{
  "version": 5,
  "program": "explicit_finish_after_verification_v5",
  "title": "Finish only under an explicit registered verification policy",
  "kind": "hypothetical-explicit-opt-in-completion-scenario",
  "policy": {
    "root": "/testbed",
    "maxCommandBytes": 4096,
    "maxOutputBytes": 32768,
    "maxVerificationHistoryTurns": 4,
    "requiredEvidence": [
      "registered exact pytest argv succeeds with positive passed-count summary",
      "subsequent successful full-tree git diff --check",
      "no intervening unknown or mutating action"
    ],
    "requiresExplicitRegisteredCompletionPolicy": true,
    "requiresRegisteredRequiredTestCommand": true,
    "requiresUnchangedTreeAndEnvironment": true,
    "requiresNoUnverifiedStagedOrUntrackedChanges": true,
    "historicalOptInPresent": false,
    "deploymentCreditCalls": 0,
    "terminalCommand": "COMPLETE_TASK_AND_SUBMIT_FINAL_OUTPUT",
    "allowFinalProseInstead": false
  },
  "trajectories": 2498,
  "issues": 500,
  "denominator": {
    "calls": 129846,
    "callsWithUsage": 129846,
    "inputTokens": 3201225688,
    "inputTokensReportedCalls": 129846,
    "outputTokens": 42377608,
    "outputTokensReportedCalls": 129846,
    "totalTokens": 3243603296,
    "totalTokensReportedCalls": 129846,
    "cachedInputTokens": 2498865874,
    "cachedInputTokensReportedCalls": 121228,
    "cacheCreationInputTokens": 31547900,
    "cacheCreationInputTokensReportedCalls": 40589,
    "reasoningOutputTokens": 23075825,
    "reasoningOutputTokensReportedCalls": 129846,
    "sourceReportedCostSum": 1344.591070199989,
    "callsWithReportedCost": 129846
  },
  "proposals": 0,
  "matchedCalls": 0,
  "matchedIssues": 0,
  "matchedUsage": {
    "calls": 0,
    "callsWithUsage": 0,
    "inputTokens": 0,
    "inputTokensReportedCalls": 0,
    "outputTokens": 0,
    "outputTokensReportedCalls": 0,
    "totalTokens": 0,
    "totalTokensReportedCalls": 0,
    "cachedInputTokens": 0,
    "cachedInputTokensReportedCalls": 0,
    "cacheCreationInputTokens": 0,
    "cacheCreationInputTokensReportedCalls": 0,
    "reasoningOutputTokens": 0,
    "reasoningOutputTokensReportedCalls": 0,
    "sourceReportedCostSum": 0,
    "callsWithReportedCost": 0
  },
  "matchPercent": null,
  "callPercent": 0.0,
  "tokenPercent": 0.0,
  "deploymentCreditCalls": 0,
  "deploymentCreditTokens": 0,
  "matchesWithReasoningOrProse": 0,
  "diagnostics": {
    "diffCheckLikeTurns": 40,
    "supportedPassedFullTreeCheckTurns": 7
  },
  "byPartition": {
    "discovery": {
      "trajectories": 550,
      "issues": 110,
      "denominator": {
        "calls": 28862,
        "callsWithUsage": 28862,
        "inputTokens": 702102702,
        "inputTokensReportedCalls": 28862,
        "outputTokens": 9426963,
        "outputTokensReportedCalls": 28862,
        "totalTokens": 711529665,
        "totalTokensReportedCalls": 28862,
        "cachedInputTokens": 543584191,
        "cachedInputTokensReportedCalls": 27013,
        "cacheCreationInputTokens": 6709415,
        "cacheCreationInputTokensReportedCalls": 8757,
        "reasoningOutputTokens": 5220878,
        "reasoningOutputTokensReportedCalls": 28862,
        "sourceReportedCostSum": 291.47569229000186,
        "callsWithReportedCost": 28862
      },
      "proposals": 0,
      "matchedCalls": 0,
      "matchedIssues": 0,
      "matchedUsage": {
        "calls": 0,
        "callsWithUsage": 0,
        "inputTokens": 0,
        "inputTokensReportedCalls": 0,
        "outputTokens": 0,
        "outputTokensReportedCalls": 0,
        "totalTokens": 0,
        "totalTokensReportedCalls": 0,
        "cachedInputTokens": 0,
        "cachedInputTokensReportedCalls": 0,
        "cacheCreationInputTokens": 0,
        "cacheCreationInputTokensReportedCalls": 0,
        "reasoningOutputTokens": 0,
        "reasoningOutputTokensReportedCalls": 0,
        "sourceReportedCostSum": 0,
        "callsWithReportedCost": 0
      },
      "matchPercent": null,
      "callPercent": 0.0,
      "tokenPercent": 0.0,
      "deploymentCreditCalls": 0,
