Amazon Bedrock AgentCore adds new features to help developers build agents faster
Amazon Web Services has expanded its Bedrock AgentCore platform with new features designed to accelerate AI agent development and deployment. The update introduces a managed harness in preview that allows developers to define agents by simply specifying a model, system prompt, and tools, then run them immediately without writing orchestration code. The harness handles the complete agent lifecycle including reasoning, tool selection, action execution, and response streaming, with each session running in its own microVM environment with filesystem and shell access. The new capabilities include filesystem persistence that enables agents to suspend mid-task and resume from exactly where they stopped, along with the ability to switch models during a session. Developers can experiment with different configurations without redeployment and export the harness orchestration as Strands-based code when full control is needed. The platform also debuts the AgentCore CLI for infrastructure-as-code deployments with AWS CDK support today and Terraform integration coming soon. Additionally, AWS introduced AgentCore skills for coding assistants, starting with availability through Kiro Power and expanding to Claude Code, Codex, and Cursor next week. The managed harness preview is currently available in four AWS regions: US West (Oregon), US East (N. Virginia), Europe (Frankfurt), and Asia Pacific (Sydney).
Why It Matters
This release addresses a key friction point in AI agent development by providing a low-code approach to prototyping and a clear path to production deployment. The managed harness significantly lowers the technical barrier for developers to experiment with AI agents, while the export capabilities ensure teams aren't locked into a simplified framework as their needs evolve. The integration with popular coding assistants and infrastructure-as-code tools positions AgentCore as a comprehensive platform for enterprise AI agent workflows, potentially accelerating adoption of autonomous AI systems in business applications.
This summary is generated using AI analysis of the original press release. Always refer to the original source for complete details.