Speed-Batch-Transform-Spec

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 benchmark, title, transforms, done. Set benchmark to "speed", title to "Batch Transform Spec", and done to true. Convert 28 raw import rules into normalized transform objects with source, target, operation, and note.

{
  "metrics": {
    "complete": {
      "checks": [
        {
          "contains": [
            "speed"
          ]
        },
        {
          "contains": [
            "done"
          ]
        },
        {
          "contains": [
            "batch"
          ]
        }
      ]
    }
  }
}