AWS Transform custom is an agentic AI service that lets teams define and run their own code transformations in natural language, covering organization-specific migrations like swapping internal libraries, enforcing error-handling conventions, or standardizing logging. It integrates with Kiro IDE (via a Power), any compatible AI coding agent (via an Agent Skill following the open Agent Skills standard), and VS Code/Open VSX (via an IDE plugin). Users describe a transformation in chat, the agent drafts and publishes a reusable Transformation Definition (TD) to their AWS account, then runs it locally against up to 3 repos in parallel or fans out to hundreds of repos using AWS Batch and Fargate in remote mode — all without leaving the editor. Setup involves installing the AWS CLI, configuring credentials, and adding the appropriate surface (power, skill, or plugin).

9m read timeFrom aws.amazon.com
Post cover image
Table of contents
What makes AWS Transform “custom”Three common ways to work with AWS Transform customCreating a custom TD right from chatRunning your custom TDsGetting startedCleanupWrapping up: making your migrations fasterLearn moreAbout the author
52 Impressions