{
  "schema": "agent-rerun-receipt.v1",
  "record": {
    "id": "native-codex-extra-regression-repair",
    "taskId": "extra-regression-repair",
    "taskTitle": "Repair money calculations across modules",
    "agent": "Codex CLI",
    "requestedModel": "gpt-6.1-sol",
    "resolvedModel": null,
    "wallSeconds": 37.556,
    "inputTokens": 114952,
    "outputTokens": 1175,
    "totalTokens": 116127,
    "costUsd": null,
    "energyWh": null,
    "assertionsPassed": 10,
    "assertionsTotal": 10,
    "observedNativeModel": "gpt-6.1-sol",
    "modelIdentityStatus": "MATCHED_NATIVE_RECEIPT",
    "modelIdentityEvidence": "native-effective-config; provider response identity not exposed",
    "providerResponseModel": null,
    "reasoningEffort": null,
    "family": "regression-repair",
    "status": "PASS",
    "tier": "native subscription",
    "diagnosticOnly": false,
    "comparisonEligible": null,
    "cohort": "original-model",
    "resultNote": null,
    "limitations": [
      "Auth uses existing native subscription/prepaid account; billing decrement unavailable.",
      "Native models differ: this is a product pilot, not a same-model comparison.",
      "Synthetic authored fixture; no statistical ranking.",
      "Custom pilot receipt has not been admitted through BenchmarkResult to JSON/DuckDB/Neon; no leaderboard promotion.",
      "Provider-response model identity not exposed; native effective selection is recorded separately and does not change acceptance status."
    ],
    "evidenceFile": "native-codex-extra-regression-repair.json"
  },
  "sourceRecord": {
    "id": "native-codex-extra-regression-repair",
    "family": "regression-repair",
    "taskId": "extra-regression-repair",
    "taskTitle": "Repair money calculations across modules",
    "agent": "Codex CLI",
    "requestedModel": "gpt-6.1-sol",
    "resolvedModel": null,
    "status": "PASS",
    "wallSeconds": 37.556,
    "inputTokens": 114952,
    "outputTokens": 1175,
    "totalTokens": 116127,
    "costUsd": null,
    "energyWh": null,
    "assertionsPassed": 10,
    "assertionsTotal": 10,
    "behaviorPassed": 9,
    "behaviorTotal": 9,
    "scopePassed": true,
    "changedPaths": [
      "cart.py",
      "invoice.py"
    ],
    "addedPaths": [],
    "processExitCode": 0,
    "oracleExitCode": 0,
    "evidencePaths": [
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/command.json",
        "sha256": "f6c5f35d54000cf4775fc637e8257a7521a75b9c565a0f644d197f10ce5a96bf"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/freeze.json",
        "sha256": "284724208cd620d00b1b6482dd2dca9e76e9d8a1bd4c55e4e32446e0cb576591"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/last_message.txt",
        "sha256": "622722f1f4ff61acd671d8920dc4fadcfaf9bb791650c81643cb045b5f5ac20a"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/native-model-identity.json",
        "sha256": "67aa81206d8044d456ee50021325331b88d1e927d9e6b73da80ec28d69d171c8"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/oracle.stderr.log",
        "sha256": "9377f94d8d8f544e85f61b5da82a33f574e58569ec47832f0ed6cf04accae8b5"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/oracle.stdout.log",
        "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/prompt.txt",
        "sha256": "b70f397a0ce904320147c810719cfc7bc623ce182820f58b92dccdfacd2924c7"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/sandbox.sb",
        "sha256": "30fc51bcfa69883d9155aa6ff865290c65bb16aa31a3a23956e9d3c8fe798cce"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/stderr.log",
        "sha256": "68dc25506cebfd4d87553dcadd637c22c2be9c6c3964b3e4ed36667ec3c1f600"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/stdout.log",
        "sha256": "e73989e0142568b84ab05826e7dc6456d64381983f55bc67304d340dff4ff2da"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/workspace/README.md",
        "sha256": "b8c8d2a4e51ac441528ef18987f81f996c60ec1367a08f0d4ee6c0e68c3f3fe5"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/workspace/cart.py",
        "sha256": "07a60e4171c13cc08f897e1c1e947049ba4219cdc50d640759731e298518955f"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/workspace/invoice.py",
        "sha256": "e352ed50de0bf49c480562ec7321c17ee553f722dad2d471d05f069c6f5adcd2"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/workspace/test_existing.py",
        "sha256": "fd05d4cf9520d0be75d3629fe64c0365c5848eed199c3d19ddacf18542cf42fb"
      }
    ],
    "limitations": [
      "Auth uses existing native subscription/prepaid account; billing decrement unavailable.",
      "Native models differ: this is a product pilot, not a same-model comparison.",
      "Synthetic authored fixture; no statistical ranking.",
      "Custom pilot receipt has not been admitted through BenchmarkResult to JSON/DuckDB/Neon; no leaderboard promotion.",
      "Provider-response model identity not exposed; native effective selection is recorded separately and does not change acceptance status."