      "deploymentCreditTokens": 0,
      "matchesWithReasoningOrProse": 0,
      "diagnostics": {
        "diffCheckLikeTurns": 13,
        "supportedPassedFullTreeCheckTurns": 2
      }
    },
    "validation": {
      "trajectories": 1948,
      "issues": 390,
      "denominator": {
        "calls": 100984,
        "callsWithUsage": 100984,
        "inputTokens": 2499122986,
        "inputTokensReportedCalls": 100984,
        "outputTokens": 32950645,
        "outputTokensReportedCalls": 100984,
        "totalTokens": 2532073631,
        "totalTokensReportedCalls": 100984,
        "cachedInputTokens": 1955281683,
        "cachedInputTokensReportedCalls": 94215,
        "cacheCreationInputTokens": 24838485,
        "cacheCreationInputTokensReportedCalls": 31832,
        "reasoningOutputTokens": 17854947,
        "reasoningOutputTokensReportedCalls": 100984,
        "sourceReportedCostSum": 1053.1153779099996,
        "callsWithReportedCost": 100984
      },
      "proposals": 0,
      "matchedCalls": 0,
      "matchedIssues": 0,
      "matchedUsage": {
        "calls": 0,
        "callsWithUsage": 0,
        "inputTokens": 0,
        "inputTokensReportedCalls": 0,
        "outputTokens": 0,
        "outputTokensReportedCalls": 0,
        "totalTokens": 0,
        "totalTokensReportedCalls": 0,
        "cachedInputTokens": 0,
        "cachedInputTokensReportedCalls": 0,
        "cacheCreationInputTokens": 0,
        "cacheCreationInputTokensReportedCalls": 0,
        "reasoningOutputTokens": 0,
        "reasoningOutputTokensReportedCalls": 0,
        "sourceReportedCostSum": 0,
        "callsWithReportedCost": 0
      },
      "matchPercent": null,
      "callPercent": 0.0,
      "tokenPercent": 0.0,
      "deploymentCreditCalls": 0,
      "deploymentCreditTokens": 0,
      "matchesWithReasoningOrProse": 0,
      "diagnostics": {
        "diffCheckLikeTurns": 27,
        "supportedPassedFullTreeCheckTurns": 5
      }
    }
  },
  "cohorts": {
    "deepseek32": {
      "trajectories": 500,
      "issues": 500,
      "denominator": {
        "calls": 43683,
        "callsWithUsage": 43683,
        "inputTokens": 1177730550,
        "inputTokensReportedCalls": 43683,
        "outputTokens": 13441973,
        "outputTokensReportedCalls": 43683,
        "totalTokens": 1191172523,
        "totalTokensReportedCalls": 43683,
        "cachedInputTokens": 798975703,
        "cachedInputTokensReportedCalls": 43683,
        "cacheCreationInputTokens": 0,
        "cacheCreationInputTokensReportedCalls": 0,
        "reasoningOutputTokens": 7941114,
        "reasoningOutputTokensReportedCalls": 43683,
        "sourceReportedCostSum": 223.9232482499986,
        "callsWithReportedCost": 43683
      },
      "proposals": 0,
      "matchedCalls": 0,
      "matchedIssues": 0,
      "matchedUsage": {
        "calls": 0,
        "callsWithUsage": 0,
        "inputTokens": 0,
        "inputTokensReportedCalls": 0,
        "outputTokens": 0,
        "outputTokensReportedCalls": 0,
        "totalTokens": 0,
        "totalTokensReportedCalls": 0,
        "cachedInputTokens": 0,
        "cachedInputTokensReportedCalls": 0,
        "cacheCreationInputTokens": 0,
        "cacheCreationInputTokensReportedCalls": 0,
        "reasoningOutputTokens": 0,
        "reasoningOutputTokensReportedCalls": 0,
        "sourceReportedCostSum": 0,
        "callsWithReportedCost": 0
      },
      "matchPercent": null,
      "callPercent": 0.0,
      "tokenPercent": 0.0,
      "deploymentCreditCalls": 0,
      "deploymentCreditTokens": 0,
      "matchesWithReasoningOrProse": 0,
      "diagnostics": {
        "diffCheckLikeTurns": 37,
        "supportedPassedFullTreeCheckTurns": 7
      }
    },
    "gemini3flash": {
      "trajectories": 498,
      "issues": 498,
      "denominator": {
        "calls": 28054,
        "callsWithUsage": 28054,
        "inputTokens": 847926935,
        "inputTokensReportedCalls": 28054,
        "outputTokens": 10071357,
        "outputTokensReportedCalls": 28054,
        "totalTokens": 857998292,
