What is Model Context Protocol (MCP)?
Model Context Protocol (MCP) is an open standard created by Anthropic that enables Large Language Models (LLMs) and autonomous AI clients to securely access external tools, databases, resources, and live context.
MCP Development
The open standard that connects AI models to tools, databases, and APIs.
Build your own MCP server from scratch.
Course Curriculum
Seven modules to go from zero to production MCP server.
1. What is MCP?
The universal connector for AI. understand the problem, architecture, and why it matters now.
2. Transports & Message Format
JSON-RPC, stdio, SSE, and Streamable HTTP. how data flows between clients and servers.
3. Building Your First MCP Server
Set up a TypeScript project, register a tool, test with MCP Inspector, connect to Claude Desktop.
4. Tools, Resources & Prompts
The three MCP primitives. when to use each, who controls them, and design best practices.
5. Connecting to Clients
Configure Claude Desktop, Cursor, VS Code, and build your own custom MCP client.
6. Real-World MCP Servers
Build database, API wrapper, multi-tool, and authenticated server patterns.
7. Production, Security & Distribution
Error handling, security hardening, npm publishing, monitoring, and the MCP registry.
Learn More
Canonical Registries (The "npm" Layer)
Community MCP Registries & Toolkits
Skills & Plugin Marketplaces
First-Party Stores & Marketplaces
- OpenAI ChatGPT Apps
- ChatGPT GPTs Store
- OpenAI Plugin & Skills Spec
- Claude Platform Marketplace
- Claude Plugins Directory
- Claude Code Official Plugins
- Anthropic Agent Skills
- Cursor Marketplace
- GitHub Copilot Plugins
- Windsurf Cascade MCP
- Cline MCP Marketplace
- Google Antigravity Plugins
- Gemini CLI Extensions
- xAI Grok Bot Docs
- Grok Plugin Marketplace
- Hermes Skills Hub (Nous Research)
- Zinc (Agent Autonomous Commerce)