feat: Add Claude Code Skills integration

Created Aegis TDD as Claude Code Skill for reusable workflow.

Global Skill (~/.claude/skills/aegis-tdd/):
- SKILL.md with TDD workflow
- YAML frontmatter with triggers
- Templates for core/services/handlers
- AI Trinity workflow integration
- Coverage targets and patterns

Project Skill (.claude/skills/aegis-tdd-example/):
- Project-specific patterns
- Established test structures
- Real examples from this project
- Tool integration (dashboard, generate-test)

Skills Features:
- Auto-activates on TDD keywords
- Provides RED-GREEN-REFACTOR guidance
- Includes mocking templates
- References Aegis AI patterns
- 95%+ coverage targets

Usage:
- Global: Works in any project
- Project: Team-shared patterns
- Auto-activation on "create tests", "TDD", etc.

Documentation:
- SKILLS_PROPOSAL.md (comparison & recommendations)
- Practical implementation guide

Recommendation: Hybrid approach
- Global skill for general TDD
- Project skill for team patterns

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

Author: hwandam77

.claude/skills/aegis-tdd-example/SKILL.md

@@ -0,0 +1,260 @@
+---
+name: aegis-tdd-example
+description: Aegis AI project-specific TDD examples and patterns. Use when working on this project to follow established patterns for handlers, services, and core modules with 99%+ coverage standards.
+allowed-tools: Read, Write, Edit, Bash, Grep, Glob
+---
+
+# 🛡️ Aegis TDD Example - Project Patterns
+
+## Purpose
+
+Provide Aegis AI project-specific TDD patterns and examples.
+
+**Project Context**: 99.33% coverage, 105 tests, LEGENDARY level 👑
+
+---
+
+## 🎯 Project-Specific Standards
+
+### Coverage Requirements
+
+- **Core modules**: 95%+ (currently 98.81%)
+- **Services**: 95%+ (currently 100%)
+- **MCP**: 100% (currently 100%)
+
+### Test Count
+
+- Minimum 10 tests per module
+- Edge cases mandatory
+- Error handling 100% covered
+
+---
+
+## 📋 Established Patterns
+
+### Pattern 1: Core Module (handlerLoader example)
+
+**Test Structure**:
+```javascript
+describe('HandlerLoader', () => {
+  let instance;
+
+  beforeEach(() => {
+    instance = require('../../src/core/handlerLoader');
+    instance.clearHandlers();  // Isolation
+  });
+
+  // Use real fixtures instead of mocks
+  test('loads handlers from directory', () => {
+    const fixturesPath = path.join(__dirname, '../fixtures/handlers-valid');
+    instance.loadHandlers(fixturesPath);
+    expect(instance.listHandlers()).toContain('handlerA');
+  });
+});
+```
+
+**Key Points**:
+- ✅ Use `clearHandlers()` for isolation
+- ✅ Real fixtures over complex mocks
+- ✅ Test actual file system operations
+
+---
+
+### Pattern 2: Service with Mocking (geminiService example)
+
+**Mocking Strategy**:
+```javascript
+const { spawn } = require('child_process');
+jest.mock('child_process');
+
+describe('GeminiService', () => {
+  let mockProcess;
+
+  beforeEach(() => {
+    mockProcess = {
+      stdout: { on: jest.fn(), setEncoding: jest.fn() },
+      stderr: { on: jest.fn() },
+      on: jest.fn(),
+      kill: jest.fn(),  // Important for timeout tests
+    };
+    spawn.mockReturnValue(mockProcess);
+  });
+
+  test('handles timeout correctly', async () => {
+    mockProcess.on.mockImplementation(() => {});  // No close event
+
+    await expect(service.execute('test', { timeout: 100 }))
+      .rejects.toThrow('timeout');
+    expect(mockProcess.kill).toHaveBeenCalled();
