Urgent.News

What's breaking now, across thousands of outlets.

AI

Build interactive MCP Apps using Amazon Bedrock AgentCore

Learn how to build and deploy an MCP App with interactive HTML widgets on Amazon Bedrock AgentCore. Because MCP Apps is a host-agnostic standard, the same server delivers the same rich experience across AI hosts like ChatGPT and Claude that support the extension.

As businesses increasingly engage with customers through AI hosts like ChatGPT and Claude, they require a solution to make their services accessible across various applications featuring rich user interfaces, not just plain text. To address this need, MCP Apps and Amazon Bedrock AgentCore present a unified approach. MC P Apps expands the Model Context Protocol (MCP) with interactive HTML widgets directly within AI hosts, while Amazon Bedrock AgentCore offers a scalable platform for building, connecting, and optimizing agents at any framework or model.

For MCP Apps specifically, AgentCore's runtime provides a secure, serverless, session-isolated environment with native MCP support. AgentCore Gateway, another capability of AgentCore, exposes this runtime through a single secure endpoint that MCP Apps-compatible hosts can access. By handling the underlying complexities, AgentCore allows developers to concentrate on business logic and widget design.

This article demonstrates how to build and deploy an MCP App with interactive HTML widgets on Amazon Bedrock AgentCore through the example of Unicorn Rentals, an application that enables users to browse available unicorns, book rentals, view active bookings, and return them. Regardless of the AI host used, such as ChatGPT or Claude, the experience remains consistent.

The deployment process involves several steps, from listing available unicorns to returning a unicorn, all facilitated by a single MCP server running on Amazon Bedrock AgentCore runtime and protected by an AgentCore Gateway.

Written by urgent.news from AWS Machine Learning's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at aws.amazon.com →

More in AI

More from Friday 11 September →