# Strands Agent Chatbot with Amazon Bedrock AgentCore

A reference architecture demonstrating a multi-agent chatbot built with Strands Agents and Amazon Bedrock AgentCore, combining tool execution, memory, and browser automation.

This repository is an AWS sample reference architecture for building agentic chatbots with Strands Agents and Amazon Bedrock AgentCore. It demonstrates a multi-agent chatbot that combines tool execution, persistent memory, browser automation, and agent-to-agent collaboration, intended as an extensible starting point for teams building similar systems on AWS.

It shows AgentCore's managed, containerized Runtime, its Memory service for persistent conversation state and summarization, its Gateway for MCP-based tool integration with JWT authentication, its Code Interpreter for secure Python execution, its Browser service for headless automation with a live view, its Identity service for OAuth delegation, a Registry for cataloging agent skills and MCP servers, and Observability for tracing agent runs. The model picker includes several Bedrock-hosted model families. Deployment is automated with Terraform, and the sample is MIT licensed.

Website: https://github.com/aws-samples/sample-strands-agent-with-agentcore