aws-transform

Executes code upgrades, SDK migrations, and framework transformations via the AWS Transform CLI.

Updated Jul 1, 2026
One-click install
npx skills add https://github.com/sakicodes/BuildFestHackathon26 --skill aws-transform-sakicodes
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: aws-transform
Source: https://github.com/sakicodes/BuildFestHackathon26/tree/main/.agents/skills/aws-transform
Command: npx skills add https://github.com/sakicodes/BuildFestHackathon26 --skill aws-transform-sakicodes

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Modernizing codebases — upgrading language versions, migrating AWS SDKs, and migrating frameworks — is slow, repetitive work that scales poorly across dozens or hundreds of repositories. This Skill automates the full transformation workflow using the AWS Transform (ATX) CLI, from repository inspection to execution and result reporting. ## Core Features & Use Cases - Automated TD Matching: Inspects repositories for signals (Java/Python/Node.js versions, boto2, AWS SDK v1/v2 usage, x86 configs) and matches them against Transformation Definitions discovered via atx custom def list --json. - Dual Execution Modes: Runs transformations locally for 1-9 repos (max 3 concurrent) or remotely at scale via AWS Batch/Fargate for up to 512 repos with auto-deployed infrastructure. - End-to-End Workflow: Handles prerequisite checks (AWS CLI, credentials, IAM permissions, CDK), private repo credential setup via Secrets Manager, runtime version verification, parallel execution monitoring, and consolidated result summaries. - Use Case: A team needs to migrate 50 Java services from AWS SDK v1 to v2. The Skill inspects each repo, presents a match report, deploys remote infrastructure with user consent, submits batch jobs in chunks of 128, and delivers a transformation summary with a CloudWatch dashboard link. ## Quick Start Ask the assistant to upgrade your Java project to Java 21 or migrate your Python code from boto2 to boto3 using AWS Transform, and provide the repository path or git URL when prompted.

Frequently Asked Questions about aws-transform

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I migrate AWS SDK for Java from v1 to v2 across multiple repositories?

Provide the repository paths or git URLs, and the Skill inspects each repo for com.amazonaws imports or aws-java-sdk dependencies, matches them to the appropriate Transformation Definition, and executes the migration locally or at scale via AWS Batch after your confirmation.

How do I upgrade Python or Node.js versions in my codebase automatically?

The Skill detects current versions from files like pyproject.toml, .python-version, package.json, or .nvmrc, then matches a version-upgrade Transformation Definition. It verifies the target runtime is installed locally or configures the remote container before executing.

Can AWS Transform run on Windows?

Native Windows is not supported. You must install Windows Subsystem for Linux (WSL) with `wsl --install` in PowerShell as Administrator, restart, and run the Skill from within a WSL terminal.

What AWS permissions does the ATX CLI need?

Local mode requires the transform-custom:* permission, which can be granted by attaching the AWSTransformCustomFullAccess managed policy. Remote mode additionally needs Lambda invoke, S3, KMS, Secrets Manager, and CloudWatch permissions, generated via the included policy generator script.

How do private GitHub repositories work with remote transformations?

For HTTPS URLs, store a GitHub PAT in Secrets Manager as atx/github-token; for SSH URLs, store an SSH private key as atx/ssh-key. The remote container fetches these credentials at startup to clone your repos, and you can delete the secrets anytime.

What are the limits for running transformations at scale?

Local mode supports up to 9 repos with a maximum of 3 running concurrently. Remote mode supports up to 512 repositories per session, submitted in chunks of up to 128 jobs per Lambda call, with AWS Batch running jobs concurrently.