{
  "run_id": "20260929T104211_claude-sonnet-5-5_0b43",
  "model": "claude-sonnet-5-5",
  "provider": "Anthropic",
  "effort": "",
  "timestamp": "2026-09-29T10:42:11.999605+00:00",
  "task_count": 6,
  "failed_count": 1,
  "total_time_seconds": 6184.21,
  "time_complete": false,
  "usage_complete": false,
  "previous_attempts": 1,
  "previous_usage": [
    {
      "task_id": "launch-day",
      "task_name": "Launch Day",
      "task_number": 4,
      "wildcard": false,
      "failed": true,
      "attempts": 1,
      "score": 0.0,
      "time_seconds": null,
      "ttft_ms": null,
      "prompt_tokens": null,
      "output_tokens": null,
      "output_tokens_basis": "unavailable",
      "reasoning_tokens": null,
      "tokens_per_second": null,
      "cost_usd": null,
      "cost_basis": ""
    }
  ],
  "prompt_tokens": 534656,
  "output_tokens": 671568,
  "output_tokens_basis": "exact",
  "reasoning_tokens": null,
  "tokens_per_second": 132.54,
  "cost_usd": null,
  "cost_basis": "",
  "cost_complete": false,
  "allowance": null,
  "tasks": [
    {
      "task_id": "aero-lab",
      "task_name": "Aero Lab",
      "task_number": 1,
      "wildcard": false,
      "failed": false,
      "attempts": 1,
      "score": 5.85,
      "time_seconds": 1146.7249704590067,
      "ttft_ms": 955887.6304170117,
      "prompt_tokens": 132791,
      "output_tokens": 147023,
      "output_tokens_basis": "exact",
      "reasoning_tokens": null,
      "tokens_per_second": 128.21121348840097,
      "cost_usd": null,
      "cost_basis": "",
      "functional": 6.5,
      "visual": 5.0,
      "code": 7.0,
      "ux": 5.0,
      "innovation": null,
      "required_pass_rate": 0.33,
      "judge": "llm-vision",
      "scored_by": "claude-code/claude-opus-5-5 (quality-1 rubric)",
      "prompt_fingerprint": "6c713ca03954f51e",
      "scoring_fingerprint": "59929116d9ab5b1e",
      "rubric": "quality-1",
      "evaluation_history": [
        {
          "score": 4.4,
          "rubric": "legacy",
          "scored_by": "claude-code/claude-opus-5-5"
        }
      ],
      "score_note": "",
      "finish_reason": "stop",
      "continuations": 0,
      "html": "01_aero-lab.html",
      "screenshot": "01_aero-lab_screenshot.jpg",
      "frames": "01_aero-lab_frames.jpg",
      "checks": [],
      "lines_changed": null,
      "repair": null
    },
    {
      "task_id": "jet-canyon",
      "task_name": "Jet Canyon",
      "task_number": 2,
      "wildcard": false,
      "failed": false,
      "attempts": 1,
      "score": 7.95,
      "time_seconds": 1183.0014474170166,
      "ttft_ms": 932384.2991670244,
      "prompt_tokens": 133033,
      "output_tokens": 161528,
      "output_tokens_basis": "exact",
      "reasoning_tokens": null,
      "tokens_per_second": 136.54083040446204,
      "cost_usd": null,
      "cost_basis": "",
      "functional": 8.5,
      "visual": 9.0,
      "code": 7.0,
      "ux": 6.5,
      "innovation": null,
      "required_pass_rate": 0.67,
      "judge": "llm-vision",
      "scored_by": "claude-code/claude-opus-5-5 (quality-1 rubric)",
      "prompt_fingerprint": "021db4866371ec15",
      "scoring_fingerprint": "59929116d9ab5b1e",
      "rubric": "quality-1",
      "evaluation_history": [
        {
          "score": 7.7,
          "rubric": "legacy",
          "scored_by": "claude-code/claude-opus-5-5"
        }
      ],
      "score_note": "",
      "finish_reason": "stop",
      "continuations": 0,
      "html": "02_jet-canyon.html",
      "screenshot": "02_jet-canyon_screenshot.jpg",
      "frames": "02_jet-canyon_frames.jpg",
      "checks": [],
      "lines_changed": null,
      "repair": null
    },
    {
      "task_id": "blast-island",
      "task_name": "Blast Island",
      "task_number": 3,
      "wildcard": false,
      "failed": false,
      "attempts": 1,
      "score": 7.7,
      "time_seconds": 1156.92135745805,