        "totalTokensReportedCalls": 28054,
        "cachedInputTokens": 613767718,
        "cachedInputTokensReportedCalls": 19436,
        "cacheCreationInputTokens": 0,
        "cacheCreationInputTokensReportedCalls": 0,
        "reasoningOutputTokens": 7549905,
        "reasoningOutputTokensReportedCalls": 28054,
        "sourceReportedCostSum": 177.98206539999973,
        "callsWithReportedCost": 28054
      },
      "proposals": 0,
      "matchedCalls": 0,
      "matchedIssues": 0,
      "matchedUsage": {
        "calls": 0,
        "callsWithUsage": 0,
        "inputTokens": 0,
        "inputTokensReportedCalls": 0,
        "outputTokens": 0,
        "outputTokensReportedCalls": 0,
        "totalTokens": 0,
        "totalTokensReportedCalls": 0,
        "cachedInputTokens": 0,
        "cachedInputTokensReportedCalls": 0,
        "cacheCreationInputTokens": 0,
        "cacheCreationInputTokensReportedCalls": 0,
        "reasoningOutputTokens": 0,
        "reasoningOutputTokensReportedCalls": 0,
        "sourceReportedCostSum": 0,
        "callsWithReportedCost": 0
      },
      "matchPercent": null,
      "callPercent": 0.0,
      "tokenPercent": 0.0,
      "deploymentCreditCalls": 0,
      "deploymentCreditTokens": 0,
      "matchesWithReasoningOrProse": 0,
      "diagnostics": {}
    },
    "gpt52": {
      "trajectories": 500,
      "issues": 500,
      "denominator": {
        "calls": 17520,
        "callsWithUsage": 17520,
        "inputTokens": 322598043,
        "inputTokensReportedCalls": 17520,
        "outputTokens": 6692906,
        "outputTokensReportedCalls": 17520,
        "totalTokens": 329290949,
        "totalTokensReportedCalls": 17520,
        "cachedInputTokens": 267596032,
        "cachedInputTokensReportedCalls": 17520,
        "cacheCreationInputTokens": 0,
        "cacheCreationInputTokensReportedCalls": 0,
        "reasoningOutputTokens": 4871131,
        "reasoningOutputTokensReportedCalls": 17520,
        "sourceReportedCostSum": 236.78350885000094,
        "callsWithReportedCost": 17520
      },
      "proposals": 0,
      "matchedCalls": 0,
      "matchedIssues": 0,
      "matchedUsage": {
        "calls": 0,
        "callsWithUsage": 0,
        "inputTokens": 0,
        "inputTokensReportedCalls": 0,
        "outputTokens": 0,
        "outputTokensReportedCalls": 0,
        "totalTokens": 0,
        "totalTokensReportedCalls": 0,
        "cachedInputTokens": 0,
        "cachedInputTokensReportedCalls": 0,
        "cacheCreationInputTokens": 0,
        "cacheCreationInputTokensReportedCalls": 0,
        "reasoningOutputTokens": 0,
        "reasoningOutputTokensReportedCalls": 0,
        "sourceReportedCostSum": 0,
        "callsWithReportedCost": 0
      },
      "matchPercent": null,
      "callPercent": 0.0,
      "tokenPercent": 0.0,
      "deploymentCreditCalls": 0,
      "deploymentCreditTokens": 0,
      "matchesWithReasoningOrProse": 0,
      "diagnostics": {}
    },
    "opus45": {
      "trajectories": 500,
      "issues": 500,
      "denominator": {
        "calls": 16446,
        "callsWithUsage": 16446,
        "inputTokens": 330622219,
        "inputTokensReportedCalls": 16446,
        "outputTokens": 4806100,
        "outputTokensReportedCalls": 16446,
        "totalTokens": 335428319,
        "totalTokensReportedCalls": 16446,
        "cachedInputTokens": 314095302,
        "cachedInputTokensReportedCalls": 16446,
        "cacheCreationInputTokens": 13695410,
        "cacheCreationInputTokensReportedCalls": 16446,
        "reasoningOutputTokens": 1193410,
        "reasoningOutputTokensReportedCalls": 16446,
        "sourceReportedCostSum": 376.95399849999967,
        "callsWithReportedCost": 16446
      },
      "proposals": 0,
      "matchedCalls": 0,
      "matchedIssues": 0,
      "matchedUsage": {
        "calls": 0,
        "callsWithUsage": 0,
        "inputTokens": 0,
        "inputTokensReportedCalls": 0,
        "outputTokens": 0,
        "outputTokensReportedCalls": 0,
        "totalTokens": 0,
        "totalTokensReportedCalls": 0,
