AWS Unveils First Managed Payment System for AI Agents in Landmark Partnership with Coinbase and Stripe

Breaking: Amazon Bedrock AgentCore Now Lets AI Agents Pay for APIs Autonomously

AWS has previewed the first managed payment capabilities for its Bedrock AgentCore platform, allowing AI agents to autonomously access and pay for APIs, MCP servers, web content, and other agents. The service, built in partnership with Coinbase and Stripe, eliminates the need for developers to build custom billing, credential management, and compliance systems.

“This is a game-changer for agent autonomy,” said Sarah Chen, AWS Vice President of AI Services. “Agents can now transact in real time without human intervention, opening up entirely new use cases like research agents that purchase live market data or coding agents that call paid APIs mid-task.”

How It Works

Users connect a Coinbase CDP wallet or a Stripe Privy wallet as a payment connection, set session-level spending limits, and the agent handles transactions autonomously during execution. The system handles billing, credential management, and compliance out of the box.

Background

Amazon Bedrock AgentCore, launched earlier this year, provides a managed runtime for AI agents. Until now, agents could only access free APIs and data sources. The new payment capability unlocks premium content and services, making agents more useful in production environments.

The partnership with Coinbase and Stripe ensures enterprise-grade security and compliance, leveraging existing wallet infrastructure and payment rails.

What This Means

Enterprises can now deploy agents that handle end-to-end workflows requiring paid subscriptions, real-time data feeds, or API calls with usage-based pricing. This reduces the “undifferentiated heavy lifting” of building custom payment integrations and accelerates agent deployment.

“It removes a major barrier to agent adoption,” added Chen. “Companies no longer need to choose between agent functionality and cost control.”

Other Major Updates in This Release

Agent Toolkit for AWS

AWS also released the Agent Toolkit for AWS, a production-ready suite of tools and guidance available at no extra charge. It helps AI coding agents build on AWS with fewer errors, lower token costs, and enterprise-grade security controls. The toolkit replaces the earlier MCP servers, plugins, and skills on AWS Labs. Users can start with the quick start guide or browse skills on GitHub.

AWS MCP Server Now Generally Available

The managed remote Model Context Protocol (MCP) server has reached general availability. It gives AI agents and coding assistants secure, authenticated access to all AWS services through a small, fixed set of tools. See Seb Stormacq’s blog post for details.

Amazon WorkSpaces for AI Agents (Preview)

A new preview lets AI agents securely access and operate desktop applications through managed WorkSpaces environments. Organizations can automate everyday workflows at scale while maintaining full governance and compliance. Read Micah Walter’s blog post for more.

New EC2 Instances: M8idn/M8idb and R8idn/R8idb

Amazon EC2 launched new instances powered by custom sixth-generation Intel Xeon Scalable processors and sixth-generation AWS Nitro cards. They deliver up to 43% better compute performance per vCPU compared to previous gen. M8idn/R8idn instances offer up to 600 Gbps network bandwidth, while M8idb/R8idb instances deliver up to 300 Gbps EBS bandwidth.

Valkey Turns Two

The open-source key-value store Valkey has surpassed 100 million Docker pulls (up 17x year over year) and attracted over 225 contributors. AWS highlights Valkey as proof that community-driven technology can outpace single-vendor models.

For a complete list of AWS announcements, visit the What’s New with AWS page.