#atom

Official implementation examples of the Model Context Protocol

Core Idea: Reference servers demonstrate standard implementations of MCP, showcasing best practices and providing usable examples for developers to build upon or integrate into their own systems.

Key Elements

Server Categories

1. Official Reference Implementations:

   • Showcase MCP features and SDK usage
   • Maintained by Anthropic
   • Designed for reliability and clarity
   • Demonstrate protocol conformance
   • Serve as templates for community implementations

2. Key Reference Servers

   • Fetch: Web content fetching and conversion for efficient LLM usage
   • Filesystem: Secure file operations with configurable access controls
   • Memory: Knowledge graph-based persistent memory system
   • PostgreSQL/SQLite: Database interaction capabilities with schema inspection
   • GitHub: Repository management, file operations, and GitHub API integration
   • Google Drive: File access and search capabilities for Google Drive
   • Time: Time and timezone conversion capabilities
   • Git: Tools to read, search, and manipulate Git repositories
   • Slack: Channel management and messaging capabilities
   • Google Maps: Location services, directions, and place details

Implementation Patterns

• Standardized Error Handling:

  • Consistent error reporting formats
  • User-friendly error messages
  • Graceful failure handling

• Authentication Management:

  • Environment variable-based credential storage
  • Token management and refresh strategies
  • Permission scope limitation

• Resource Definition Conventions:

  • Document metadata standardization
  • Content chunking and formatting patterns
  • Embedding and search optimization

• Tool Documentation:

  • JSON Schema for parameter definitions
  • Example usage for each tool
  • Edge case handling explanations

Usage in Host Applications

• Direct npx Execution:

  npx -y @modelcontextprotocol/server-memory
  

• Python Execution:

  uvx mcp-server-git
  # or
  pip install mcp-server-git
  python -m mcp_server_git
  

• Client Configuration:

  {
    "mcpServers": {
      "filesystem": {
        "command": "npx",
        "args": ["-y", "@modelcontextprotocol/server-filesystem", "/path/to/allowed/files"]
      },
      "git": {
        "command": "uvx",
        "args": ["mcp-server-git", "--repository", "path/to/git/repo"]
      }
    }
  }
  

Development Patterns

• Tool-first Development: Reference servers focus on exposing functionality as tools first
• Progressive Enhancement: Adding resources and prompts as secondary features
• Security Boundaries: Implementing strict access controls and permission models
• Minimal Dependencies: Keeping dependency chains manageable for reliability
• Comprehensive Documentation: Providing detailed usage examples and configuration options

Connections

• Related Concepts: Model Context Protocol (MCP), MCP Server, MCP Frameworks
• Implementation Examples: Building MCP Servers, MCP Implementation with Python
• Security Aspects: MCP Security Considerations
• Integration Patterns: MCP Client Integration, Tool Orchestration

References

1. MCP Servers Repository: github.com/modelcontextprotocol/servers
2. MCP Server Documentation: modelcontextprotocol.io/servers
3. TypeScript SDK: github.com/modelcontextprotocol/typescript-sdk
4. Python SDK: github.com/modelcontextprotocol/python-sdk
5. Claude Desktop MCP Integration Guide: docs.anthropic.com/claude/docs/claude-desktop-mcp-integration

#MCP #ReferenceImplementation #AITools #ServerImplementation #DevTools #Anthropic #ModelContextProtocol