Refactor-Small-Condition-Extractor

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 repeated eligibility checks into a named predicate. Preserve null handling and boolean result.

{
  "metrics": {
    "visible": {
      "checks": [
        {
          "contains": [
            "predicate"
          ]
        },
        {
          "contains": [
            "eligibility"
          ]
        },
        {
          "contains": [
            "boolean"
          ]
        }
      ]
    },
    "hidden": {
      "checks": [
        {
          "contains": [
            "null handling"
          ]
        },
        {
          "contains": [
            "repeated"
          ]
        },
        {
          "contains": [
            "same result"
          ]
        }
      ]
    },
    "intent": {
      "checks": [
        {
          "contains": [
            "extract function"
          ]
        },
        {
          "contains": [
            "duplication"
          ]
        },
        {
          "contains": [
            "intent"
          ]
        }
      ]
    }
  }
}