DigestAI news desk

Cut through the AI noise.

Agents & Tools11 min read

Amazon Bedrock AgentCore runtime lets coding agents preview web apps in real time

Amazon Bedrock AgentCore runtime now lets developers preview web apps built by coding agents directly in their browser without deploying them. The Session endpoints feature runs each agent session in an isolated microVM, providing an authenticated URL that forwards HTTP requests, WebSockets, and updates to the ports developers specify. This eliminates the need for manual redeployment, enabling…

1 source primary source

Key points

  • Session endpoints in Amazon Bedrock AgentCore runtime let developers preview web apps in real time via authenticated URLs
  • Developers can watch live updates from coding agents modifying Vite-based websites without manual redeployment
  • Requires configuring allowed ports (e.g., 8090 for Vite) and using a local endpoint with AWS credentials for security

The setup requires configuring allowed ports (e.g., 8090 for the Vite dev server) and using a local endpoint to forward requests with AWS credentials. While the Session endpoint authenticates every request via SigV4 or JWT, the preview page itself remains isolated to localhost. The feature supports HTTP/1.1, HTTP/2, and WebSocket connections, and developers can inspect internal ports (e.g., /status) for debugging. AWS emphasizes security: requests are logged in CloudTrail, and the local endpoint restricts access to prevent unauthorized use. The sample code, available in the agentcore-samples GitHub repo, demonstrates deploying a Node.js agent with a Vite website and testing changes like switching to a dark theme.

Model page: Claude Haiku 5.5 →

Full story from AWS Machine Learning Blog · by Sundar Raghavan primary sourceOpen source ↗

Preview web apps live as your coding agent builds them with Amazon Bedrock AgentCore runtime

AWS Machine Learning Blog · 30 October 2026

Loading the full article…

This text was published by AWS Machine Learning Blog and written by Sundar Raghavan. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗

Topics · follow one to build your own front page

The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.

Comments

via GitHub Discussions

More in Agents & Tools

All →

Related stories