gensokyo-devandKaoru0822
c23e161629
Share compiled schema conversion while recording all four built-in catalog profiles. Preserve canonical hashes, output validation and custom-budget fallback. Refs #723. Co-authored-by: Kaoru0822 <208281322+kaoru0822-kitauji@users.noreply.github.com>
809 lines
31 KiB
JSON
809 lines
31 KiB
JSON
{
|
|
"issue": 723,
|
|
"source_base": "d532b18123916a9e88f98c7880ba4f984e23d715",
|
|
"runtimes": {
|
|
"baseline-runtime-d532b1812": {
|
|
"source_revision": "d532b18123916a9e88f98c7880ba4f984e23d715",
|
|
"archive_sha256": "79832e2571112400892ec1aae429b030de38c8c672a8acfa6680a2568362fed3",
|
|
"manifest_sha256": "89cbc1688e4702a47b9a5cc66ac38d18de46b74eb550e44227f0ec2119c5e7af"
|
|
},
|
|
"candidate-runtime-1": {
|
|
"source_base": "d532b18123916a9e88f98c7880ba4f984e23d715",
|
|
"patch_sha256": "cfb0eee1ee7f3ceecff7cb77ac63e569dfcfc020e4252cb11869644628f4f743",
|
|
"archive_sha256": "7e4275b1db784545335a4eab373a3db11b1eb5cc6bac7611521a698b0cffce56",
|
|
"manifest_sha256": "356b3baa153650b330cea6bc7e28163565f3201b8d80aa2d6bb69f66faa8705a"
|
|
}
|
|
},
|
|
"node": "24.18.0",
|
|
"npm": "11.16.0",
|
|
"mcp_sdk": "2.3.1",
|
|
"client": "Codex app-server 0.163.0",
|
|
"negotiated_protocol": "2025-06-18",
|
|
"model_turns_started": 0,
|
|
"actual_model_usage": null,
|
|
"task": "Complete 209-tool discovery, then verify no target open, no cleanup pending, no bound provider, package version and tool count twice.",
|
|
"run_order": [
|
|
"baseline-d532-1",
|
|
"candidate-1-1 (reversed profiles)",
|
|
"baseline-d532-2 (reversed profiles)",
|
|
"candidate-1-2"
|
|
],
|
|
"resource_bounds": {
|
|
"cpu_affinity": [15],
|
|
"cpu_quota_percent": 100,
|
|
"hard_memory_gib": 4,
|
|
"early_stop_gib": 3,
|
|
"node_heap_gib": 2,
|
|
"swap": 0,
|
|
"task_ceiling": 192,
|
|
"accounting": "Kernel cgroup process-family peak including file cache, Codex, controller and server; CPU counters at payload exit with only observer alive. Final whole-scope counters unavailable.",
|
|
"oom_events": 0,
|
|
"verified_payload_cleanup_recordings": 16
|
|
},
|
|
"tiktoken": "0.12.0",
|
|
"encoding": "cl100k_base",
|
|
"token_proxy_limit": "Input-facing catalog + RPC params + text result serialization; no model turn/usage, schema injection, prompts or reasoning measured. Identity paths/timestamps cause small reply token differences.",
|
|
"scripts_sha256": {
|
|
"codex-host-probe.py": "d0da9b39e0343b5e84f05328f56df3e82b712fd3fb32ee375cf3735a448fb80e",
|
|
"digest-probe.mjs": "e2f8950014bb40806aef147196ca5d432c6400941d3901e1c0a0e78ce5266bcf",
|
|
"build-digest-probe.mjs": "46060b75c014ca39f9efe317a0445a0220ea7166a08dbdacf04706670e958e6d",
|
|
"run-cohort.py": "1cc5db864b4fe1b2d657ddfe72d468d9e72bddf5118b0bf10c9fce344ec45e9c"
|
|
},
|
|
"profile_digest_oracle": {
|
|
"full": {
|
|
"full": {
|
|
"tools_sha256": "c3e7950408c9005ae3ae0633966609bf1c0f72a36392bc0e0c0fdfe9ea584ae6",
|
|
"prompts_sha256": "7152916ab48bd2a9677a5ad9cdbde0a51dc53c113934952ad096e93f5f18aa7e",
