The Model Context Protocol (MCP) lets AI assistants call tools, read resources, and use prompts from your server. This skill shows you how to build those servers using Node.js and TypeScript. You will learn to register tools, set up resources, and create prompt templates.
You will also choose between a local connection or a remote one. For local use like Claude Desktop, use stdio. For remote clients like Cursor, use Streamable HTTP. The SDK changes over time, so check the official docs or Context7 for the latest method names.
Zod validation helps you check inputs. Follow best practices like schema first and idempotent tools to make reliable servers. This skill guides you through installation, setup, and examples.
Global
mkdir -p ~/.claude/skills/mcp-server-patternsProject
mkdir -p .claude/skills/mcp-server-patternsSource Repository
Find Skillsvercel-labs/skills
Find and install the perfect skill to extend your AI agent
Microsoft Foundrymicrosoft/azure-skills
Build, deploy, and improve AI agents on Microsoft Foundry from start to finish
Azure Aimicrosoft/azure-skills
Search, transcribe, and analyze with Azure AI tools for smarter apps
Azure Hosted Copilot Sdkmicrosoft/azure-skills
Build, deploy, and manage your Copilot SDK apps on Azure with ease
Triagemattpocock/skills
Triage issues with a state machine driven by clear roles and agent briefs
Handoffmattpocock/skills
Hand off your work to another AI agent with a clear summary
Image Editagentspace-so/runcomfy-agent-skills
Smart router picks the best AI model for your image editing needs
Agentspaceagentspace-so/skills
See your AI agent's live folder from any browser instantly