Sec-Crypto-Utils

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 crypto utility code that hashes passwords with unsalted SHA-1 and compares secrets with ===.

{
  "metrics": {
    "correct": {
      "checks": [
        {
          "contains": [
            "sha-1"
          ]
        },
        {
          "contains": [
            "salt"
          ]
        },
        {
          "contains": [
            "password"
          ]
        },
        {
          "contains": [
            "constant time"
          ]
        },
        {
          "contains": [
            "timing"
          ]
        }
      ]
    },
    "hidden": {
      "checks": [
        {
          "contains": [
            "argon2"
          ]
        },
        {
          "contains": [
            "bcrypt"
          ]
        },
        {
          "contains": [
            "crypto.timingSafeEqual"
          ]
        },
        {
          "contains": [
            "unique salt"
          ]
        },
        {
          "contains": [
            "do not roll your own"
          ]
        }
      ]
    }
  }
}