|
|
"combined_sha256": "e0b2715af8312fb565f14d75296adc947b94252e53643f08d831ca8fc8aee941"
|
|
},
|
|
"off": {
|
|
"tools_sha256": "5ad649af53472a1d789b26997cb26a4e8d5d1f2a97c587ad1a6b16030ef5d764",
|
|
"prompts_sha256": "7152916ab48bd2a9677a5ad9cdbde0a51dc53c113934952ad096e93f5f18aa7e",
|
|
"combined_sha256": "cd84cff526daa360e30e299002e7d14427fde9c2b00f59f622fccf89ff2535c7"
|
|
}
|
|
},
|
|
"compact": {
|
|
"full": {
|
|
"tools_sha256": "774c1b7c2c2f57beea9bed62171f046da0f17f2b3bbd7900a90855ecdd26ba79",
|
|
"prompts_sha256": "7152916ab48bd2a9677a5ad9cdbde0a51dc53c113934952ad096e93f5f18aa7e",
|
|
"combined_sha256": "2cbc9ae351fe4f1204a4c2b5e2e92fff8a2ecd87bc14c7ef068d6691c17dff15"
|
|
},
|
|
"off": {
|
|
"tools_sha256": "efe7a17bad5617171162414df88529ff54807b870b5d510e2838ce18e2ba8bd8",
|
|
"prompts_sha256": "7152916ab48bd2a9677a5ad9cdbde0a51dc53c113934952ad096e93f5f18aa7e",
|
|
"combined_sha256": "07d379697201b70d45579d0811e137a0df088be88498fc9fae6806adba26c976"
|
|
}
|
|
}
|
|
},
|
|
"recordings": [
|
|
{
|
|
"recording": "baseline-d532-1-full-full",
|
|
"verified_complete": true,
|
|
"profiles": ["full", "full"],
|
|
"catalog_tools": 209,
|
|
"catalog_bytes": 3322571,
|
|
"catalog_semantic_sha256": "799cc997a9573b5c3dfce3ac22e1fb5eb1637e81b489392f37cc81767b334359",
|
|
"discovery_ms": 5074.2612921167165,
|
|
"first_status_ms": 1699.8079400509596,
|
|
"warm_status_ms": 981.8687399383634,
|
|
"wall_seconds": 8.919,
|
|
"cpu_seconds": 8.480822,
|
|
"kernel_family_peak_bytes": 802246656,
|
|
"token_proxy": {
|
|
"input_facing_catalog": 187005,
|
|
"arguments": 86,
|
|
"text_replies": 22702,
|
|
"combined": 209793,
|
|
"complete_discovered_catalog": 845872
|
|
},
|
|
"raw_sha256": {
|
|
"report.json": "f36f889ab241f704977241443ad55176de6607805f6c82013a287e968dbb428d",
|
|
"catalog.json": "ad5dcfb828d4524f97b494884deed42389ab3e29e2f7839a12b5e95bd1c50ff1",
|
|
"transcript.json": "7d019f379100f45ff1c3f55e66d79e233f7b736886be1823da11e4b843d5fdc1",
|
|
"baseline-d532-1-full-full.log": "3709d55fe5412f49ee880bfee4be818711c977b026ad3133a3f8abb91111f9a7"
|
|
}
|
|
},
|
|
{
|
|
"recording": "baseline-d532-2-full-full",
|
|
"verified_complete": true,
|
|
"profiles": ["full", "full"],
|
|
"catalog_tools": 209,
|
|
"catalog_bytes": 3322571,
|
|
"catalog_semantic_sha256": "799cc997a9573b5c3dfce3ac22e1fb5eb1637e81b489392f37cc81767b334359",
|
|
"discovery_ms": 5847.158181015402,
|
|
"first_status_ms": 1123.013640055433,
|
|
"warm_status_ms": 911.4040599670261,
|
|
"wall_seconds": 8.905,
|
|
"cpu_seconds": 8.452489,
|
|
"kernel_family_peak_bytes": 803254272,
|
|
"token_proxy": {
|
|
"input_facing_catalog": 187004,
|
|
"arguments": 86,
|
|
"text_replies": 22702,
|
|
"combined": 209792,
|
|
"complete_discovered_catalog": 845872
|
|
},
|
|
"raw_sha256": {
|
|
"report.json": "b85792959c6838010ebd622d1cf213d2eb1cd9562da14fc49a2555bda03a0eee",
|
|
"catalog.json": "f588658a13bc1b559008508954f2ce7f34815b326686b4c167c0c27b521cad1b",
|
|
