Easytrieve Modernization

Transform mainframe Broadcom Easytrieve (EZT) programs to Java 17 (Spring Boot 3.x) using AWS Transform custom.

---

Two Solution Paths

Path	Best for	How
Kiro Power (recommended)	End-to-end automation, team use	Install this repo as a Kiro Power — Kiro orchestrates everything
Manual (CDK/CLI)	Learning, customization, debugging	Deploy EC2 via CDK, run atx CLI commands manually

---

Path 1: Kiro Power (recommended)

Kiro handles everything: workspace setup, BRE extraction, Java transformation, byte-by-byte validation — all running serverless on AWS Batch.

Step 1: Install Kiro

Download from kiro.dev. Sign in with your AWS Builder ID.

Step 2: Install the aws-transform Power

1. Open Kiro → Powers panel (left sidebar)
2. Search for "aws-transform" → Click Install

Step 3: Install this Power

Point Kiro at this GitHub repo:

https://github.com/aws-samples/sample-mainframe-easytrieve-transform

Or install manually — clone and run:

git clone https://github.com/aws-samples/sample-mainframe-easytrieve-transform
cd sample-mainframe-easytrieve-transform/mcp-server
npm install

Then tell Kiro to use it, or copy to your powers directory:

Platform	Copy to
Windows	%LOCALAPPDATA%\Kiro\powers\sample-mainframe-easytrieve-transform\
macOS	~/Library/Application Support/Kiro/powers/sample-mainframe-easytrieve-transform/
Linux	~/.local/share/Kiro/powers/sample-mainframe-easytrieve-transform/

Restart Kiro. You should see "Easytrieve Modernization" in the Powers panel.

Step 4: Configure AWS Credentials

aws configure
# Region: us-east-1
# Requires: AWSTransformCustomFullAccess policy
aws sts get-caller-identity  # verify

Step 5: Deploy Remote Infrastructure (one-time per account)

Open a Kiro chat and say:

"Set up AWS Transform custom remote infrastructure for EZT transformations"

Or see POWER.md for manual deployment instructions.

Step 6: Use It

Open a Kiro chat:

"I want to transform my Easytrieve programs to Java. Source: s3://my-bucket/ezt-source/ Input data: s3://my-bucket/input-data/ Baseline output: s3://my-bucket/output-data/"

Kiro does the rest (~10 min per program, 128 concurrent).

---

Path 2: Manual (CDK + CLI)

For manual step-by-step execution, see README-CDK.md.

This deploys an EC2 instance with atx CLI pre-installed. You run the APG epics manually via SSH/SSM:

1. Deploy EC2 via CDK/CloudFormation
2. Create transformation definitions using atx CLI
3. Run BRE extraction
4. Run transformation + validation
5. Review results

---

Repository Structure

sample-mainframe-easytrieve-transform/
├── POWER.md                          # Kiro Power definition
├── README.md                         # This file
├── README-CDK.md                     # Manual path instructions (CDK/CLI)
├── mcp.json                          # MCP server config for Kiro
├── hooks/                            # Kiro agent hooks (workflow enforcement)
├── steering/                         # Phase-specific workflow guides
├── mcp-server/                       # Deterministic orchestration server (Node.js)
│   ├── src/                          # TypeScript source
│   └── dist/                         # Compiled JS (ready to run)
├── test/                             # Sample EZT test workload
├── documents/                        # Transformation definition source files
│   ├── transformation_definition.md
│   ├── bre_transformation_definition.md
│   ├── summaries.md
│   └── ca-easytrieve-report-generator-11-6.txt
├── app.py                            # CDK app entry point
├── easytrieve_transform_stack.py     # CDK stack (EC2 infrastructure)
├── easytrieve-transform-stack.yaml   # CloudFormation template
├── cdk.json                          # CDK config
└── requirements.txt                  # CDK Python dependencies

---

Performance (Validated)

Metric	Result
Single EZT program (low complexity)	~10 min
Parallel capacity	128 concurrent per batch
Max per session	512 programs
Infrastructure cost when idle	$0 (serverless)

---

Links

• APG Pattern
• AWS Transform Custom Docs
• AWS Transform Pricing