In this week's AWS roundup, we dive into exciting new capabilities that empower AI agents to handle payments autonomously, a production-ready toolkit for building on AWS, new EC2 instances with massive performance boosts, and open-source Valkey's impressive growth. Here are answers to the most pressing questions about these updates.

What is the new payment capability in Amazon Bedrock AgentCore?

Amazon Bedrock AgentCore has introduced managed payment capabilities in preview, allowing AI agents to autonomously pay for APIs, MCP servers, web content, and even other agents. This feature, built in partnership with Coinbase and Stripe, removes the heavy lifting of custom billing, credential management, and compliance systems. Developers can connect a Coinbase CDP wallet or Stripe Privy wallet as a payment connection, set session-level spending limits, and let the agent transact autonomously during execution. For example, a research agent could pay for real-time market data on the fly, or a coding agent could call paid APIs mid-task. This opens up possibilities for fully autonomous workflows without manual payment intervention. To get started, visit the AgentCore documentation and the AgentCore CLI.

What is the Agent Toolkit for AWS and how does it help developers?

The Agent Toolkit for AWS is a production-ready suite of tools and guidance, available at no additional charge, that helps AI coding agents build on AWS with fewer errors, lower token costs, and enterprise-grade security controls. It is the successor to the MCP servers, plugins, and skills previously available on AWS Labs. The toolkit provides a quick start guide and a set of skills and plugins hosted on GitHub. Developers can use it to accelerate AI agent development, reduce integration complexity, and ensure compliance. The toolkit covers everything from authentication to payment handling, making it easier to deploy agents in production. For more details, see the quick start guide and browse the available skills on GitHub.

How does the new AWS MCP Server GA simplify AI agent access to AWS services?

The AWS MCP Server is now generally available as a managed remote Model Context Protocol (MCP) server. It gives AI agents and coding assistants secure, authenticated access to all AWS services through a small, fixed set of tools. This is part of the Agent Toolkit for AWS. The GA release ensures that agents can interact with AWS resources (like S3, DynamoDB, Lambda) without needing custom integrations for each service. The server handles authentication and authorization via AWS Identity and Access Management (IAM), providing a consistent interface. This simplifies agent development because developers no longer need to manage individual API calls or credential rotations. For a deeper dive, read Seb Stormacq's blog post.

What can AI agents do with Amazon WorkSpaces in the new preview?

Amazon WorkSpaces for AI agents is now in preview, allowing AI agents to securely access and operate desktop applications through managed WorkSpaces environments. This capability enables organizations to automate everyday workflows at scale while maintaining full enterprise-grade governance and compliance. For example, an agent could log into a WorkSpaces virtual desktop, open an accounting application, process invoices, and generate reports—all without human intervention. The preview ensures that all actions are logged and can be audited, meeting security and compliance requirements. To learn more, read Micah Walter's blog post.

What performance improvements do the new EC2 M8idn/R8idb instances offer?

AWS announced the new EC2 M8idn/M8idb and R8idn/R8idb instances, powered by custom sixth-generation Intel Xeon Scalable processors available only on AWS, along with the latest sixth-generation AWS Nitro cards. These instances deliver up to 43% better compute performance per vCPU compared to previous-generation instances. The M8idn and R8idn variants offer up to 600 Gbps network bandwidth, while the M8idb and R8idb variants provide up to 300 Gbps EBS bandwidth. This makes them ideal for memory-intensive and compute-intensive workloads such as databases, in-memory caches, and real-time analytics. These improvements help reduce costs and latency for high-performance applications. For a full list of announcements, check the What's New with AWS page.

What milestones has Valkey reached on its second anniversary?

Valkey, the open-source key-value store, celebrated its second anniversary with impressive milestones. It stands as proof that open, community-driven technology can innovate faster, scale further, and deliver more value than any single-vendor model. Valkey has surpassed 100 million Docker pulls, a 17x increase year-over-year. It has also attracted more than 225 contributors who have submitted over numerous improvements. The project continues to gain traction as a high-performance alternative for caching, session management, and real-time data needs. These achievements underscore the strength of the open-source community and Valkey's growing adoption in production environments. For the full story, visit the Valkey blog.