An MCP Server is a lightweight program that exposes specific capabilities through the standardized Model Context Protocol. AWS MCP Servers expose a portfolio of servers that give AI applications access to AWS documentation, contextual guidance, and best practices within development, deployment, and cloud-native workflows. They integrate with MCP clients running in IDEs, chatbots, or copilots to maintain 1:1 connections, enabling access to local data sources and remote services to enrich model outputs. The AWS MCP Servers provide access to up-to-date AWS capabilities, docs, What's New posts, and well-architected guidance, bringing cloud-native best practices directly into your development or AI-assisted workflows. They are designed to work with standard stdio transport (with streamable HTTP support in development) and align with MCP specification requirements. Note that Server Sent Events (SSE) support was removed in May 2025 to improve backwards compatibility, and future versions aim to enhance transport options while preserving security and reliability in AWS integrations.