    ],
    "observedNativeModel": "gpt-6.1-sol",
    "modelIdentityEvidence": "native-effective-config; provider response identity not exposed",
    "providerResponseModel": null,
    "threeStoreAdmission": "NOT_VERIFIED",
    "admitted": false,
    "tier": "native subscription",
    "modelIdentityStatus": "MATCHED_NATIVE_RECEIPT",
    "acceptanceStatus": "PASS"
  },
  "sourceFile": "artifacts/agent_comparison_rerun_20261016/native/reviewed-summary.json",
  "sourceFileSha256": "d470470c759ba7357132831170430dcd2876add878a4fb767db2fe55d9163a70",
  "files": [
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/command.json",
      "sha256": "f6c5f35d54000cf4775fc637e8257a7521a75b9c565a0f644d197f10ce5a96bf",
      "bytes": 1077
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/freeze.json",
      "sha256": "284724208cd620d00b1b6482dd2dca9e76e9d8a1bd4c55e4e32446e0cb576591",
      "bytes": 859,
      "content": "{\n  \"frozenAt\": \"2026-10-16T03:57:34.732565+00:00\",\n  \"syntheticInput\": true,\n  \"promptSha256\": \"b70f397a0ce904320147c810719cfc7bc623ce182820f58b92dccdfacd2924c7\",\n  \"fixtureContractSha256\": \"133564a7ef2a4d40230de10d32c24c9650e1c9f60d910db4a45b4df9301e91a8\",\n  \"fixtureHashes\": {\n    \"invoice.py\": \"511f665ac9f198741d8ba1731979923a0d9c0c52f11d05d4eb7825b30b414f61\",\n    \"cart.py\": \"d639d79d64eba17b909d01e3bd0e4c0fb43c313b2dc7e4748af0149fca185867\",\n    \"README.md\": \"b8c8d2a4e51ac441528ef18987f81f996c60ec1367a08f0d4ee6c0e68c3f3fe5\",\n    \"test_existing.py\": \"fd05d4cf9520d0be75d3629fe64c0365c5848eed199c3d19ddacf18542cf42fb\"\n  },\n  \"requestedModel\": \"gpt-6.1-sol\",\n  \"taskId\": \"extra-regression-repair\",\n  \"workingDirectory\": \"<temporary-workspace>\",\n  \"oracleSha256\": \"168cb53da5c1645a4afd68d5a7bf6918a9165017fc65c13d9dcf29bed4788204\"\n}\n",
      "contentRedacted": true,
      "publishedContentSha256": "3b317f9c521ba16575d18f8dfef27ecd1ad1073a92f71ee4a90b21fa5af34a46"
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/last_message.txt",
      "sha256": "622722f1f4ff61acd671d8920dc4fadcfaf9bb791650c81643cb045b5f5ac20a",
      "bytes": 363
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/native-model-identity.json",
      "sha256": "67aa81206d8044d456ee50021325331b88d1e927d9e6b73da80ec28d69d171c8",
      "bytes": 410,
      "content": "{\n  \"threadId\": \"01a142db-abe1-7280-91ad-b5523058ed7a\",\n  \"effectiveModels\": [\n    \"gpt-6.1-sol\"\n  ],\n  \"nativeModelProviders\": [\n    \"openai\"\n  ],\n  \"nativeTurnContexts\": [\n    {\n      \"model\": \"gpt-6.1-sol\",\n      \"effort\": \"low\"\n    }\n  ],\n  \"providerResponseModel\": null,\n  \"evidenceBoundary\": \"Native effective selection recorded by the installed CLI; remote provider response identity is not exposed.\"\n}\n",
      "contentRedacted": false,
      "publishedContentSha256": "67aa81206d8044d456ee50021325331b88d1e927d9e6b73da80ec28d69d171c8"
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/oracle.stderr.log",
      "sha256": "9377f94d8d8f544e85f61b5da82a33f574e58569ec47832f0ed6cf04accae8b5",
      "bytes": 783,
      "content": "test_decimal_type (__main__.RegressionOracle.test_decimal_type) ... ok\ntest_empty_cart_type (__main__.RegressionOracle.test_empty_cart_type) ... ok\ntest_generator (__main__.RegressionOracle.test_generator) ... ok\ntest_invalid_quantity (__main__.RegressionOracle.test_invalid_quantity) ... ok\ntest_multiply_before_rounding (__main__.RegressionOracle.test_multiply_before_rounding) ... ok\ntest_negative_price (__main__.RegressionOracle.test_negative_price) ... ok\ntest_round_each_line (__main__.RegressionOracle.test_round_each_line) ... ok\ntest_round_half_up (__main__.RegressionOracle.test_round_half_up) ... ok\ntest_zero_quantity (__main__.RegressionOracle.test_zero_quantity) ... ok\n\n----------------------------------------------------------------------\nRan 9 tests in 0.001s\n\nOK\n",
      "contentRedacted": false,
      "publishedContentSha256": "9377f94d8d8f544e85f61b5da82a33f574e58569ec47832f0ed6cf04accae8b5"
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/oracle.stdout.log",
      "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
      "bytes": 0,
      "content": "",
      "contentRedacted": false,
