Design Patterns MCP Server 🎯

An intelligent MCP (Model Context Protocol) server that provides design pattern recommendations using semantic search and vector embeddings. This project offers access to a comprehensive catalog of 555+ design patterns through a natural language interface.

📋 Overview

The Design Patterns MCP Server is a specialized server that integrates with AI assistants (like Claude, Cursor) to provide intelligent design pattern recommendations. It uses advanced semantic search technologies to find the most appropriate patterns based on natural language problem descriptions.

✨ Key Features

🔍 Intelligent Semantic Search: Find patterns using natural problem descriptions
📚 Comprehensive Catalog: 555+ patterns organized in 20+ categories
🎯 Contextual Recommendations: Suggestions based on programming language and domain
⚡ Vector Search: Uses SQLite with vector extensions for efficient search
🌐 Multi-language: Support for multiple programming languages
🔧 MCP Integration: Compatible with Claude Code, Cursor and other MCP clients
🚀 High Performance: Object Pool pattern prevents memory leaks, 30-40% faster queries
💾 Smart Caching: LRU cache with 85%+ hit rate
🏗️ SOLID Architecture: Clean, maintainable, and testable codebase

🆕 Recent Improvements (v0.2.1)

Architecture Refactoring (October 2025)

✅ Object Pool Pattern: Eliminates memory leaks with bounded prepared statements
✅ Service Layer: Centralized business logic with PatternService
✅ Facade Pattern: Simplified handlers via PatternHandlerFacade
✅ Dependency Injection: Full DI Container integration for testability
✅ Performance: 30-40% faster on repeated queries with smart caching
✅ Code Quality: 40% reduction in main server file (704→422 lines)
✅ Pattern Catalog: Expanded to 555+ patterns with code examples

🗂️ Available Pattern Categories

GoF Patterns: Classic patterns (Creational, Structural, Behavioral)
Architectural Patterns: MVC, MVP, MVVM, Clean Architecture, Hexagonal
Microservices Patterns: Circuit Breaker, Event Sourcing, CQRS, Saga
Cloud Patterns: Auto-scaling, Load Balancing, Service Discovery
AI/ML Patterns: Model Training, RAG, Few-Shot Learning, Continual Learning
React Patterns: Hooks, Server Components, Suspense, React 19 features
Enterprise Patterns: Repository, Unit of Work, Dependency Injection
Security Patterns: Authentication, Authorization, Data Protection
Performance Patterns: Caching, Lazy Loading, Object Pool, Connection Pooling
Concurrency Patterns: Producer-Consumer, Thread Pool, Actor Model
Integration Patterns: Message Queue, Event Bus, API Gateway
Data Access Patterns: Active Record, Data Mapper, Query Object
Testing Patterns: Test Double, Page Object, Builder Pattern for tests
Functional Patterns: Monads, Functors, Higher-Order Functions
Reactive Patterns: Observer, Publisher-Subscriber, Reactive Streams
DDD Patterns: Aggregate, Value Object, Domain Service, Bounded Context
Game Development Patterns: State Machine, Component System, Object Pool
Mobile Patterns: Model-View-Intent, Redux-like patterns, Offline-First
IoT Patterns: Device Twin, Telemetry Ingestion, Edge Processing
Blockchain/Web3 Patterns: DeFi, NFT, DAO, Cross-chain
Anti-Patterns: Practices to avoid and their solutions

🏗️ Project Architecture

Refactored Architecture (v0.2.x)

src/ ├── adapters/ # Adapters for external services (LLM, Embeddings) ├── builders/ # Builders for complex objects ├── cli/ # Command line interface ├── core/ # Core domain logic and DI Container │ └── container.ts # Dependency Injection Container with TOKENS ├── db/ # Database configuration and migrations ├── facades/ # Facade pattern implementations │ └── pattern-handler-facade.ts # Simplifies MCP handlers ├── factories/ # Factories for object creation ├── lib/ # Auxiliary libraries and MCP utilities ├── models/ # Data models and types (unified Pattern interface) ├── repositories/ # Data access layer (Repository Pattern) │ ├── interfaces.ts # Repository contracts │ └── pattern-repository.ts # SQLite implementation ├── services/ # Business services and orchestration │ ├── cache.ts # LRU Cache service │ ├── database-manager.ts # Database operations with Object Pool │ ├── pattern-service.ts # Service Layer for business logic │ ├── statement-pool.ts # Object Pool for prepared statements │ └── semantic-search.ts # Semantic search operations ├── strategies/ # Strategy pattern implementations ├── types/ # TypeScript type definitions ├── utils/ # Utility functions └── mcp-server.ts # MCP server data/ ├── patterns/ # JSON files with 555+ pattern definitions └── design-patterns.db # SQLite database with embeddings

🔧 Main Components

Core Services

DatabaseManager: SQLite operations with Object Pool (prevents memory leaks)
StatementPool: LRU-based pool for prepared statements (max 100)
CacheService: In-memory LRU cache with TTL and metrics