"transcript.json": "5326f37ffdeae98ad4679cf53f462a41fddd71eb9891040753e716489dcc1b68",
|
|
"baseline-d532-2-full-full.log": "147000d226f6cfb3450af51ade419b20c6b0b7d7409c22e876c98830762b39eb"
|
|
}
|
|
},
|
|
{
|
|
"recording": "baseline-d532-1-compact-full",
|
|
"verified_complete": true,
|
|
"profiles": ["compact", "full"],
|
|
"catalog_tools": 209,
|
|
"catalog_bytes": 2723927,
|
|
"catalog_semantic_sha256": "d18f02920b84d39b90dc488db89a8f403f6ec35139a39a3329375a18078af1ea",
|
|
"discovery_ms": 11450.75011299923,
|
|
"first_status_ms": 4571.937991073355,
|
|
"warm_status_ms": 1836.9912200141698,
|
|
"wall_seconds": 18.99,
|
|
"cpu_seconds": 16.522676,
|
|
"kernel_family_peak_bytes": 808927232,
|
|
"token_proxy": {
|
|
"input_facing_catalog": 43507,
|
|
"arguments": 82,
|
|
"text_replies": 22714,
|
|
"combined": 66303,
|
|
"complete_discovered_catalog": 702182
|
|
},
|
|
"raw_sha256": {
|
|
"report.json": "b27141e7af05b3dd43930c0df30f4872709d8db72cc953478150084e99e032f4",
|
|
"catalog.json": "8ff7b639d3409b52761d483283c674e3670e42169ddfda0c1e2a87a207c4dcbd",
|
|
"transcript.json": "1f91daa9703495ca6519459124e9564f0c96f5542b61563bf10029a1f6337926",
|
|
"baseline-d532-1-compact-full.log": "fb187f2b9255e4fc1ecd2c15f9df3c7f1efc76aaba528589f5eed81e7e80e835"
|
|
}
|
|
},
|
|
{
|
|
"recording": "baseline-d532-2-compact-full",
|
|
"verified_complete": true,
|
|
"profiles": ["compact", "full"],
|
|
"catalog_tools": 209,
|
|
"catalog_bytes": 2723927,
|
|
"catalog_semantic_sha256": "d18f02920b84d39b90dc488db89a8f403f6ec35139a39a3329375a18078af1ea",
|
|
"discovery_ms": 5926.411310909316,
|
|
"first_status_ms": 2899.308810941875,
|
|
"warm_status_ms": 1365.2960499748588,
|
|
"wall_seconds": 11.148,
|
|
"cpu_seconds": 10.613801,
|
|
"kernel_family_peak_bytes": 816377856,
|
|
"token_proxy": {
|
|
"input_facing_catalog": 43507,
|
|
"arguments": 82,
|
|
"text_replies": 22714,
|
|
"combined": 66303,
|
|
"complete_discovered_catalog": 702182
|
|
},
|
|
"raw_sha256": {
|
|
"report.json": "df856c2a486c513923b3c67db30bc60ca28ba3b1d0e3418d28d9d6429eca9366",
|
|
"catalog.json": "407ff33fd8e57f2842b76ce190c2089930add6c9a4f618cc90320045f6a67ed2",
|
|
"transcript.json": "d11dfb4c28c5355e733943e3504d3b4e11f056f0f7625e835e5cdc916a451716",
|
|
"baseline-d532-2-compact-full.log": "0485a806e815ecb91862d37cb9200b7cb0877a3d8c69882e8299100d13769a57"
|
|
}
|
|
},
|
|
{
|
|
"recording": "baseline-d532-1-full-off",
|
|
"verified_complete": true,
|
|
"profiles": ["full", "off"],
|
|
"catalog_tools": 209,
|
|
"catalog_bytes": 832191,
|
|
"catalog_semantic_sha256": "e537c521b5dd0ac369dee706815579d9d7803497a8fc7f7d683fec267042b81c",
|
|
"discovery_ms": 11546.154452953488,
|
|
"first_status_ms": 4629.906311165541,
|
|
"warm_status_ms": 2057.3992100544274,
|
|
"wall_seconds": 19.588,
|
|
"cpu_seconds": 16.963204,
|
|
"kernel_family_peak_bytes": 709857280,
|
|
"token_proxy": {
|
|
"input_facing_catalog": 187005,
|
|
"arguments": 86,
|
|
"text_replies": 22710,
|
|
"combined": 209801,
|
|