      "publishedContentSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855"
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/prompt.txt",
      "sha256": "b70f397a0ce904320147c810719cfc7bc623ce182820f58b92dccdfacd2924c7",
      "bytes": 624,
      "content": "Repair this standard-library-only Python invoice package. invoice.line_total(price, quantity) returns a Decimal rounded to cents with ROUND_HALF_UP after multiplying the exact decimal price by quantity; quantity must be an int >= 0 and bool is invalid (raise ValueError). cart.total(lines) sums individually rounded line totals into a Decimal including Decimal('0.00') for an empty cart. Existing callers of invoice.line_total and cart.total must remain compatible. Fix only invoice.py and cart.py, do not modify README.md or add dependencies. Run the existing tests if useful. Do not replace monetary arithmetic with float.",
      "contentRedacted": false,
      "publishedContentSha256": "b70f397a0ce904320147c810719cfc7bc623ce182820f58b92dccdfacd2924c7"
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/sandbox.sb",
      "sha256": "30fc51bcfa69883d9155aa6ff865290c65bb16aa31a3a23956e9d3c8fe798cce",
      "bytes": 753
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/stderr.log",
      "sha256": "68dc25506cebfd4d87553dcadd637c22c2be9c6c3964b3e4ed36667ec3c1f600",
      "bytes": 5356
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/stdout.log",
      "sha256": "e73989e0142568b84ab05826e7dc6456d64381983f55bc67304d340dff4ff2da",
      "bytes": 5963
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/workspace/README.md",
      "sha256": "b8c8d2a4e51ac441528ef18987f81f996c60ec1367a08f0d4ee6c0e68c3f3fe5",
      "bytes": 91
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/workspace/cart.py",
      "sha256": "07a60e4171c13cc08f897e1c1e947049ba4219cdc50d640759731e298518955f",
      "bytes": 373,
      "content": "from decimal import Decimal, localcontext\n\nfrom invoice import line_total\n\ndef total(lines):\n    result = Decimal('0.00')\n    with localcontext() as context:\n        for price, qty in lines:\n            amount = line_total(price, qty)\n            context.prec = max(context.prec, result.adjusted() + 4, amount.adjusted() + 4)\n            result += amount\n    return result\n",
      "contentRedacted": false,
      "publishedContentSha256": "07a60e4171c13cc08f897e1c1e947049ba4219cdc50d640759731e298518955f"
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/workspace/invoice.py",
      "sha256": "e352ed50de0bf49c480562ec7321c17ee553f722dad2d471d05f069c6f5adcd2",
      "bytes": 694,
      "content": "from decimal import Decimal, ROUND_HALF_UP, localcontext\n\ndef line_total(price, quantity):\n    if isinstance(quantity, bool) or not isinstance(quantity, int) or quantity < 0:\n        raise ValueError('quantity must be an int >= 0')\n    amount = Decimal(str(price))\n    with localcontext() as context:\n        # Allow both exact multiplication and quantizing large amounts to cents.\n        quantity_digits = len(str(quantity))\n        context.prec = max(\n            context.prec,\n            len(amount.as_tuple().digits) + quantity_digits + 2,\n            amount.adjusted() + quantity_digits + 4,\n        )\n        return (amount * quantity).quantize(Decimal('0.01'), rounding=ROUND_HALF_UP)\n",
      "contentRedacted": false,
      "publishedContentSha256": "e352ed50de0bf49c480562ec7321c17ee553f722dad2d471d05f069c6f5adcd2"
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/codex-extra-regression-repair/workspace/test_existing.py",
      "sha256": "fd05d4cf9520d0be75d3629fe64c0365c5848eed199c3d19ddacf18542cf42fb",
      "bytes": 262,
      "content": "import unittest\nfrom decimal import Decimal\nfrom invoice import line_total\n\nclass ExistingTests(unittest.TestCase):\n    def test_whole_amount(self):\n        self.assertEqual(line_total('2.00', 2), Decimal('4.00'))\n\nif __name__ == '__main__':\n    unittest.main()\n",
      "contentRedacted": false,
      "publishedContentSha256": "fd05d4cf9520d0be75d3629fe64c0365c5848eed199c3d19ddacf18542cf42fb"
    }
  ],
  "disclosure": "Synthetic inputs and observed execution. Original-byte hashes retained; selected source and oracle outputs reproduced. Private provider transcripts remain local."
}