AWS runs Claude Code as a remote agent on Bedrock AgentCore: one microVM per task, up to three attempts, and a draft pull request unless verification passes
A sample architecture for handing a coding task to Claude Code in your own AWS account, closing the laptop and coming back to a pull request, with the security controls spelled out step by step.
// Key points
- Each registered repository gets its own AgentCore runtime; every task runs in its own microVM inside a VPC with private endpoints to Amazon Bedrock, and the container holds no other AWS credentials
- Claude Code plans with Claude Opus, writes the change with Claude Sonnet, runs the repository's own verification, then reviews with Opus and can send the task back to planning, up to three attempts
- The agent never holds a long-lived GitHub credential: AgentCore Identity issues a user-to-server token that expires after 8 hours; task prompts are screened by a Bedrock Guardrails guardrail first
- The pull request records the verification command, exit code and attempts used, opens as a draft unless verification passed, and the agent never merges
Builder's takePlan with the big model, write with the cheaper one, review with the big one again: that split is worth copying even if you never touch AgentCore. For PandaClaws' backend I'd adopt the two hard rules first, short-lived repo tokens and "draft unless verified", before worrying about where the sandbox runs.
// Background · from #Coding agents
Full timeline →- Nov 7 Foundry Local runs Qwen3.8-27B at 4-bit in about 18 GB, on 24 GB consumer GPUs with up to 64K context, as a GitHub Copilot model provider
- Nov 7 The Copilot usage metrics API now reports how many Copilot code review comments developers acted on, and how many reviewed PRs merged with no later human change request
- Nov 6 Microsoft Research's PDD Kit turns intent into a Verus spec and proves agent-written code against it, verifying a Rust storefront's login logic