"complete_discovered_catalog": 195042
|
|
},
|
|
"raw_sha256": {
|
|
"report.json": "4963559eafd97d6aa72e2265651d9f6d3d0cd3a04965677d6fdedfe0378e7c11",
|
|
"catalog.json": "aef1822a088446790e14373a6b9b1039cda640598d19620fdd776f6d7c49d890",
|
|
"transcript.json": "203f7f00eca7617cdaa8c7fd45fdde3d4502cd29cf36ee0c724181ad3178410b",
|
|
"baseline-d532-1-full-off.log": "84e5a67f6f1e97fb3c429d96d7b3de33e08eddc56b21b453fa2b3abf19dbd15b"
|
|
}
|
|
},
|
|
{
|
|
"recording": "baseline-d532-2-full-off",
|
|
"verified_complete": true,
|
|
"profiles": ["full", "off"],
|
|
"catalog_tools": 209,
|
|
"catalog_bytes": 832191,
|
|
"catalog_semantic_sha256": "e537c521b5dd0ac369dee706815579d9d7803497a8fc7f7d683fec267042b81c",
|
|
"discovery_ms": 5198.2202511280775,
|
|
"first_status_ms": 2726.173529867083,
|
|
"warm_status_ms": 1434.7798710223287,
|
|
"wall_seconds": 10.325,
|
|
"cpu_seconds": 9.75455,
|
|
"kernel_family_peak_bytes": 703328256,
|
|
"token_proxy": {
|
|
"input_facing_catalog": 187005,
|
|
"arguments": 86,
|
|
"text_replies": 22710,
|
|
"combined": 209801,
|
|
"complete_discovered_catalog": 195042
|
|
},
|
|
"raw_sha256": {
|
|
"report.json": "e603556898f606d6804c73a227ef6d809f09f516316255b316e5afb473170ff9",
|
|
"catalog.json": "8200cccc7a0d7bdcf288cab0225269acadbd532aed69da7d7e8a3b4cf34b9c16",
|
|
"transcript.json": "4d2a862a24282cac2581a36ca46fd99474835cba247203556c04015ea02096fe",
|
|
"baseline-d532-2-full-off.log": "90b720806f3f5756408501a1f9d5dd724108647706e38aca10cbc0a35e56a37a"
|
|
}
|
|
},
|
|
{
|
|
"recording": "baseline-d532-1-compact-off",
|
|
"verified_complete": true,
|
|
"profiles": ["compact", "off"],
|
|
"catalog_tools": 209,
|
|
"catalog_bytes": 233547,
|
|
"catalog_semantic_sha256": "0b81bdd81251ddd527a9c9ecbb5e6048e9c8a3be1f6195465a803d206f03fe73",
|
|
"discovery_ms": 12976.338863139972,
|
|
"first_status_ms": 6152.441991027445,
|
|
"warm_status_ms": 3121.8184099998325,
|
|
"wall_seconds": 23.47,
|
|
"cpu_seconds": 19.234341,
|
|
"kernel_family_peak_bytes": 665423872,
|
|
"token_proxy": {
|
|
"input_facing_catalog": 43507,
|
|
"arguments": 82,
|
|
"text_replies": 22690,
|
|
"combined": 66279,
|
|
"complete_discovered_catalog": 51352
|
|
},
|
|
"raw_sha256": {
|
|
"report.json": "c3c21635cf212eba3902ad64f956244ddfe122425fa3b39ec280b5296ebe23c6",
|
|
"catalog.json": "e5a4444ea91913e68705a6bd6afee2b8ea0a43b2f91f8628ada597e57afab5db",
|
|
"transcript.json": "58df156c85a8b1ec325945db08e4e84e8174dfd739f48e4d83b2a3537f7a52b9",
|
|
"baseline-d532-1-compact-off.log": "8d5b2ddcf5b31ad4635269837e217dd2242a91d82ea961d01e2ab3a2076b9e35"
|
|
}
|
|
},
|
|
{
|
|
"recording": "baseline-d532-2-compact-off",
|
|
"verified_complete": true,
|
|
"profiles": ["compact", "off"],
|
|
"catalog_tools": 209,
|
|
"catalog_bytes": 233547,
|
|
"catalog_semantic_sha256": "0b81bdd81251ddd527a9c9ecbb5e6048e9c8a3be1f6195465a803d206f03fe73",
|
|
"discovery_ms": 5907.094460912049,
|
|
"first_status_ms": 2939.780990127474,
|
|
"warm_status_ms": 1040.7937010750175,
|
|
