Sec-Abac-Rule-Engine

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 verdict, risk, mitigation. Review an ABAC rule engine that treats missing attributes as true and falls back to allow on parser errors.

{
  "metrics": {
    "correct": {
      "checks": [
        {
          "contains": [
            "ABAC"
          ]
        },
        {
          "contains": [
            "missing attribute"
          ]
        },
        {
          "contains": [
            "fail closed"
          ]
        },
        {
          "contains": [
            "deny"
          ]
        },
        {
          "contains": [
            "parser"
          ]
        }
      ]
    },
    "hidden": {
      "checks": [
        {
          "contains": [
            "default deny"
          ]
        },
        {
          "contains": [
            "audit"
          ]
        },
        {
          "contains": [
            "policy"
          ]
        },
        {
          "contains": [
            "least privilege"
          ]
        },
        {
          "contains": [
            "error"
          ]
        }
      ]
    }
  }
}