow Orchestrators)
-├── plan-mode      → delegates to → frontend-developer (or project-specific)
-├── act-mode       → delegates to → frontend-developer (or project-specific)
-└── eval-mode      → delegates to → code-reviewer
+├── plan-mode      → delegates to → [Dynamic Primary Agent]
+├── act-mode       → delegates to → [Dynamic Primary Agent]
+└── eval-mode      → delegates to → code-reviewer (always)
 
-Delegate Agents (Implementation Experts)
-├── frontend-developer     # React/Next.js expertise
+Primary Agents (Implementation Experts) - role.type: "primary"
+├── frontend-developer     # React/Next.js expertise (default)
 ├── backend-developer      # Multi-language backend expertise
-└── code-reviewer         # Quality evaluation expertise
+├── agent-architect        # AI agent framework expertise
+└── devops-engineer        # Infrastructure expertise
 
 Specialist Agents (Domain Experts)
 ├── architecture-specialist
@@ -84,13 +155,15 @@ Specialist Agents (Domain Experts)
 └── ... (other specialists)
 ```
 
+**Dynamic Resolution**: Primary Agent is dynamically determined based on prompt content, project configuration, and file context.
+
 ### Mode Agent Details
 
 | Mode Agent | Workflow | Delegates To | Purpose |
 |------------|----------|--------------|---------|
-| **plan-mode** | PLAN | frontend-developer (configurable) | Analysis and planning without changes |
-| **act-mode** | ACT | frontend-developer (configurable) | Full development with all tools |
-| **eval-mode** | EVAL | code-reviewer | Code quality evaluation |
+| **plan-mode** | PLAN | Dynamic Primary Agent | Analysis and planning without changes |
+| **act-mode** | ACT | Dynamic Primary Agent | Full development with all tools |
+| **eval-mode** | EVAL | code-reviewer (fixed) | Code quality evaluation |
 
 **Key Features:**
 - **Seamless Integration**: Works with OpenCode agent system
@@ -105,9 +178,9 @@ Specialist Agents (Domain Experts)
 ```bash
 # OpenCode CLI example
 /agent plan-mode
-새로운 기능을 만들어줘
+Create a new feature
 