"wall_seconds": 10.742,
|
|
"cpu_seconds": 10.176561,
|
|
"kernel_family_peak_bytes": 667668480,
|
|
"token_proxy": {
|
|
"input_facing_catalog": 43507,
|
|
"arguments": 82,
|
|
"text_replies": 22690,
|
|
"combined": 66279,
|
|
"complete_discovered_catalog": 51352
|
|
},
|
|
"raw_sha256": {
|
|
"report.json": "14aed217fcad50915b211e4ee9894ebf9cb4abd8d4c4a4e007cc6fe2b397d44a",
|
|
"catalog.json": "e55a5b2cba59b09ac4f63831c2af6682d162bf5b05259b3c0f9db8eb2bc67e44",
|
|
"transcript.json": "e9a9ac8b657da35501e577bf1d1e18c89923c5bcdbb1307c8254d512bbdc7b53",
|
|
"baseline-d532-2-compact-off.log": "e7c0c68bd1088f11107a6713df1efb74f01ff3479d1913977fe22935814c912c"
|
|
}
|
|
},
|
|
{
|
|
"recording": "candidate-1-1-full-full",
|
|
"verified_complete": true,
|
|
"profiles": ["full", "full"],
|
|
"catalog_tools": 209,
|
|
"catalog_bytes": 3322571,
|
|
"catalog_semantic_sha256": "799cc997a9573b5c3dfce3ac22e1fb5eb1637e81b489392f37cc81767b334359",
|
|
"discovery_ms": 7035.046560922638,
|
|
"first_status_ms": 1280.1523089874536,
|
|
"warm_status_ms": 962.2103709261864,
|
|
"wall_seconds": 10.309,
|
|
"cpu_seconds": 9.82015,
|
|
"kernel_family_peak_bytes": 806117376,
|
|
"token_proxy": {
|
|
"input_facing_catalog": 187004,
|
|
"arguments": 84,
|
|
"text_replies": 22690,
|
|
"combined": 209778,
|
|
"complete_discovered_catalog": 845872
|
|
},
|
|
"raw_sha256": {
|
|
"report.json": "4e6d4cb20d4b435b47164535930c60902eed44c4efec0665d9eeb2bd0635fd39",
|
|
"catalog.json": "f0700961a2a999671107dc5e5274098a5d97047e834654c7c3c7ee0c420a91b5",
|
|
"transcript.json": "4d0b64606c913325ce83bd797b2286d0e9758f09c526dacfce422756cf5212d8",
|
|
"candidate-1-1-full-full.log": "9d37d44776cfd9f41811e0a608c79ed63257383218e01f602317c2dc5a9de10f"
|
|
}
|
|
},
|
|
{
|
|
"recording": "candidate-1-2-full-full",
|
|
"verified_complete": true,
|
|
"profiles": ["full", "full"],
|
|
"catalog_tools": 209,
|
|
"catalog_bytes": 3322571,
|
|
"catalog_semantic_sha256": "799cc997a9573b5c3dfce3ac22e1fb5eb1637e81b489392f37cc81767b334359",
|
|
"discovery_ms": 5431.887832004577,
|
|
"first_status_ms": 1290.1408409234136,
|
|
"warm_status_ms": 1058.6841800250113,
|
|
"wall_seconds": 8.819,
|
|
"cpu_seconds": 8.06667,
|
|
"kernel_family_peak_bytes": 801071104,
|
|
"token_proxy": {
|
|
"input_facing_catalog": 187004,
|
|
"arguments": 78,
|
|
"text_replies": 22690,
|
|
"combined": 209772,
|
|
"complete_discovered_catalog": 845872
|
|
},
|
|
"raw_sha256": {
|
|
"report.json": "e2853484e579376a8edb168f84ad2add2ffdd30ce78e123b2dfcdd17d524344e",
|
|
"catalog.json": "5c918daeb3f8b10625927a87f3f95367a2a2823cc7c098aa8ad3d8d5b47aa6ad",
|
|
"transcript.json": "1e22387d227134741eac842b8000c5bb73a42e2368c3c53e21bc1d45c17ed819",
|
|
"candidate-1-2-full-full.log": "877a3bdfe67875bb5ab1981850da2fe519428708184eef55a28c5fa18bb727ea"
|
|
}
|
|
},
|
|
{
|
|
"recording": "candidate-1-1-compact-full",
|
|
"verified_complete": true,
|
|
"profiles": ["compact", "full"],
|
|
"catalog_tools": 209,
|
|
"catalog_bytes": 2723927,
