Refactor-Data-Transformer

The model receives the prompt (and optional system message). The run uses scorer rubric_json_metrics with the JSON configuration below. Pass/fail and partial credit are determined entirely by that scorer against the model output; no human grading.

Return JSON only with keys refactor, rationale, tests. Refactor ETL record transformation into parse, normalize, and project steps. Preserve rejection reporting.

{
  "metrics": {
    "visible": {
      "checks": [
        {
          "contains": [
            "parse"
          ]
        },
        {
          "contains": [
            "normalize"
          ]
        },
        {
          "contains": [
            "project"
          ]
        }
      ]
    },
    "hidden": {
      "checks": [
        {
          "contains": [
            "rejection reporting"
          ]
        },
        {
          "contains": [
            "ETL"
          ]
        },
        {
          "contains": [
            "record"
          ]
        }
      ]
    },
    "intent": {
      "checks": [
        {
          "contains": [
            "pipeline"
          ]
        },
        {
          "contains": [
            "transformation"
          ]
        },
        {
          "contains": [
            "stages"
          ]
        }
      ]
    }
  }
}