{
  "analysis_primary": {
    "bootstrap_resamples": 10000,
    "bootstrap_seed": 20260829,
    "difference": -0.020000000000000004,
    "estimand": "equal-weight mean of 20 task-level official-MCP minus controlled functional-success differences",
    "exact_two_sided_sign_flip_p": 0.881103515625,
    "task_bootstrap_95_interval": [
      -0.15000000000000002,
      0.10000000000000002
    ],
    "tasks_negative": 7,
    "tasks_positive": 7,
    "tasks_tied": 6
  },
  "conditions": {
    "task-diversity-confirmatory-controlled-v2": {
      "records": 100,
      "scored_actor_failures": [],
      "statuses": {
        "complete": 100
      },
      "unscorable_actor_failures": []
    },
    "task-diversity-confirmatory-official-carbon-mcp-v2": {
      "records": 100,
      "scored_actor_failures": [],
      "statuses": {
        "complete": 100
      },
      "unscorable_actor_failures": []
    }
  },
  "eligible": true,
  "format": "ds-eval/task-diversity-confirmatory-v2-result-audit@1",
  "hard_cost_ceiling_usd": 140.0,
  "identity_snapshots": [
    {
      "phase": "preflight",
      "protocol": "2025-11-25",
      "server": {
        "name": "carbon-mcp",
        "title": "Carbon MCP",
        "vendor": "IBM Carbon Design System",
        "version": "1.12.0"
      },
      "tool_list_sha256": "241b836971e64f3502ac7a4b36b6308bebd4c0902be44d141c8786e5862f9da5"
    },
    {
      "phase": "postflight",
      "protocol": "2025-11-25",
      "server": {
        "name": "carbon-mcp",
        "title": "Carbon MCP",
        "vendor": "IBM Carbon Design System",
        "version": "1.12.0"
      },
      "tool_list_sha256": "241b836971e64f3502ac7a4b36b6308bebd4c0902be44d141c8786e5862f9da5"
    }
  ],
  "issues": [],
  "mcp_tool_calls": {
    "by_tool": {
      "mcp__carbon-mcp__code_search": 60,
      "mcp__carbon-mcp__docs_search": 11
    },
    "errors_by_tool": {
      "mcp__carbon-mcp__code_search": 0,
      "mcp__carbon-mcp__docs_search": 0
    },
    "mean_latency_ms": 666.1439154929578,
    "returned_evidence_bytes": 1024953,
    "total": 71
  },
  "recorded_cost_usd": 72.1244016,
  "replacement_attempts": [
    {
      "classification": "unscorable",
      "condition": "task-diversity-confirmatory-controlled-v2",
      "run": 4,
      "task_id": "carbon-notification-design-to-code-r3"
    }
  ],
  "study_id": "carbon-task-diversity-confirmatory-v2"
}
