Amazon Bedrock AgentCore

Amazon Bedrock AgentCore (Preview)

Deploy and operate AI agents securely at scale - using any framework and model

Get started with Amazon Bedrock AgentCore

What is Amazon Bedrock AgentCore?

Amazon Bedrock AgentCore enables you to deploy and operate highly capable AI agents securely, at scale. It offers infrastructure purpose-built for dynamic agent workloads, powerful tools to enhance agents, and essential controls for real-world deployment. AgentCore services can be used together or independently and work with any framework including CrewAI, LangGraph, LlamaIndex, and Strands Agents, as well as any foundation model in or outside of Amazon Bedrock, giving you ultimate flexibility. AgentCore eliminates the undifferentiated heavy lifting of building specialized agent infrastructure, so you can accelerate agents to production.

Bedrock AgentCore Capabilities

Deploy agents securely at scale

Deploy and scale dynamic AI agents securely using any framework, protocol, or model in or outside of Amazon Bedrock. Benefit from complete session isolation and industry leading support for long-running workloads up to 8 hours, enabling complex, multi-step agent tasks. Manage agent identity and access seamlessly across your services with native integration to your existing identity providers for automated authentication and permission delegation.

Enhance agents with tools and memory

Enable intelligent, personalized experiences with persistent memory that maintains agent knowledge across interactions. Easily connect agents to your existing tools and services with a gateway that transforms them into agent-ready capabilities with minimal code. Enhance your AI agents with powerful built-in tools including a secure browser runtime to help agents execute complex web-based workflows and a code interpreter to securely execute code for tasks such as generating visualizations.

Monitor in production

Gain deep operational insights with real-time visibility into your agents' performance through comprehensive observability dashboards powered by Amazon CloudWatch. Track key metrics such as token usage, latency, session duration, and error rates to support optimal operations. Debug issues, audit agent decisions, and support compliance with visibility into the entire agent workflow. Integrates seamlessly with your existing monitoring systems via OpenTelemetry.

Bedrock AgentCore Benefits

Faster time to value

Accelerate from prototype to production with fully-managed services that eliminate infrastructure complexity, so you can bring groundbreaking agentic solutions to market faster.

Flexibility and choice

Build agents your way using any framework, model, or tool—while maintaining complete control over how your agents operate and integrate with existing systems.

Security and trust

Deploy with confidence using enterprise-grade security, complete session isolation, and comprehensive controls that help your agents operate reliably and securely at scale.

See Bedrock AgentCore in Action

Swami Sivasubramanian, VP of Agentic AI at AWS, explains how AgentCore simplifies the journey of taking AI agents from prototype to production. Watch the team demonstrate how you can leverage AgentCore's modular services to deploy agents using your preferred open-source frameworks and models.

Bedrock AgentCore Services and Tools

Runtime

Deploy agents on secure, serverless infrastructure

Prevent data leakage with complete session isolation. Handle any agent use case with support for low latency real-time iterations and 8-hour asynchronous workloads for long-running tasks.

Learn more

Gateway

Transform existing APIs into agent-ready tools with a few lines of code

Simplifies the process of connecting AI agents to a wide range of tools and services, enhancing their ability to perform diverse tasks.

Learn more

Memory

Maintain context across interactions without managing infrastructure

Provides customers with full control over what the AI agent remembers, delivering context-aware experiences with both short-term and long-term memory, all with zero infrastructure management.

Learn more

Identity

Secure, scalable agent identity and access management

Allow your agents to securely access AWS resources and third-party tools and services on behalf of users or by themselves with pre-authorized user consent.

Learn more

Observability

Monitor agent behavior through intuitive dashboards

Provides operational visibility into AI agents for monitoring, debugging, and supporting compliance. Integrates with preferred dashboards via OpenTelemetry compatibility.

Learn more

Code Interpreter

Execute code securely across multiple languages

Enables AI agents to execute code securely in sandbox environments, enhancing their accuracy and expanding their ability to solve complex end-to-end tasks.

Learn more

Browser tool

Fast, secure, and serverless browser runtime for agents

Enables agents to dynamically interact with web applications. Offers sub-second latency and compute-based session isolation and observability (Live View, Session Replay).

Learn more

Customers

Itaú Unibanco

“Itaú has pioneered using AI/ML for years and already delivered agentic AI to provide customers with a hyper-personalized banking experience, enhanced security, and automation. We are excited about how Amazon Bedrock AgentCore services will expand the possibilities and enable secure and scalable financial services transformation. Amazon Bedrock AgentCore is expected to support our effort to meet customer commitments at scale by giving developers the tools and flexibility to build and manage agents, while ensuring compliance. These agentic ai services will help us deliver an intuitive banking experience with the efficiency of automation and personalization customers expect.“

Carlos Eduardo Mazzei, Chief Technology Officer, Itaú Unibanco

Innovaccer

"AI has massive potential in healthcare, but getting the foundation right is key. That's why we're building HMCP (Healthcare Model Context Protocol) on Amazon Bedrock AgentCore Gateway, which has been a game-changer, automatically converting our existing APIs into MCP-compatible tools and scaling seamlessly as we grow. It gives us the secure, flexible base we need to make sure AI agents can safely and responsibly interact with healthcare data, tools, and workflows. With this partnership, we're accelerating AI innovation with trust, compliance, and real-world impact at the core."

Abhinav Shashank, CEO & Co-founder, Innovaccer

Epsilon

"Marketers need intelligent solutions that deliver personalized customers experiences with speed, accuracy, and scale. With Amazon Bedrock AgentCore, Epsilon plans to accelerate intelligent agent development for dynamic content creation and customer journeys, expecting to cut campaign build times by up to 30% and boost engagement metrics across key verticals."

Prashanth Athota SVP Software Engineering, Epsilon

Box

"In addition to requiring control over how AI agents securely interact with proprietary data, enterprises need the confidence that their AI agents’ results are accurate, transparent, and compliant. Leveraging Strands Agents to build, Box is experimenting with deploying agents with Bedrock AgentCore runtime, allowing our customers to scale AI capabilities to meet enterprise demands, all while maintaining top tier security and compliance standards. Future enterprises will be built on a foundation of Intelligent Content Management powered by agentic AI, and together Box and AWS are leading the way in making this vision a reality."

Ben Kus CTO, Box

Pricing

Tailor AgentCore to your needs—mix and match services, use them independently or together, and pay for what you use as your AI initiatives grow. Learn more on the AgentCore pricing page.