      "ttft_ms": 896714.076333039,
      "prompt_tokens": 133825,
      "output_tokens": 154758,
      "output_tokens_basis": "exact",
      "reasoning_tokens": null,
      "tokens_per_second": 133.7670871078301,
      "cost_usd": null,
      "cost_basis": "",
      "functional": 8.5,
      "visual": 8.5,
      "code": 7.0,
      "ux": 6.0,
      "innovation": null,
      "required_pass_rate": 0.56,
      "judge": "llm-vision",
      "scored_by": "claude-code/claude-opus-5-5 (quality-1 rubric)",
      "prompt_fingerprint": "a282381696456fc9",
      "scoring_fingerprint": "59929116d9ab5b1e",
      "rubric": "quality-1",
      "evaluation_history": [
        {
          "score": 7.7,
          "rubric": "legacy",
          "scored_by": "claude-code/claude-opus-5-5"
        }
      ],
      "score_note": "",
      "finish_reason": "stop",
      "continuations": 0,
      "html": "03_blast-island.html",
      "screenshot": "03_blast-island_screenshot.jpg",
      "frames": "03_blast-island_frames.jpg",
      "checks": [],
      "lines_changed": null,
      "repair": null
    },
    {
      "task_id": "launch-day",
      "task_name": "Launch Day",
      "task_number": 4,
      "wildcard": false,
      "failed": true,
      "attempts": 2,
      "score": 0.0,
      "time_seconds": 1117.2532613329968,
      "ttft_ms": null,
      "prompt_tokens": null,
      "output_tokens": null,
      "output_tokens_basis": "unavailable",
      "reasoning_tokens": null,
      "tokens_per_second": null,
      "cost_usd": null,
      "cost_basis": "",
      "functional": 0.0,
      "visual": 0.0,
      "code": 0.0,
      "ux": 0.0,
      "innovation": 0.0,
      "required_pass_rate": 0.0,
      "judge": "error",
      "scored_by": "",
      "prompt_fingerprint": "2787714823d7edc7",
      "scoring_fingerprint": "9ca92a92adfa86a1",
      "rubric": "",
      "evaluation_history": [],
      "score_note": "",
      "finish_reason": "error",
      "continuations": 0,
      "html": null,
      "screenshot": null,
      "frames": null,
      "checks": [],
      "lines_changed": null,
      "repair": null
    },
    {
      "task_id": "camera-shop",
      "task_name": "Alex Optics",
      "task_number": 5,
      "wildcard": false,
      "failed": false,
      "attempts": 1,
      "score": 7.6,
      "time_seconds": 1079.835201665992,
      "ttft_ms": 782753.7366659963,
      "prompt_tokens": 133645,
      "output_tokens": 142142,
      "output_tokens_basis": "exact",
      "reasoning_tokens": null,
      "tokens_per_second": 131.63304898812376,
      "cost_usd": null,
      "cost_basis": "",
      "functional": 8.0,
      "visual": 8.0,
      "code": 7.5,
      "ux": 6.5,
      "innovation": null,
      "required_pass_rate": 0.56,
      "judge": "llm-vision",
      "scored_by": "claude-code/claude-opus-5-5 (quality-1 rubric)",
      "prompt_fingerprint": "e519ec9dadc0a186",
      "scoring_fingerprint": "59929116d9ab5b1e",
      "rubric": "quality-1",
      "evaluation_history": [
        {
          "score": 6.9,
          "rubric": "legacy",
          "scored_by": "claude-code/claude-opus-5-5"
        }
      ],
      "score_note": "",
      "finish_reason": "stop",
      "continuations": 0,
      "html": "05_camera-shop.html",
      "screenshot": "05_camera-shop_screenshot.jpg",
      "frames": "05_camera-shop_frames.jpg",
      "checks": [],
      "lines_changed": null,
      "repair": null
    },
    {
      "task_id": "wild-human-pyramid-balance-puzzle",
      "task_name": "Human Pyramid Balance Puzzle",
      "task_number": 6,
      "wildcard": true,
      "failed": false,
      "attempts": 1,
      "score": 6.95,
      "time_seconds": 500.47653749998426,
      "ttft_ms": 382900.120708975,
      "prompt_tokens": 1362,
      "output_tokens": 66117,
      "output_tokens_basis": "exact",
      "reasoning_tokens": null,
      "tokens_per_second": 132.10809108109703,
      "cost_usd": null,