Business Logic

PatternService: Service Layer orchestrating pattern operations
PatternRepository: Data access abstraction (Repository Pattern)
SemanticSearchService: Semantic search with embeddings
PatternMatcher: Pattern matching and ranking logic

Integration

PatternHandlerFacade: Facade simplifying MCP handlers
VectorOperationsService: Vector search using sqlite-vec
LLMBridgeService: Interface for language models (optional)
EmbeddingServiceAdapter: Adapter for embedding services

Infrastructure

SimpleContainer: Dependency Injection container
MigrationManager: Database migrations
PatternSeeder: Initial data seeding

🚀 Installation and Setup

Prerequisites

Node.js >= 18.0.0
npm >= 8.0.0 or Bun >= 1.0.0

Installation

# Clone the repository git clone https://github.com/your-org/design-patterns-mcp.git cd design-patterns-mcp # Install dependencies npm install # Configure environment variables (optional) cp .env.example .env # Build the project npm run build # Setup the database npm run db:setup

MCP Configuration

Add to your MCP configuration file (.mcp.json or Claude Desktop config):

{ "mcpServers": { "design-patterns": { "command": "node", "args": ["dist/src/mcp-server.js"], "cwd": "/path/to/design-patterns-mcp", "env": { "LOG_LEVEL": "info", "DATABASE_PATH": "./data/design-patterns.db" } } } }

📖 Usage

Finding Patterns with Natural Language

Use natural language descriptions to find appropriate design patterns through Claude Code:

For object creation problems:

"I need to create complex objects with many optional configurations"
"How can I create different variations of similar objects?"
"What pattern helps with step-by-step object construction?"

For behavioral problems:

"I need to notify multiple components when data changes"
"How to decouple command execution from the invoker?"
"What pattern helps with state-dependent behavior?"

For architectural problems:

"How to structure a microservices communication system?"
"What pattern helps with distributed system resilience?"
"How to implement clean separation between layers?"

For React development:

"How to manage state in React 18/19?"
"What patterns work with React Server Components?"
"How to optimize React performance?"

MCP Tool Functions

find_patterns: Semantic search for patterns using problem descriptions
- Returns ranked recommendations with confidence scores
- Supports category filtering and programming language preferences
search_patterns: Keyword or semantic search with filtering options
- Supports hybrid search (keyword + semantic)
- Filter by category, tags, complexity
get_pattern_details: Get comprehensive information about specific patterns
- Includes code examples in multiple languages
- Shows similar patterns and relationships
- Displays implementations and use cases
count_patterns: Statistics about available patterns by category
- Optional detailed breakdown by category

🛠️ Available Commands

# Development npm run build # Build for production npm run dev # Run in development mode npm start # Start production server # Testing & Quality npm test # Run all tests npm run lint # Check code quality npm run lint:fix # Fix linting issues npm run typecheck # Check TypeScript types # Database npm run db:setup # Complete database setup (migrate + seed + embeddings) npm run migrate # Run database migrations npm run seed # Populate with initial data npm run generate-embeddings # Generate embeddings for semantic search

🎯 Usage Examples

Problem-Based Pattern Discovery

Distributed Systems:

"I need a pattern for handling service failures gracefully" → Circuit Breaker, Bulkhead
"How to implement eventual consistency in distributed data?" → Event Sourcing, CQRS
"What pattern helps with service discovery and load balancing?" → Service Registry, API Gateway

Data Validation:

"I need to validate complex business rules on input data" → Specification Pattern
"How to compose validation rules dynamically?" → Chain of Responsibility
"What pattern separates validation logic from business logic?" → Strategy Pattern

Performance Optimization:

"I need to cache expensive computations efficiently" → Cache-Aside, Write-Through
"How to implement lazy loading for large datasets?" → Lazy Loading, Virtual Proxy
"What pattern helps with connection pooling?" → Object Pool Pattern

Category-Specific Searches

Enterprise Applications:

"Show me enterprise patterns for data access" → Repository, Unit of Work, Data Mapper
"What patterns help with dependency injection?" → DI Container, Service Locator
"How to implement domain-driven design?" → Aggregate, Value Object, Bounded Context

Security Implementation:

"I need authentication and authorization patterns" → RBAC, OAuth 2.0, JWT
"What patterns help with secure data handling?" → Encryption at Rest, Defense in Depth
"How to implement role-based access control?" → RBAC Pattern, Policy-Based Access

🔧 Advanced Configuration

Environment Variables

# Database configuration DATABASE_PATH=./data/design-patterns.db # Logging configuration LOG_LEVEL=info # debug | info | warn | error # LLM integration (optional) ENABLE_LLM=false LLM_PROVIDER=ollama LLM_MODEL=llama3.2 # Performance tuning MAX_CONCURRENT_REQUESTS=10 CACHE_MAX_SIZE=1000 CACHE_TTL=3600000 # 1 hour in ms POOL_MAX_SIZE=100 # Prepared statement pool size