        "cachedInputTokens": 0,
        "cachedInputTokensReportedCalls": 0,
        "cacheCreationInputTokens": 0,
        "cacheCreationInputTokensReportedCalls": 0,
        "reasoningOutputTokens": 0,
        "reasoningOutputTokensReportedCalls": 0,
        "sourceReportedCostSum": 0,
        "callsWithReportedCost": 0
      },
      "matchPercent": null,
      "callPercent": 0.0,
      "tokenPercent": 0.0,
      "deploymentCreditCalls": 0,
      "deploymentCreditTokens": 0,
      "matchesWithReasoningOrProse": 0,
      "diagnostics": {}
    },
    "sonnet45": {
      "trajectories": 500,
      "issues": 500,
      "denominator": {
        "calls": 24143,
        "callsWithUsage": 24143,
        "inputTokens": 522347941,
        "inputTokensReportedCalls": 24143,
        "outputTokens": 7365272,
        "outputTokensReportedCalls": 24143,
        "totalTokens": 529713213,
        "totalTokensReportedCalls": 24143,
        "cachedInputTokens": 504431119,
        "cachedInputTokensReportedCalls": 24143,
        "cacheCreationInputTokens": 17852490,
        "cacheCreationInputTokensReportedCalls": 24143,
        "reasoningOutputTokens": 1520265,
        "reasoningOutputTokensReportedCalls": 24143,
        "sourceReportedCostSum": 328.9482491999996,
        "callsWithReportedCost": 24143
      },
      "proposals": 0,
      "matchedCalls": 0,
      "matchedIssues": 0,
      "matchedUsage": {
        "calls": 0,
        "callsWithUsage": 0,
        "inputTokens": 0,
        "inputTokensReportedCalls": 0,
        "outputTokens": 0,
        "outputTokensReportedCalls": 0,
        "totalTokens": 0,
        "totalTokensReportedCalls": 0,
        "cachedInputTokens": 0,
        "cachedInputTokensReportedCalls": 0,
        "cacheCreationInputTokens": 0,
        "cacheCreationInputTokensReportedCalls": 0,
        "reasoningOutputTokens": 0,
        "reasoningOutputTokensReportedCalls": 0,
        "sourceReportedCostSum": 0,
        "callsWithReportedCost": 0
      },
      "matchPercent": null,
      "callPercent": 0.0,
      "tokenPercent": 0.0,
      "deploymentCreditCalls": 0,
      "deploymentCreditTokens": 0,
      "matchesWithReasoningOrProse": 0,
      "diagnostics": {
        "diffCheckLikeTurns": 3
      }
    }
  },
  "split": {
    "method": "SHA256('autotoken-v5:'+issue)%5==0 discovery; remaining validation; issue grouped across models",
    "policyFrozenBeforeValidation": true,
    "notUntouchedExternalCorpus": true
  },
  "sourceManifestSha256": "73cccd8079f5c3150015207a7d102f1a4409c7996f0ea9ae1305f1b45239247b",
  "plannerSha256": "797ed0b7305b8ccf1577b3cd510ebf7d8ee2d6cf0f6c9c4271773a8463df33bf",
  "limitations": [
    "Historical explicit completion opt-in and required-test registration are absent: deployment credit is zero even when a conditional syntax match exists.",
    "The scenario treats the exact prior pytest argv as hypothetically registered required verification, never a registration inferred for deployment.",
    "A positive pytest summary alone never proposes completion: a later successful full-tree diff check is required within four turns and only supported read-only operations can intervene.",
    "Pipelines, partial-file checks, redirects other than stderr merge, unknown/mutating actions, failed tests, missing results and stale verification abstain.",
    "Historical check does not prove staged/untracked changes are approved or that the test selection is sufficient. Runtime policy must register these conditions.",
    "Exact singleton native submission is the primary label; final prose and submission bundles are not equivalent.",
    "Submission may contain needed reasoning; removing it can prematurely finish an incomplete task.",
    "Matched original usage is a conditional gross opportunity, not validated savings. Router, policy registration, reasoning loss and changed future context/cache costs are excluded.",
    "Cache/reasoning counts are subsets of input/output totals; no double counting.",
    "No recorded commands executed, tasks submitted or paid inference used."
  ]
}