      "cost_basis": "",
      "functional": 8.0,
      "visual": 7.5,
      "code": 6.5,
      "ux": 5.0,
      "innovation": null,
      "required_pass_rate": 0.17,
      "judge": "llm-vision",
      "scored_by": "claude-code/claude-opus-5-5 (quality-1 rubric)",
      "prompt_fingerprint": "b2b5473f9f99c565",
      "scoring_fingerprint": "59929116d9ab5b1e",
      "rubric": "quality-1",
      "evaluation_history": [
        {
          "score": 7.7,
          "rubric": "legacy",
          "scored_by": "claude-code/claude-opus-5-5"
        }
      ],
      "score_note": "",
      "finish_reason": "stop",
      "continuations": 0,
      "html": "06_wild-human-pyramid-balance-puzzle.html",
      "screenshot": "06_wild-human-pyramid-balance-puzzle_screenshot.jpg",
      "frames": "06_wild-human-pyramid-balance-puzzle_frames.jpg",
      "checks": [],
      "lines_changed": null,
      "repair": null
    }
  ],
  "score": 7.28,
  "functional": 7.88,
  "visual": 7.62,
  "code": 7.12,
  "ux": 6.0,
  "innovation": 0.0,
  "required_pass_rate": 0.42,
  "wildcard_average": 6.95,
  "ranked": true,
  "folder": "20260929t104211_claude-sonnet-5-5_0b43",
  "ran_with": "claude-code",
  "evidence": {
    "kind": "benchmark",
    "model_details": {
      "display_name": "claude-sonnet-5-5",
      "parameters_b": null,
      "active_parameters_b": null,
      "quantization": "",
      "hardware": "",
      "location": "cloud"
    },
    "official_score": null,
    "edition": {
      "id": "100d2dda6497af05",
      "name": "VisualForge Core",
      "tasks": [
        {
          "id": "aero-lab",
          "name": "Aero Lab",
          "number": 1,
          "prompt_fingerprint": "6c713ca03954f51e",
          "rubric_fingerprint": "435131efc3382411",
          "weight": 1
        },
        {
          "id": "blast-island",
          "name": "Blast Island",
          "number": 3,
          "prompt_fingerprint": "a282381696456fc9",
          "rubric_fingerprint": "1e527bc8bc737f25",
          "weight": 1
        },
        {
          "id": "camera-shop",
          "name": "Alex Optics",
          "number": 5,
          "prompt_fingerprint": "e519ec9dadc0a186",
          "rubric_fingerprint": "879611a0955f8590",
          "weight": 1
        },
        {
          "id": "jet-canyon",
          "name": "Jet Canyon",
          "number": 2,
          "prompt_fingerprint": "021db4866371ec15",
          "rubric_fingerprint": "e7dd4d0ee5b37b36",
          "weight": 1
        },
        {
          "id": "launch-day",
          "name": "Launch Day",
          "number": 4,
          "prompt_fingerprint": "2787714823d7edc7",
          "rubric_fingerprint": "a732407575bbc051",
          "weight": 1
        }
      ],
      "generation": {
        "max_continuations": 2,
        "max_html_size_bytes": 5242880,
        "max_tokens": 262144,
        "recording_duration_seconds": 15,
        "sandbox_timeout_seconds": 30,
        "screenshot_height": 1080,
        "screenshot_width": 1920,
        "temperature": null
      },
      "aggregation": "equal-weight-mean-v1"
    },
    "scoring_revision": "",
    "track": "58360bd9df67e558",
    "track_label": "Direct · mixed attempts",
    "coverage": {
      "scored": 4,
      "total": 5
    },
    "evidence_status": "incomplete",
    "attempts": {
      "aero-lab": 1,
      "blast-island": 1,
      "camera-shop": 1,
      "jet-canyon": 1,
      "launch-day": 2
    },
    "execution_harness": "direct",
    "scoring_metadata": "recorded"
  },
  "experiment": null,
  "run_average": 7.28,
  "official_score": null,
  "quality": {
    "standard": "quality-1",
    "label": "Quality v1",
    "score": 7.28,
    "coverage": {
      "scored": 4,
      "total": 5
    },
    "dimensions": {
      "functional": 7.88,
      "visual": 7.62,
      "code": 7.12,
      "ux": 6.0
    },
    "scorers": [
      "claude-code/claude-opus-5-5"
    ]
  }
}
