{
  "dataset": "lobbyagent-streaming-prompt-readiness",
  "version": "0.1.1-seed",
  "coordinate_unit": "unicode_codepoint",
  "generated_at": "2026-09-07T11:44:03.883Z",
  "source_file": "data/streaming-ready/dialogues.jsonl",
  "source_sha256": "57c327c5980bbc0161f7c93fb3eca9455fc93c2a62dcb112f90c337e418a07d3",
  "stats": {
    "dialogues": 60,
    "ready_intervals": 52,
    "labelled_chars": {
      "ready": 15169,
      "not_ready": 21519
    },
    "splits": {
      "train": {
        "dialogues": 42,
        "with_ready": 29,
        "without_ready": 13
      },
      "dev": {
        "dialogues": 10,
        "with_ready": 5,
        "without_ready": 5
      },
      "test": {
        "dialogues": 8,
        "with_ready": 6,
        "without_ready": 2
      }
    },
    "predictability": {
      "high": 51,
      "medium": 1,
      "low": 0
    }
  },
  "files": {
    "ready-train.jsonl": {
      "rows": 42,
      "bytes": 144081,
      "sha256": "02b0e65bb16a47bd940d5b824caa2d2561933240d7e278a33989a82c31a2ba6c"
    },
    "ready-dev.jsonl": {
      "rows": 10,
      "bytes": 35073,
      "sha256": "be8706266b76e95a3f832068f2aa93be4b01f2b1b390fe56995e5b5d3342471c"
    },
    "ready-test.jsonl": {
      "rows": 8,
      "bytes": 27724,
      "sha256": "60f92d31b9a8fd7506d502dc05c0a8d05e1328cc8aef2961bdec39b22a900a8f"
    }
  },
  "label_definition": "At a character position, READY=1 iff the visible prefix is sufficient to start drafting an executable, bounded, parallel-safe Coding Agent prompt without inventing unresolved product decisions.",
  "warnings": [
    "Synthetic seed data is for pipeline validation. Review boundary annotations before scaling or drawing product conclusions.",
    "candidate_prompt is intentionally absent from student training data.",
    "Low-predictability intervals should be excluded from the first student model."
  ]
}