aws/bedrock-agentcore-starter-toolkit

Python CLI toolkit for Amazon Bedrock AgentCore (legacy). For new projects, use the AgentCore CLI: https://github.com/aws/agentcore-cli

What it solves

It simplifies the process of building, deploying, and operating production-ready AI agents on Amazon Bedrock AgentCore. It provides a set of tools to help developers move agents from development to a secure, scalable, and reliable production environment without having to manage the underlying infrastructure.

How it works

This toolkit acts as a starting point for developers to create projects using various agent SDK frameworks (such as Strands, LangGraph, or LangChain) and model providers. It integrates with a suite of modular Bedrock AgentCore services, including:

  • Runtime: A serverless environment for deploying agents and tools.
  • Memory: Managed short-term and long-term memory for context-aware conversations.
  • Gateway: A managed Model Context Protocol (MCP) server that converts APIs and Lambda functions into tools.
  • Code Interpreter: A secure sandbox for executing code.
  • Browser: A cloud-based browser for interacting with websites.
  • Observability & Evaluation: Tools for tracing, debugging, and measuring agent performance metrics like correctness and goal success.
  • Identity & Policy: Systems for secure access management and deterministic control over agent actions using the Cedar policy language.

Who it’s for

Developers who want to build AI agents using open-source frameworks and deploy them to Amazon Bedrock AgentCore, specifically those who prefer a Python-based workflow (though the project now recommends the AgentCore CLI for new projects).

Highlights

  • Framework Agnostic: Works with popular frameworks like LangGraph, CrewAI, and Strands.
  • Serverless Scaling: Handles infrastructure and security via the AgentCore Runtime.
  • MCP Integration: Uses the Gateway service to scale agents to use hundreds or thousands of tools.
  • Enterprise Security: Includes built-in identity management and a secure code execution sandbox.

Related

  • Project
    awslabs/agentcore-samples

    A collection of samples and tutorials for Amazon Bedrock AgentCore, providing the infrastructure to deploy and operate framework- and model-agnostic AI agents securely at scale on AWS.

  • Project
    aws/bedrock-agentcore-sdk-python

    A Python SDK for deploying local AI agents to Amazon Bedrock AgentCore, providing serverless infrastructure, enterprise security, and integrated tools like memory and code execution.

  • Project
    awslabs/amazon-bedrock-agent-samples

    A collection of reference implementations and best practices for building single and multi-agent AI assistants using Amazon Bedrock Agents.

  • Project
    langchain-ai/langchain-aws

    A collection of LangChain and LangGraph components that integrate AWS services like Bedrock, SageMaker, and DynamoDB for building AI agents and RAG applications.

  • Project
    awslabs/fullstack-solution-template-for-agentcore

    A full-stack starter template for AgentCore that provides a secured React frontend and AWS infrastructure to accelerate the deployment of AI agent applications.