mock-generation

Example: mock generation from a real execution

Capture a real, completed Step Functions execution and turn its per-state outputs into mock data you can replay locally — plus basic history inspection. See CLI & history.

Requirements

AWS credentials + a region (see the Setup guide) and a real, completed execution ARN of one of your own state machines. There's no bundled ASL, and no role_arn is needed — this only reads the execution. No Docker.

What to do

Point it at one of your executions via the EXECUTION_ARN env var:

EXECUTION_ARN=arn:aws:states:<region>:<account>:execution:MyStateMachine:run-1 \
  uv run --python=3.13 --with aws-stepfunctions-toolkit python run.py

This writes mock_data/{history.json,input.json,state_outputs.json}. state_outputs.json holds each state's recorded output, ready to back StaticMockResponseStrategy entries when you wire up a runner. Pair with running a sub-range to replay a failure from the middle of a real run.

The same thing from the CLI:

uv run --with aws-stepfunctions-toolkit sfn-toolkit generate-mock \
  arn:aws:states:<region>:<account>:execution:MyStateMachine:run-1