Sec-Csp-Parser

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 a CSP parser that treats unknown directives as allow-all and collapses duplicate directives incorrectly.

{
  "metrics": {
    "correct": {
      "checks": [
        {
          "contains": [
            "CSP"
          ]
        },
        {
          "contains": [
            "directive"
          ]
        },
        {
          "contains": [
            "duplicate"
          ]
        },
        {
          "contains": [
            "allow"
          ]
        },
        {
          "contains": [
            "parser"
          ]
        }
      ]
    },
    "hidden": {
      "checks": [
        {
          "contains": [
            "fail closed"
          ]
        },
        {
          "contains": [
            "report"
          ]
        },
        {
          "contains": [
            "default-src"
          ]
        },
        {
          "contains": [
            "unknown"
          ]
        },
        {
          "contains": [
            "normalize"
          ]
        }
      ]
    }
  }
}