Reason-Constraint-Feature-Flag-Priority

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. Flag A enables a feature for all users. Flag B explicitly disables the same feature for free-tier users. A user is on the free tier. Which flag wins and is the feature enabled or disabled for this user?

{
  "metrics": {
    "accuracy": {
      "checks": [
        {
          "contains": [
            "disabled"
          ]
        },
        {
          "contains": [
            "deny"
          ]
        },
        {
          "contains": [
            "Flag B"
          ]
        }
      ]
    },
    "evidence": {
      "checks": [
        {
          "contains": [
            "free-tier"
          ]
        },
        {
          "contains": [
            "overrides"
          ]
        },
        {
          "contains": [
            "explicit"
          ]
        }
      ]
    },
    "constraint": {
      "checks": [
        {
          "contains": [
            "more specific"
          ]
        },
        {
          "contains": [
            "deny wins"
          ]
        },
        {
          "contains": [
            "flag priority"
          ]
        }
      ]
    }
  }
}