|
|
"catalog_semantic_sha256": "d18f02920b84d39b90dc488db89a8f403f6ec35139a39a3329375a18078af1ea",
|
|
"discovery_ms": 5128.74938105233,
|
|
"first_status_ms": 1351.6553710214794,
|
|
"warm_status_ms": 1321.080510970205,
|
|
"wall_seconds": 8.815,
|
|
"cpu_seconds": 8.478922,
|
|
"kernel_family_peak_bytes": 781271040,
|
|
"token_proxy": {
|
|
"input_facing_catalog": 43507,
|
|
"arguments": 80,
|
|
"text_replies": 22702,
|
|
"combined": 66289,
|
|
"complete_discovered_catalog": 702182
|
|
},
|
|
"raw_sha256": {
|
|
"report.json": "ee38fcca794a959271af78a6c65377f411fdbe3b7b97068835b36586ec804ddf",
|
|
"catalog.json": "db8f1cbaa97c09672537d87ac48457bdb599ae439bca2d8a9216a032edbbdd88",
|
|
"transcript.json": "22d7a8b00909ad82e40ca7beb21d793bff61dacd88ba0d345ef82eeacf177516",
|
|
"candidate-1-1-compact-full.log": "ae17a479c9f094e522e15b58617302bf098464a4f55cea21cc913d18ba02afd2"
|
|
}
|
|
},
|
|
{
|
|
"recording": "candidate-1-2-compact-full",
|
|
"verified_complete": true,
|
|
"profiles": ["compact", "full"],
|
|
"catalog_tools": 209,
|
|
"catalog_bytes": 2723927,
|
|
"catalog_semantic_sha256": "d18f02920b84d39b90dc488db89a8f403f6ec35139a39a3329375a18078af1ea",
|
|
"discovery_ms": 5686.0627620480955,
|
|
"first_status_ms": 1266.2633110303432,
|
|
"warm_status_ms": 1025.3590901847929,
|
|
"wall_seconds": 8.893,
|
|
"cpu_seconds": 8.191988,
|
|
"kernel_family_peak_bytes": 780988416,
|
|
"token_proxy": {
|
|
"input_facing_catalog": 43507,
|
|
"arguments": 82,
|
|
"text_replies": 22702,
|
|
"combined": 66291,
|
|
"complete_discovered_catalog": 702182
|
|
},
|
|
"raw_sha256": {
|
|
"report.json": "1ae8930195ef5d5b6d5118ffaad3d0d2bedaaf114b3887a28357504be023821f",
|
|
"catalog.json": "7e65fcac2fe7257a2f1c058f2b9ad1200772072fd7cabd4ea315bf4c48307303",
|
|
"transcript.json": "5a0220b5fe36ae31af91f9eb7f8006290fcf790c774096b838f7287c472e94c1",
|
|
"candidate-1-2-compact-full.log": "2e4baa5d7e627be27f87739d03f2a4e762ae1675a63e14cc7b5adf883c24192c"
|
|
}
|
|
},
|
|
{
|
|
"recording": "candidate-1-1-full-off",
|
|
"verified_complete": true,
|
|
"profiles": ["full", "off"],
|
|
"catalog_tools": 209,
|
|
"catalog_bytes": 832191,
|
|
"catalog_semantic_sha256": "e537c521b5dd0ac369dee706815579d9d7803497a8fc7f7d683fec267042b81c",
|
|
"discovery_ms": 5637.611771002412,
|
|
"first_status_ms": 1178.2668600790203,
|
|
"warm_status_ms": 1036.6536700166762,
|
|
"wall_seconds": 8.791,
|
|
"cpu_seconds": 8.201682,
|
|
"kernel_family_peak_bytes": 656801792,
|
|
"token_proxy": {
|
|
"input_facing_catalog": 187005,
|
|
"arguments": 82,
|
|
"text_replies": 22698,
|
|
"combined": 209785,
|
|
"complete_discovered_catalog": 195042
|
|
},
|
|
"raw_sha256": {
|
|
"report.json": "cd706d0ab9a3dd9662ec73ecc45deefdbd40a4a42cd9ee2735be0f8798cc1933",
|
|
"catalog.json": "9d1fd380d490edf4b6403e3daed015b72bc654b96dfb06dbe7d776e6a6146d76",
|
|
"transcript.json": "0e3f0c66a2b638729d341af8d8a379576d267b53c1dcea5563e1afa4f9245cc8",
|
|
"candidate-1-1-full-off.log": "1baf3976c762923680cd17cbef36778b3982e90005eb34bb8ec3ce8030ebd3fb"
