Claude Code plugins and agent skills for AWS development — IaC(CDK/SST), serverless, cost ops, and Bedrock AgentCore

Curated archive of agentic AI samples, reference implementations, and tooling from the Agentic AI Acceleration program in Korea. AWS-native patterns covering Bedrock, AgentCore, and end-to-end agentic AI pipelines.

Agentic AI Robot: Industrial Safety Monitoring and Control

This sample demonstrates building a logistics tracking agent using Strands Agents SDK and an OpenAI model hosted on Amazon Bedrock AgentCore Runtime with VPC connectivity. The implementation shows how to create an intelligent agent that uses local tools to query a private Amazon RDS database running in an Amazon VPC to track shipments and delays.

Hands-on workshop: Build AI agents that automate browser tasks using Amazon Bedrock AgentCore and Browser-Use SDK with Live View.

Build and deploy a multi-agent AI system on AWS using Bedrock AgentCore, 4 agents collaborate to research, analyze, and write reports

Production deployment of Deep Research Agents to AWS Bedrock AgentCore using Terraform IaC, with LangFuse integration for comprehensive observability and monitoring.

Internal agent platform on Amazon Bedrock AgentCore: interactive Claude Code cloud workspaces + headless agent kernels behind one portal, with publishing, scheduling, channels, memory, evals and governance

Deploy your own personal OpenClaw on AWS Bedrock AgentCore — serverless, ~$9/mo, one-click CloudFormation, Discord/WhatsApp/Telegram 🦞

Universal three-layer evaluation SDK for AI agents, with a reference deployment on Amazon Bedrock AgentCore and Strands.

Multi-agent SRE on-call investigator that auto-triages Slack/Discord infrastructure alerts via AWS Bedrock AgentCore, fanning out to specialized agents (CloudWatch, EKS, Slack/Discord scanners) for parallel investigation.

AI-powered AWS cost analysis and optimization agent using natural language — built with Amazon Bedrock AgentCore and Strands Agents SDK

Enterprise identity on Amazon Bedrock AgentCore with Lark (Feishu) as the IdP. Per-user Lark tokens live in the AgentCore Identity Token Vault (OAuth 3LO) — stored, refreshed, and injected natively, no custom interceptor. Downstream MCP tools act as the user against Lark, reaching only what that user can.

Production SecOps agents as configuration on Amazon Bedrock AgentCore Harness — three-layer reference (strategy / simulation / foundation), live-validated Layer 1.

Configuration-driven, provider-agnostic runtime for declaring AI agents in YAML and deploying them on AWS — built on the Strands Agents SDK and Amazon Bedrock AgentCore.

A Slack frontend for AI agents on Amazon Bedrock AgentCore.

Enterprise identity on Amazon Bedrock AgentCore with Lark (Feishu) as the IdP. A Gateway Request Interceptor forwards the user's identity and injects a per-user credential — the agent never holds it — so downstream MCP tools act as the user against Lark and reach only what that user can.

AI co-pilot for insurance advisors built on Amazon Bedrock AgentCore — text + voice agents, MCP tools, and coverage-aware customer insight.

Production security-operations agents built as configuration on Amazon Bedrock AgentCore — a harness for defining, testing, and running SecOps agents.