+  });
+});
+```
+
+**Key Points**:
+- ✅ Mock `child_process.spawn`
+- ✅ Include `kill` for timeout tests
+- ✅ Test error paths thoroughly
+
+---
+
+### Pattern 3: MCP Protocol (index.js example)
+
+**JSON-RPC Testing**:
+```javascript
+describe('MCP Server', () => {
+  test('handles tools/list request', async () => {
+    const request = {
+      jsonrpc: '2.0',
+      method: 'tools/list',
+      id: 1,
+    };
+
+    const response = await server.handleRequest(request);
+
+    expect(response.jsonrpc).toBe('2.0');
+    expect(response.id).toBe(1);
+    expect(response.result).toHaveProperty('tools');
+  });
+
+  test('returns error for invalid jsonrpc version', async () => {
+    const response = await server.handleRequest({
+      jsonrpc: '1.0',
+      method: 'tools/list',
+      id: 1,
+    });
+
+    expect(response.error.code).toBe(-32600);
+  });
+});
+```
+
+**Key Points**:
+- ✅ Test all error codes (-32600, -32601, -32602, -32603, -32700)
+- ✅ Verify JSON-RPC 2.0 compliance
+- ✅ 100% coverage mandatory
+
+---
+
+## 🧪 Project Test Fixtures
+
+**Location**: `tests/fixtures/`
+
+**Example**:
+```
+tests/fixtures/
+├── handlers-valid/
+│   ├── handlerA.js
+│   └── handlerB.js
+└── handlers-mixed/
+    ├── handlerC.js
+    ├── config.json  # Should be ignored
+    └── README.md    # Should be ignored
+```
+
+**Usage**:
+```javascript
+const fixturesPath = path.join(__dirname, '../fixtures/handlers-valid');
+handlerLoader.loadHandlers(fixturesPath);
+```
+
+---
+
+## 🎮 Project Tools
+
+### Dashboard
+
+```bash
+npm run dashboard
+```
+
+Shows:
+- Current coverage by module
+- TDD level (LEGENDARY 👑)
+- Progress bars
+- Achievement percentage
+
+### Test Generator
+
+```bash
+npm run generate-test <moduleName> <category>
+
+# Example
+npm run generate-test authHandler handlers
+```
+
+Generates:
+- Test file with template
+- Source file template
+- TDD workflow guidance
+
+---
+
+## 📊 Achieved Results
+
+**Coverage by Layer**:
+- Core: 98.81% (5 modules)
+- Services: 100% (3 services) 🏆
+- MCP: 100% (index.js) 🏆
+
+**100% Coverage Modules** (7):
+1. index.js
+2. stageOrchestrator.js
+3. stateManager.js
+4. workflowEngine.js
+5. geminiService.js
+6. qwenService.js
+7. codexService.js
+
+---
+
+## 💡 When to Use This Skill
+
+**Auto-activate when user**:
+- Asks to "create tests"
+- Mentions "TDD" or "test-driven"
+- Wants to add a new module/service
+- Asks about test coverage
+- Mentions "follow Aegis AI pattern"
+
+**Steps**:
+1. Identify module category (core/services/handlers)
+2. Use appropriate pattern from this skill
+3. Follow TDD cycle
+4. Aim for 95%+ coverage
+5. Use project tools (dashboard, generate-test)
+
+---
+
+## 🏆 Success Criteria
+
+- [ ] Tests written before implementation
+- [ ] Coverage 95%+
+- [ ] All tests passing
+- [ ] Edge cases covered
+- [ ] Mocking strategy appropriate
+- [ ] JSDoc complete
+- [ ] CI/CD passing
+
+---
+
+## 📚 Reference Files
+
+**Within this project**:
+- Examples: All files in `tests/`
+- Patterns: `src/core/`, `src/services/`
+- Documentation: `docs/TDD_POLICY.md`
+- Benefits: `docs/PRACTICAL_BENEFITS.md`
+
+**Commands**:
+```bash
+npm test                 # Run tests
+npm run test:coverage    # Coverage report
+npm run dashboard        # Visual dashboard
+npm run generate-test    # Generate new test
+```
+
+---
+
+**Level**: 👑 LEGENDARY (99.33% coverage)
+**Tests**: 105 (100% pass)
+**Status**: Production Ready ✅