Claude Code and Amazon Bedrock in AWS GovCloud
Anthropic Claude Opus 5.5 and Claude Sonnet 5.5 are now accessible on Amazon Bedrock within AWS GovCloud (US) Regions, bringing agentic coding capabilities to regulated workloads.

Bringing Advanced AI Models to Regulated Workloads
The introduction of Anthropic Claude Opus 5.5 and Claude Sonnet 5.5 in the AWS GovCloud (US) Regions delivers a robust on-ramp for AI-assisted development tailored for workloads carrying strict regulatory or compliance requirements, including International Traffic in Arms Regulations (ITAR). Organizations managing sensitive operations can now leverage these advanced models through <a href="https://aws.amazon.com/bedrock">Amazon Bedrock</a> to accelerate daily coding tasks while upholding compliance standards.
According to the official announcement, Claude Sonnet 5 holds FedRAMP Class D certification alongside Department of Defense Impact Level 4 and 5 (IL4/IL5) authorization. Meanwhile, Claude Opus 5.5 and Claude Sonnet 5.5 have also achieved FedRAMP Class D certification on <a href="https://aws.amazon.com/bedrock">Amazon Bedrock</a>, ensuring that government agencies and defense contractors can deploy generative AI capabilities safely within authorized frameworks.
Security Features and Endpoint Architectures
Built specifically for US customers navigating elevated compliance needs, <a href="https://aws.amazon.com/govcloud-us/">AWS GovCloud (US) Regions</a> provide secure environments backed by rigorous data protection mechanisms. Customer content remains uncompromised, as it is never stored, logged, used for training AWS models, or shared with third parties.
To accommodate diverse architectural designs, <a href="https://aws.amazon.com/bedrock">Amazon Bedrock</a> in AWS GovCloud (US) supports two separate endpoint surfaces: bedrock-runtime and bedrock-mantle. Both options are powered by the underlying Mantle inference engine featuring a Zero Operator Access architecture. The bedrock-runtime endpoint utilizes the AWS SDK and supports comprehensive audit trails through invocation logging, whereas bedrock-mantle natively supports the Anthropic Messages API and unlocks capabilities like server-side tools and background inference.
Empowering Developers with Claude Code
Engineers working within these secure environments can pair the cloud infrastructure with <a href="https://code.claude.com">Claude Code</a>, Anthropic's agentic coding assistant. Operating directly within developer terminals, integrated development environments like VS Code and JetBrains, and background pipelines via the Claude Agent SDK, the tool reads codebases, edits files, and executes commands.
The agentic tool is capable of writing code across multiple files, fixing bugs, resolving Git merge conflicts, and answering intricate architectural questions. Furthermore, developers can extend its functionality by connecting to external data sources using the Model Context Protocol, allowing seamless integration with tools like the AWS CLI, Terraform, and Kubernetes.
Getting Started and Configuration Guidelines
Implementing this setup requires an <a href="https://aws.amazon.com/govcloud-us/">AWS GovCloud (US) Regions</a> account with proper Identity and Access Management roles configured for <a href="https://aws.amazon.com/bedrock">Amazon Bedrock</a>. Teams must verify that model access for Claude Opus 5.5, Claude Sonnet 5.5, and Claude Sonnet 5 is explicitly enabled within their regional environment.
Once the AWS CLI is configured with valid session credentials, developers can install <a href="https://code.claude.com">Claude Code</a> and navigate to their local code repositories. During the interactive login process, users select the third-party platform option followed by <a href="https://aws.amazon.com/bedrock">Amazon Bedrock</a>, utilizing the setup wizard to specify the us-gov-west-1 region and pin the necessary model configurations.
Sources
- AWS Machine Learning BlogSupercharge regulated workloads with Claude Code and Amazon Bedrock
Continue chronologically