|
|
}
|
|
},
|
|
{
|
|
"recording": "candidate-1-2-full-off",
|
|
"verified_complete": true,
|
|
"profiles": ["full", "off"],
|
|
"catalog_tools": 209,
|
|
"catalog_bytes": 832191,
|
|
"catalog_semantic_sha256": "e537c521b5dd0ac369dee706815579d9d7803497a8fc7f7d683fec267042b81c",
|
|
"discovery_ms": 5316.555531928316,
|
|
"first_status_ms": 1383.681601146236,
|
|
"warm_status_ms": 1108.902801061049,
|
|
"wall_seconds": 8.698,
|
|
"cpu_seconds": 8.19166,
|
|
"kernel_family_peak_bytes": 662822912,
|
|
"token_proxy": {
|
|
"input_facing_catalog": 187004,
|
|
"arguments": 82,
|
|
"text_replies": 22698,
|
|
"combined": 209784,
|
|
"complete_discovered_catalog": 195042
|
|
},
|
|
"raw_sha256": {
|
|
"report.json": "89c5f579cb8f8a1860b3f605042ffc51db4aaae69f7a22213256f93e42b91672",
|
|
"catalog.json": "0301c40f667a17a9238ad8d9abdbb8e8c7b07a7d0dcc6ae062f5885e0b19dfc7",
|
|
"transcript.json": "7e884d99d3335707f9108a63528fd2830b1c21e7888cfc320852d75924d833cb",
|
|
"candidate-1-2-full-off.log": "ba34788aa92aa60e79e832659c48f304396d1df41f493a509ff474a152fa7c6c"
|
|
}
|
|
},
|
|
{
|
|
"recording": "candidate-1-1-compact-off",
|
|
"verified_complete": true,
|
|
"profiles": ["compact", "off"],
|
|
"catalog_tools": 209,
|
|
"catalog_bytes": 233547,
|
|
"catalog_semantic_sha256": "0b81bdd81251ddd527a9c9ecbb5e6048e9c8a3be1f6195465a803d206f03fe73",
|
|
"discovery_ms": 4922.329220920801,
|
|
"first_status_ms": 813.2044598460197,
|
|
"warm_status_ms": 1415.0590698700398,
|
|
"wall_seconds": 7.982,
|
|
"cpu_seconds": 7.401054,
|
|
"kernel_family_peak_bytes": 629260288,
|
|
"token_proxy": {
|
|
"input_facing_catalog": 43507,
|
|
"arguments": 80,
|
|
"text_replies": 22678,
|
|
"combined": 66265,
|
|
"complete_discovered_catalog": 51352
|
|
},
|
|
"raw_sha256": {
|
|
"report.json": "89ae06a9c0e8f4c9b85aa711fc6f71204798c990e582f5702713f78612c4fce2",
|
|
"catalog.json": "d6a231b2babd7885d58f9712881f96fa9dc2b763cc7efd802bc5676dd71fe28a",
|
|
"transcript.json": "65fbdb48a9bc2aa65a815ea405d1210e298f908ddbab5e60f815c4e01a0f0e98",
|
|
"candidate-1-1-compact-off.log": "76501971a3698b13435b96647b59ee092689397012d453e62c821e4ecf40e988"
|
|
}
|
|
},
|
|
{
|
|
"recording": "candidate-1-2-compact-off",
|
|
"verified_complete": true,
|
|
"profiles": ["compact", "off"],
|
|
"catalog_tools": 209,
|
|
"catalog_bytes": 233547,
|
|
"catalog_semantic_sha256": "0b81bdd81251ddd527a9c9ecbb5e6048e9c8a3be1f6195465a803d206f03fe73",
|
|
"discovery_ms": 4839.713891968131,
|
|
"first_status_ms": 1174.165149917826,
|
|
"warm_status_ms": 1409.5584598835558,
|
|
"wall_seconds": 8.289,
|
|
"cpu_seconds": 7.764389,
|
|
"kernel_family_peak_bytes": 627363840,
|
|
"token_proxy": {
|
|
"input_facing_catalog": 43507,
|
|
"arguments": 80,
|
|
"text_replies": 22678,
|
|
"combined": 66265,
|
|
"complete_discovered_catalog": 51352
|
|
},
|
|
"raw_sha256": {
|
|
"report.json": "a4ef1220b73c39421ccfc22c2102c613fd0a33900cbc8683dad1dc3bc034cf5a",
|
|
