Reasoning-Mas-Blue-Green-Rollback

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 answer, evidence, constraints. Blue-green deploy notes say traffic was shifted to green, migrations are backward compatible, but feature flags were enabled only in blue. Explain the rollback hazard.

{
  "metrics": {
    "accuracy": {
      "checks": [
        {
          "contains": [
            "feature flags"
          ]
        },
        {
          "contains": [
            "blue"
          ]
        },
        {
          "contains": [
            "rollback"
          ]
        }
      ]
    },
    "evidence": {
      "checks": [
        {
          "contains": [
            "green"
          ]
        },
        {
          "contains": [
            "traffic shifted"
          ]
        },
        {
          "contains": [
            "migration"
          ]
        }
      ]
    },
    "constraint": {
      "checks": [
        {
          "contains": [
            "configuration drift"
          ]
        },
        {
          "contains": [
            "backward compatible"
          ]
        },
        {
          "contains": [
            "hazard"
          ]
        }
      ]
    }
  }
}