-/agent act-mode  
+/agent act-mode
 ACT
 
 /agent eval-mode
@@ -121,15 +194,16 @@ When using the `parse_mode` MCP tool, you receive enhanced response with Mode Ag
 ```json
 {
   "mode": "PLAN",
-  "originalPrompt": "새로운 기능을 만들어줘",
-  "instructions": "설계 우선 접근...",
+  "originalPrompt": "Create a new feature",
+  "instructions": "Design-first approach...",
   "rules": [{"name": "rules/core.md", "content": "..."}],
   "warnings": ["No keyword found, defaulting to PLAN"],
   "agent": "plan-mode",
   "delegates_to": "frontend-developer",
+  "primary_agent_source": "default",
   "delegate_agent_info": {
     "name": "Frontend Developer",
-    "description": "React/Next.js 전문가, TDD 및 디자인 시스템 경험",
+    "description": "React/Next.js expert with TDD and design system experience",
     "expertise": ["React", "Next.js", "TDD", "TypeScript"]
   }
 }
@@ -146,6 +220,7 @@ When using the `parse_mode` MCP tool, you receive enhanced response with Mode Ag
 | `warnings` | array | No | Parsing warnings (e.g., missing keyword) |
 | `agent` | string | No | Mode Agent name (e.g., "plan-mode") |
 | `delegates_to` | string | No | Delegate agent name (e.g., "frontend-developer") |
+| `primary_agent_source` | string | No | How Primary Agent was selected: "explicit", "config", "context", "default" |
 | `delegate_agent_info` | object | No | Delegate agent details (name, description, expertise) |
 
 ### Agent Priority System
@@ -200,6 +275,89 @@ Unified specialist agents organized by domain:
 
 ## Agent Details
 
+### Solution Architect (`solution-architect.json`)
+
+> **Note**: This is a **Primary Agent** for PLAN mode, specializing in high-level system design.
+
+**Expertise:**
+
+- System Architecture Design
+- Technology Selection
+- Integration Patterns
+- Scalability Planning
+- Trade-off Analysis
+
+**Development Philosophy:**
+
+- **Brainstorm-First**: Always start with `superpowers:brainstorming` skill
+- **Multiple Options**: Present 2-3 design approaches with trade-offs
+- **Incremental Validation**: Present design in sections (200-300 words) and validate with user
+- **Document-Driven**: Save validated designs to `docs/plans/YYYY-MM-DD-<topic>-design.md`
+
+**Responsibilities:**
+
+- Analyze requirements and constraints
+- Design high-level system architecture
+- Evaluate technology options
+- Define component boundaries
+- Delegate to domain specialists (Frontend/Backend/DevOps)
+
+**Workflow:**
+
+1. Invoke `superpowers:brainstorming` skill
+2. Understand project context (files, docs, commits)
+3. Ask clarifying questions one at a time
+4. Propose 2-3 approaches with trade-offs
+5. Present design in sections with user validation
+6. Document to `docs/plans/`
+7. Offer handoff to Technical Planner
+
+---
+
+### Technical Planner (`technical-planner.json`)
+
+> **Note**: This is a **Primary Agent** for PLAN mode, specializing in detailed implementation planning.
+
+**Expertise:**
+
+- Implementation Planning
+- TDD Strategy
+- Task Decomposition
+- Code Structure Design
+- Test Design
+
+**Development Philosophy:**
+
+- **Bite-Sized Tasks**: Each task is 2-5 minutes of work
+- **TDD-First**: Red-Green-Refactor-Commit structure per task
+- **Complete Code**: Plans include full code, no placeholders
+- **Exact Paths**: Specify exact file paths for all changes
+
+**Responsibilities:**
+
+- Break down designs into bite-sized tasks (2-5 minutes each)
+- Define exact file paths and code changes
+- Design test cases with TDD approach
+- Create executable implementation plans
+- Ensure plans are context-complete for engineers
+
+**Workflow:**
+
+1. Invoke `superpowers:writing-plans` skill
+2. Read design document or requirements
+3. Identify all components and dependencies
+4. Break into bite-sized tasks (2-5 minutes each)
+5. For each task: exact files, complete code, test commands
+6. Save to `docs/plans/YYYY-MM-DD-<feature>.md`
+7. Offer execution choice (subagent vs parallel session)
+
+**Execution Options:**
+
+- **Subagent-Driven**: Execute in current session with `superpowers:subagent-driven-development`
+- **Parallel Session**: Execute in separate session with `superpowers:executing-plans`
+
+---
+
 ### Primary Developer Agent Example: Frontend Developer (`frontend-developer.json`)
 
 > **Note**: This is an example Primary Developer Agent for React/Next.js projects. Create your own agent (e.g., `backend-developer.json`, `mobile-developer.json`) following this pattern for other tech stacks.
@@ -317,6 +475,40 @@ Unified specialist agents organized by domain:
 
 ---
 
+### Agent Architect (`agent-architect.json`)
+
+> **Note**: This is a **Primary Agent** for managing AI agent configurations, schemas, and validation.
+
+**Expertise:**
+
+- Agent Schema Design (JSON/YAML)
+- Workflow Orchestration
+- Quality Assurance Automation
+- TDD for Configuration
+- Meta-Agent Patterns
+
+**Responsibilities:**
+
+- Generate new agent definitions from requirements
+- Validate agent configurations against AgentProfile schema
+- Audit code against agent mandatory_checklist
+- Optimize agent workflows and delegation chains
+- Maintain agent registry consistency
+- Design specialized agents for specific domains
+
+**Workflow:**
+
+- **Agent Creation**: TDD approach - Define schema → Create minimal JSON → Validate → Enhance → Test → Document
+- **Agent Validation**: Multi-layer validation (Schema, Checklist, Reference integrity, Documentation)
+- **Checklist Audit**: Systematic verification against mandatory_checklist items
+
+**Activation Patterns:**
+
+- Korean: "create agent", "validate agent", "audit checklist"
+- English: "create agent", "validate agent", "audit checklist"
+
+---
+
 ### Code Reviewer (`code-reviewer.json`)
 
 **Expertise:**
@@ -659,11 +851,14 @@ All agent files are located directly in `.ai-rules/agents/` directory without su
 
 ```
 .ai-rules/agents/
-├── frontend-developer.json          # Primary Developer Agent example (auto-activated)
-├── backend-developer.json           # Primary Developer Agent for backend (auto-activated)
-├── code-reviewer.json               # Core agent (auto-activated)
+├── solution-architect.json          # Primary Agent for PLAN mode (architecture)
+├── technical-planner.json           # Primary Agent for PLAN mode (implementation)
+├── frontend-developer.json          # Primary Agent for ACT mode (default)
+├── backend-developer.json           # Primary Agent for ACT mode (backend)
+├── agent-architect.json             # Primary Agent for agent management
+├── devops-engineer.json             # Primary Agent for infrastructure
+├── code-reviewer.json               # Core agent (EVAL mode, fixed)
 ├── code-quality-specialist.json     # Utility agent
-├── devops-engineer.json             # Utility agent
 ├── accessibility-specialist.json    # Domain specialist
 ├── architecture-specialist.json     # Domain specialist
 ├── ui-ux-designer.json              # Domain specialist

package/.ai-rules/agents/accessibility-specialist.json

@@ -1,6 +1,10 @@
 {
   "name": "Accessibility Specialist",
   "description": "Accessibility expert for Planning, Implementation, and Evaluation modes - unified specialist for WCAG 2.1 AA compliance, ARIA attributes, and keyboard navigation",
+  "model": {
+    "preferred": "claude-sonnet-4-20250514",
+    "reason": "Suitable model for accessibility analysis"
+  },
   "role": {
     "title": "Accessibility Engineer",
     "expertise": [

package/.ai-rules/agents/act-mode.json

@@ -2,6 +2,11 @@
   "name": "Act Mode Agent",
   "description": "ACT 모드 전용 에이전트 - 실제 구현 실행에 특화",
 
+  "model": {
+    "preferred": "claude-sonnet-4-20250514",
+    "reason": "Suitable model for code implementation"
+  },
+
   "role": {
     "title": "Act Mode Agent",
     "mode": "ACT",