Language:JapaneseEnglish

migration-to-aws Claude Code Plugin

Author
Amazon Web Services
Category
Development
Topics
Migration & Modernization · Cloud, Deployment & CI/CD · AI Agents & AI App Development
First cataloged
2026-07-09 (UTC)
Explanation last updated
2026-08-19 (UTC)
Source (GitHub) last updated
2026-08-26 (UTC) (yesterday)

The explanation below is AI-generated. Please verify it against the sources.

This plugin equips AI coding agents (Claude Code, Codex, Cursor) to plan and generate a migration from Google Cloud Platform or Heroku to AWS, including moving OpenAI/Gemini AI workloads to Amazon Bedrock. According to the README, it runs a six-phase process that discovers existing resources from Terraform files, application code, billing data, or a read-only CLI inventory, then designs an AWS architecture, estimates costs across pricing tiers, and produces artifacts such as Terraform code, migration scripts, and an HTML assessment report. It also bundles three additional skills: a read-only Terraform security policy checker, a tool that rewrites AI SDK calls to target Bedrock, and an advisor that recommends where to run AI agents on AWS. The item description states that all processing happens locally, so source data stays within the user's own environment. Overall it functions as an assessment-and-code-generation aid rather than an automated migration executor that changes live infrastructure on its own.

About the service

The tool bridges three cloud platforms rather than being a SaaS product itself: per the README, it reads resource and billing information from Google Cloud Platform and Heroku accounts through their official command-line tools (with read-only, consent-gated access), and it maps that information onto Amazon Web Services resources and Amazon Bedrock for AI workloads, using live AWS pricing data during cost estimation.

What you can do with migration-to-aws

  • Discover existing GCP or Heroku resources from Terraform files, application code, billing exports, or live read-only CLI inventory
  • Get a proposed AWS architecture mapping (e.g., compute, database, and networking equivalents) for the discovered resources
  • Receive AWS cost estimates across multiple pricing tiers compared against current spend
  • Generate Terraform configuration, numbered migration scripts, and a self-contained HTML migration report
  • Map and migrate AI/agentic workloads (models, agents, tools) from OpenAI or Gemini to Amazon Bedrock, including code rewrites
  • Run a read-only policy check on generated Terraform for basic security misconfigurations
  • Get guidance on selecting an AWS runtime (such as AgentCore, ECS, EKS, or Lambda) for deploying AI agents

Sources

Back to list