AWS releases Claude Code remote execution sample
Understand this story
AWS released a tutorial and sample, sample-remote-coding-agent, showing how to run Claude Code as a remote coding agent on Amazon Bedrock AgentCore: each task runs in an isolated microVM session and returns a GitHub pull request when complete. Usage requires an AWS account (preferably dedicated) in a region offering AgentCore runtime V2, including US N. Virginia, Ohio, Oregon, Asia Pacific Tokyo and Europe Ireland. Deployment takes about 15–25 minutes, and registering a code repository takes another 8–12 minutes. AWS notes that an idle deployment still incurs costs through NAT gateway, interface endpoints, AWS WAF, Cognito user pool, DNS Firewall, images and logs; the deployment sets a default monthly Amazon Bedrock budget of $200 that only sends email alerts and does not cap actual spending.
AI overview based on source reports · Updated 15 min ago
Coverage timeline
Follow the coverage to explore different perspectives.
- AWS Machine Learning BlogExtend Claude Code to remote sessions on Amazon Bedrock AgentCore with security controls
This post explains how to run Claude Code as a remote agent on Amazon Bedrock AgentCore runtime, where each task executes in its own isolated microVM session inside your AWS account and returns a GitHub pull request for human review. It walks through the accompanying sample-remote-coding-agent, covering the architecture (API Gateway, Lambda, DynamoDB, SQS, AgentCore runtime) and security controls including Bedrock Guardrails prompt screening, short-lived GitHub tokens via AgentCore Identity, private VPC networking, and least-privilege IAM. The sample suits scoped features, bug fixes, and tests on trusted repositories, but not untrusted code or open-ended work, and the agent never merges pull requests.
Activity for this story
Not enough continuous observations to show a trend yet.