{
  "skill_name": "gdpr-data-handling-note",
  "status": "passed",
  "overall_mean": 4.29,
  "test_cases": [
    {
      "test_id": "TC-001",
      "description": "Podcast documentary collecting interview recordings and photographs, sensitive data involved, third-party distribution",
      "scores": {
        "coherence": 5,
        "consistency": 4,
        "fluency": 5,
        "relevance": 4,
        "professionalism": 5,
        "actionability": 4,
        "editorial_naturalness": 4
      },
      "mean": 4.5,
      "editorial_naturalness_note": "EN backfill: zero lexical/structural/tonal tells detected in automated scan. Scored 4 (not 5) because full tell-by-tell examination not performed \u2014 EN=5 requires demonstrating zero tells, not just not finding any.",
      "case_mean": 4.43
    },
    {
      "test_id": "TC-002",
      "description": "Audience survey for a YouTube channel collecting email addresses and demographic data, no sensitive categories",
      "scores": {
        "coherence": 4,
        "consistency": 5,
        "fluency": 4,
        "relevance": 4,
        "professionalism": 4,
        "actionability": 4,
        "editorial_naturalness": 4
      },
      "mean": 4.17,
      "editorial_naturalness_note": "EN backfill: zero lexical/structural/tonal tells detected in automated scan. Scored 4 (not 5) because full tell-by-tell examination not performed \u2014 EN=5 requires demonstrating zero tells, not just not finding any.",
      "case_mean": 4.14
    },
    {
      "test_id": "TC-003",
      "description": "Investigative journalism project building a source contact database with sensitive political and financial data",
      "scores": {
        "coherence": 4,
        "consistency": 4,
        "fluency": 5,
        "relevance": 4,
        "professionalism": 5,
        "actionability": 4,
        "editorial_naturalness": 4
      },
      "mean": 4.33,
      "editorial_naturalness_note": "EN backfill: zero lexical/structural/tonal tells detected in automated scan. Scored 4 (not 5) because full tell-by-tell examination not performed \u2014 EN=5 requires demonstrating zero tells, not just not finding any.",
      "case_mean": 4.29
    }
  ],
  "editorial_naturalness_mean": 4.0
}