Overview
You can bring your existing MCP server built for other AI hosts, or build a new MCP server that exposes your service’s capabilities. Alexa+ acts as the MCP client. You return standard MCP responses, and Alexa+ handles natural language understanding, response generation, and user interface (UI) rendering. Alexa+ for Builders supports the 2025-11-25 version of the MCP specification. Alexa+ also supports the MCP Apps extension, enabling you to render interactive UIs for MCP tools directly in the conversation view. This is useful when text-only responses fall short or when data is too complex for native views. Examples include layered map information, multi-step flows, games, immersive learning experiences, or dashboards. Rendering these views inline keeps users engaged within a single, trusted interaction point. You can onboard your existing MCP Apps and use a webview provided by Alexa+ to deliver rich visuals.How it all connects
To make your MCP server available to Alexa+ customers, you create an Alexa+ MCP add-on. This add-on acts as a bridge between your MCP server and Alexa+, handling tool discovery and protocol translation so that Alexa+ can seamlessly understand and invoke your capabilities. Your MCP server defines the tools, resources, and prompts you want to expose. You deploy and configure your Alexa+ add-on using the add-on Agent Skill (agentic onboarding experience) or directly with the Alexa AI CLI. On the Alexa+ side, an add-on registry and discovery layer detects your registered capabilities. The Alexa+ AI reasoning determines when and how to invoke them based on customer intent. The orchestrator coordinates between discovery, reasoning, and the customer to deliver a natural and intelligent experience. At runtime, when a customer makes a request intended for your service, Alexa+ routes the request through your add-on to your MCP server using streamable HTTP. It invokes the appropriate tools and returns the result back to the customer. If you modify tools, configurations, or other MCP server details, redeploy your add-on withalexa-ai deploy. Your new tools and other changes take effect after redeployment.
Architecture components
