What problem does it solve? Starting a new agent project on AWS involves many decisions — framework choice, project scaffolding, local development, and first deployment. This Skill walks developers through the entire journey from "I want to build an agent" to a running, deployed agent on Amazon Bedrock AgentCore, eliminating guesswork about CLI flags, naming rules, and framework trade-offs. ## Core Features & Use Cases - Framework Selection Guidance: Compares Strands, LangGraph, Google ADK, and OpenAI Agents with decision criteria, plus a BYO Container path for unsupported frameworks and languages like Java, TypeScript, or Go. - Project Scaffolding: Constructs and validates the agentcore create command, including project name rules (max 23 alphanumeric characters), protocol, build type, model provider, and memory options. - Local Dev and First Deploy: Explains agentcore dev port behavior, local limitations (no memory or gateway URLs), and walks through agentcore deploy and agentcore invoke. - Use Case: A developer says "build a customer support agent." The Skill recommends Strands with Bedrock, validates the project name, presents the create command for confirmation, explains the generated main.py structure, and guides the first deploy. ## Quick Start Ask the assistant to help you create a new AgentCore agent project, optionally naming a framework like Strands or LangGraph.