Using the Refactored Server

import { createDesignPatternsServer, TOKENS } from './mcp-server.js'; const server = createDesignPatternsServer({ databasePath: './data/design-patterns.db', logLevel: 'info', enableLLM: false, maxConcurrentRequests: 10, }); await server.initialize(); await server.start(); // Access services via DI Container (for testing) const container = server.getContainer(); const patternService = container.get(TOKENS.PATTERN_SERVICE); const cache = container.get(TOKENS.CACHE_SERVICE);

Performance Monitoring

// Get Object Pool metrics const db = container.get(TOKENS.DATABASE_MANAGER); const poolMetrics = db.getPoolMetrics(); console.log(poolMetrics); // { // size: 87, // hits: 15420, // misses: 234, // evictions: 12, // hitRate: 0.985 // 98.5% // } // Get Cache metrics const cache = container.get(TOKENS.CACHE_SERVICE); const cacheStats = cache.getStats(); console.log(cacheStats); // { // hits: 8765, // misses: 1234, // size: 876, // hitRate: 0.876 // 87.6% // }

📊 Performance and Scalability

Performance Characteristics

Vector Search: Uses sqlite-vec for efficient search in large volumes
Object Pool: Bounded prepared statement cache (max 100) prevents memory leaks
Intelligent Cache: LRU cache with 85%+ hit rate in production
Query Performance: 30-40% faster on repeated queries vs uncached
Optimized Indexes: Specific indexes for different search types
Pagination: Support for large result sets
Metrics: Built-in performance and usage metrics

Benchmarks (from tests)

Database Queries: - COUNT query: 5.03ms - SELECT with LIMIT: 2.08ms - Filtered SELECT: 3.94ms - Concurrent queries (5): 0.95ms total, 0.19ms avg Cache Operations: - Set operation: 0.09ms - Get operation (hit): 0.08ms - Load test (1000 ops): 1.99ms total, 0.002ms avg Pattern Matching: - First query: 1526ms (includes embedding generation) - Subsequent queries: 100-300ms - Cached queries: 0.05ms (2767x speedup) Throughput: - Sustained operations: 13,592 ops/second - Memory usage: Stable at 16-38MB

🧪 Testing

The project includes a comprehensive test suite with 116 passing tests:

Contract Tests: Validate MCP protocol compliance
Integration Tests: Test interaction between components
Performance Tests: Evaluate search and vectorization performance
Unit Tests: Test individual components in isolation

# Run specific test suites npm run test:unit -- --grep "PatternMatcher" npm run test:integration -- --grep "database" npm run test:performance -- --timeout 30000 npm run test:contract # MCP protocol compliance

Test Coverage

MCP Protocol: ✅ 100%
Core Services: ✅ 95%+
Performance: ✅ Comprehensive benchmarks
Database: ✅ Full migration & seeding tests

🏗️ Architecture Patterns Used

This project practices what it preaches by implementing:

Pattern	Location	Purpose
Repository	`repositories/pattern-repository.ts`	Data access abstraction
Service Layer	`services/pattern-service.ts`	Business logic orchestration
Object Pool	`services/statement-pool.ts`	Resource management
Facade	`facades/pattern-handler-facade.ts`	Simplified interface
Dependency Injection	`core/container.ts`	Inversion of control
Strategy	`strategies/search-strategy.ts`	Interchangeable algorithms
Factory	`factories/service-factory.ts`	Object creation
Singleton	Via DI Container	Single instance management
Adapter	`adapters/llm-adapter.ts`	External service integration

🤝 Contributing

We welcome contributions! Here's how:

Fork the project
Create a feature branch (git checkout -b feature/amazing-feature)
Make your changes following our code style
Run tests (npm test) and ensure they pass
Run linting (npm run lint:fix)
Commit your changes (git commit -am 'Add amazing feature')
Push to the branch (git push origin feature/amazing-feature)
Open a Pull Request

Development Guidelines

Follow SOLID principles
Write tests for new features
Update documentation
Use TypeScript strict mode
Follow existing code patterns

📜 License

This project is licensed under the MIT License. See LICENSE for details.

🔗 Useful Links

📞 Support

🐛 Issues: Report bugs through GitHub Issues
💬 Discussions: Join GitHub Discussions
📧 Email: apolosan@protonmail.com
📚 Documentation: Comprehensive architecture and refactoring details available in project documentation

🙏 Acknowledgments

Design patterns from the software engineering community
MCP protocol by Anthropic
SQLite and sqlite-vec for efficient storage and search
Open source contributors

Version: 0.2.1
Last Updated: October 2025
Patterns: 555+
Tests: 116 passing
Performance: 30-40% improvement vs v0.1.x

Deploy Server

HTTP connection URL

security – no known vulnerabilities

license - not tested

quality - confirmed to work

How are these scores calculated?

local-only server

The server can only run on the client's local machine because it depends on local resources.

Tools

Provides intelligent design pattern recommendations using semantic search through a comprehensive catalog of 200+ patterns across 20 categories. Users can describe programming problems in natural language to discover appropriate design patterns with contextual recommendations.

Design Patterns MCP Server