Claude Platform on AWS is Now Generally Available — Native Anthropic Experience Accessible via AWS Account

AWS has announced the general availability of Claude Platform on AWS, a new service that allows developers to directly access Anthropic’s native Claude Platform experience through their AWS account.

Previously, using the Claude API required a separate contract and authentication credentials with Anthropic. With Claude Platform on AWS, developers can leverage their existing AWS account, billing, and IAM policies. AWS is the first cloud provider to offer a native Claude Platform experience.

(Source: AWS Machine Learning Blog)

Available Features and APIs

Claude Platform on AWS provides the same APIs, features, and console experience as those directly offered by Anthropic. Specifically, the following features are included:

  • Messages API
  • Claude Managed Agents (beta)
  • Advisor tool (beta)
  • Web search and web fetch
  • MCP connector (beta)
  • Agent Skills (beta)
  • Code execution
  • Files API (beta)

These features enable developers to build more advanced applications that utilize agent functionality and tool integration, in addition to traditional text generation.

(Source: AWS Machine Learning Blog)

Setup and Authentication Methods

Getting started with Claude Platform on AWS involves three steps. First, enable your account on AWS Marketplace. Then, create a workspace to separate projects and environments, and set up access control using IAM policies.

Two authentication methods are provided: IAM authentication using AWS Signature Version 4 and traditional API key authentication. Using IAM authentication allows for fine-grained access control using the workspace ARN in IAM policies.

Claude Platform on AWS is operated by Anthropic, and requests and data are processed outside the AWS security boundary. This makes it suitable for teams without specific regional data residency requirements.

(Source: AWS Machine Learning Blog)

Comparison with Claude Models on Amazon Bedrock

Claude Platform on AWS complements Claude models on Amazon Bedrock. Amazon Bedrock is designed for enterprises with regional data residency or stricter security requirements.

On the other hand, Claude Platform on AWS is suitable for developers who want to utilize Anthropic’s full feature set, including agent functionality, tool integration, and MCP connectors. If data processing outside the AWS security boundary is acceptable, developers can access a richer set of features.

Developers can choose the approach that best fits their project requirements.

(Source: AWS Machine Learning Blog)

Summary

  • Through Claude Platform on AWS, developers can access Anthropic’s full feature set using their existing AWS account, without the need for separate contracts or authentication credentials, and build applications that leverage Claude Managed Agents and MCP connectors.
  • By combining IAM authentication and workspace ARNs, enterprises can implement fine-grained access control for multiple teams and environments, creating a secure AI development environment.
  • Understanding the characteristics of Amazon Bedrock (regional data residency-focused) and Claude Platform on AWS (feature-rich) allows developers to choose the optimal Claude utilization strategy for their project requirements.