Overview
What is Amazon Bedrock AgentCore?
Amazon Bedrock AgentCore is an agentic platform for building, deploying, and operating highly effective agents securely at scale using any framework and foundation model. With AgentCore, you can enable agents to take actions across tools and data with the right permissions and governance, run agents securely at scale, and monitor agent performance and quality in production - all without any infrastructure management. AgentCore services work together or independently with any open-source framework such as CrewAI, LangGraph, LlamaIndex, and Strands Agents and with any foundation model, so you don’t have to choose between open-source flexibility and enterprise-grade security and reliability.
Core services in Amazon Bedrock AgentCore
Amazon Bedrock AgentCore includes the following modular services and capabilities that you can use together or independently:
| Service | Description | Integrations |
|---|---|---|
| Runtime | A secure, serverless runtime environment purpose-built for deploying and scaling dynamic AI agents and tools. Runtime provides fast cold starts for real-time interactions, extended runtime support for asynchronous agents, true session isolation, built-in identity, and support for multi-modal and multi-agent agentic workloads. |
AgentCore Runtime works with custom frameworks and any open-source framework, including CrewAI, LangGraph, LlamaIndex, Google ADK, OpenAI Agents SDK, and Strands Agents, any foundation model in or outside of Amazon Bedrock including OpenAI, Google's Gemini, Anthropic's Claude, Amazon Nova, Meta Llama, and Mistral models, and popular protocols like MCP and A2A. |
| Memory | A way to build context-aware agents with complete control over what the agent remembers and learns. Supports for both short-term memory for multi-turn conversations and long-term memory that persists across sessions, with the ability to not only share memory stores across agents but also learn from experiences. |
AgentCore Memory works with LangGraph, LangChain, Strands, LlamaIndex |
| Gateway | A secure way to convert your APIs, Lambda functions, and existing services into Model Context Protocol (MCP)-compatible tools and also connect to pre-existing MCP servers, making them available to AI agents through Gateway endpoints with just a few lines of code. |
Any APIs, MCP tools, Lamda, and popular integrations including Salesfore, Zoom, JIRA, Slack etc. |
| Identity | A secure, scalable agent identity, access and authentication management service which is compatible with existing identity providers, eliminating needs for user migration or rebuilding authentication flows. |
Any IdP and credential providers such as Amazon Cognito, Okta, Microsoft Azure Entra ID, Auth0 etc. |
| Code Interpreter | An isolated sandbox environment for agents to execute code enhancing their accuracy and expanding their ability to solve complex end-to-end tasks. |
Multiple languages including Python, JavaScript and TypeScript |
| Browser | A fast and secure cloud-based browser runtime environment to enable AI agents to interact with web applications, fill forms, navigate websites, and extract information in a fully managed environment. |
Any foundation model or popular browser automation frameworks including Playwright and BrowserUse |
| Observability | A unified view to trace, debug and monitor agent performance in production. It offers detailed visualizations of each step in the agent workflow, enabling you to inspect an agent's execution path, audit intermediate outputs, and debug performance bottlenecks and failures. |
Any monitoring and observability stack that integrate with telemetry data emitted in standardized OpenTelemetry (OTEL)-compatible format |
| Evaluations | An evaluation system that continuously inspects agent quality based on real-world behavior. With AgentCore Evaluations, you can either use thirteen built-in evaluators for common quality dimensions or use custom evaluators for specific business requirements. |
All results integrated into AgentCore Observability powered by Amazon CloudWatch for unified monitoring. |
| Policy | A capability that provides deterministic control to ensure agents operate within defined boundaries and business rules without slowing them down. Easily author fine-grained rules using natural language or Cedar |
Integrates with AgentCore Gateway, to intercept every tool call before execution. You can define which tools agents can access, what actions they can perform, and under what conditions. |
What can you build with Amazon Bedrock AgentCore?
With Amazon Bedrock AgentCore, developers can accelerate AI agents into production with the scale, reliability, and security, critical to real-world deployment. Some common use cases for which you must consider leveraging AgentCore are:
-
Agents
Build autonomous AI apps that reason, use tools, and maintain context. Deploy agents for customer support, workflow automation, data analysis, or coding assistance. Your agent runs serverless with isolated sessions, persistent memory, and built-in observability.
-
Tools and Model Context Protocol (MCP) Servers
Transform existing APIs, databases, or services into tools that any MCP-compatible agent can use. Deploy a gateway that wraps your Lambda functions or OpenAPI specs making your backend instantly accessible to agents without rewriting code.
-
Agent Platforms
Provide your internal developers or customers with a paved path to build and deploy agents using approved tools, shared memory stores, and governed access to enterprise services. Centralize observability, authentication, and compliance while enabling teams to ship agent-powered features faster.
Pricing for Amazon Bedrock AgentCore
AgentCore offers flexible, consumption-based pricing with no upfront commitments or minimum fees. For more information, see AgentCore pricing
Next Steps
If you are a first-time user of Amazon Bedrock AgentCore, we recommend that you begin by reading the following sections:
