Based solely on the provided article, Amazon Bedrock AgentCore harness handles the orchestration loop, tool routing, memory management, and compute environment for building an AI agent. The post walks through building a serverless image editor where users upload a photo, describe an edit in plain English, and receive the result in seconds. The agent runs on AgentCore harness without custom orchestration code. The full solution, including authentication, encrypted storage, three image editing tools, and a React frontend, is deployed with a single deployment command. The infrastructure is defined using AWS Cloud Development Kit (AWS CDK). The application accepts prompts like “change the car color to blue” or “extend the image 200 pixels to the right.” An agent powered by Claude Sonnet 4.6 breaks the requirement into a series of steps and orchestrates tool calling, each associated with a different Stability AI model. AgentCore memory stores conversation history in the AgentCore service for 30 days. The architecture has four layers: a React frontend hosted on AWS Amplify, an AWS Lambda proxy, and three AWS Lambda-backed tools exposed through Model Context Protocol (MCP). After each edit, a Python script runs directly on the AgentCore runtime microVM to add a watermark.