Scaling cloud migrations with agentic AI on Amazon Bedrock AgentCore
AWS published a post describing a four-agent pattern for cloud migrations on Amazon Bedrock AgentCore, built by AWS Professional Services and using the Strands Agents SDK. The post says the pattern reduced infrastructure as code (IaC) development time from 3 to 4 weeks per application to minutes, based on internal project tracking data, on one enterprise program spanning more than 300 applications with a fixed fiscal year deadline.
The pattern is positioned as running alongside AWS Transform, not replacing it. According to the post, AWS Transform covers migration and modernization for server, network, mainframe, .NET, and application code workloads, while AWS Database Migration Service covers the database tier. The custom agents attach to those services and require sources and destinations reached through Model Context Protocol (MCP) tools the organization builds and maintains.
The four agents are the Intake Agent, which reads migration inputs from document and collaboration systems through MCP tools; the IaC Agent, which generates IaC composing approved internal modules; the Migration Intelligence and Governance Agent, which reports and governs inside program tools including Jira, Confluence, and Webex; and a Site Reliability Engineering Agent for operations after cutover. The post groups them into a migration journey and an operations journey.
Each agent is defined by a foundation model, a system prompt, and tools. AgentCore runtime hosts them serverlessly with session isolation and multi-agent orchestration. Agents call MCP tools through AgentCore Gateway; AgentCore Identity authenticates calls via scoped IAM roles and the identity provider; AgentCore memory stores session state and shared context across the applications.
The post states the program had three conditions: MCP-connected sources and destinations including an internal wiki holding security standards, a ticketing system, a collaboration platform, and an in-house provisioning API; organization-specific IaC composition against a security-office-reviewed internal module library; and work continuing past cutover.
It also describes a curated policy set, served by an AWS Lambda function and exposed as an MCP tool named get_policies, with versioned rules and wave exceptions carrying expiry dates. Policy in AgentCore evaluates Cedar rules before tool calls, while the curated set determines what generated infrastructure satisfies.
Based on reporting from the original publisher. Visit the source for full context and later updates.
Publisher excerpt
Learn how AWS Professional Services uses a multi-agent framework built on Amazon Bedrock AgentCore to automate enterprise cloud migrations end to end. Purpose-built AI agents handle discovery, infrastructure as code generation, portfolio governance, and post-migration operations, reducing IaC development time from weeks to minutes.