"catalog.json": "39ae1d7385fe8684f037e354ec2ab1126243fd31667be1b1cc559a9d1e40149c",
|
|
"transcript.json": "d60eae5fdbfb4095f040d54103a6e9b1d2ff8b8f64126976d14b0a3e024969f7",
|
|
"candidate-1-2-compact-off.log": "0e195ca040565e29e1f6d25b087a706d0d500ff1752a24bff54ad30704bd6d6b"
|
|
}
|
|
}
|
|
],
|
|
"digest_phases": [
|
|
{
|
|
"recording": "baseline-d532-1-compact-full-digest",
|
|
"input": "compact",
|
|
"output": "full",
|
|
"import_ms": 3827.9035209999997,
|
|
"digest_ms": 2916.01332,
|
|
"digest_cpu_us": {
|
|
"user": 2587474,
|
|
"system": 151462
|
|
},
|
|
"raw_sha256": "28a8e0b5c847978be9d71e4d5ed6953758e266e697b820f45f1d2141861c95e3"
|
|
},
|
|
{
|
|
"recording": "baseline-d532-1-compact-off-digest",
|
|
"input": "compact",
|
|
"output": "off",
|
|
"import_ms": 3873.5359620000004,
|
|
"digest_ms": 3790.342601,
|
|
"digest_cpu_us": {
|
|
"user": 3121749,
|
|
"system": 212457
|
|
},
|
|
"raw_sha256": "b88dd91f159a1763d547f1e26637afaf3a03b28faad3295c2deff8c90e7d7707"
|
|
},
|
|
{
|
|
"recording": "baseline-d532-1-full-full-digest",
|
|
"input": "full",
|
|
"output": "full",
|
|
"import_ms": 1766.4657209999998,
|
|
"digest_ms": 0.06419000000005326,
|
|
"digest_cpu_us": {
|
|
"user": 65,
|
|
"system": 12
|
|
},
|
|
"raw_sha256": "0f18c0d63def7df73b6183240c49011aca9cd85899a221fdb01908da1d79120f"
|
|
},
|
|
{
|
|
"recording": "baseline-d532-1-full-off-digest",
|
|
"input": "full",
|
|
"output": "off",
|
|
"import_ms": 5377.792571,
|
|
"digest_ms": 3209.784441,
|
|
"digest_cpu_us": {
|
|
"user": 2585090,
|
|
"system": 252879
|
|
},
|
|
"raw_sha256": "2de37e6aac00f37003f619c4b4a5a747637e0b188e9892d5a37a091ae568400a"
|
|
},
|
|
{
|
|
"recording": "baseline-d532-2-compact-full-digest",
|
|
"input": "compact",
|
|
"output": "full",
|
|
"import_ms": 1807.87952,
|
|
"digest_ms": 1512.0799399999999,
|
|
"digest_cpu_us": {
|
|
"user": 1421207,
|
|
"system": 88507
|
|
},
|
|
"raw_sha256": "a330124a6cec474e5337aa5a40c33d31fa0bdce304554c1c68f79954bf5cdd33"
|
|
},
|
|
{
|
|
"recording": "baseline-d532-2-compact-off-digest",
|
|
"input": "compact",
|
|
"output": "off",
|
|
"import_ms": 1854.8867300000002,
|
|
"digest_ms": 1402.711411,
|
|
"digest_cpu_us": {
|
|
"user": 1284446,
|
|
"system": 116933
|
|
},
|
|
"raw_sha256": "02f19af3b9836796d85f81effd5e86bab0aa4978eeae0c59b8c35e57d03dd17b"
|
|
},
|
|
{
|
|
"recording": "baseline-d532-2-full-full-digest",
|
|
"input": "full",
|
|
"output": "full",
|
|
"import_ms": 1572.84796,
|
|
"digest_ms": 0.0849799999998595,
|
|
"digest_cpu_us": {
|
|
"user": 85,
|
|
"system": 15
|
|
},
|
|
"raw_sha256": "a9fca1656b8763f5ef305548eceb7bbd27cdeb426c6d6934a891c82823a54f31"
|
|
},
|
|
{
|
|
"recording": "baseline-d532-2-full-off-digest",
|
|
"input": "full",
|
|
"output": "off",
|
|
"import_ms": 1814.550491,
|
|
"digest_ms": 1280.248369,
|
|
"digest_cpu_us": {
|
|
"user": 1166910,
|
|
"system": 108321
|
|
},
|
|
"raw_sha256": "350a276f6f408de517db2b19af7cd7490c243b71338854d4acb54418f1a00b64"
|
|
},
|
|
{
|
|
"recording": "candidate-1-1-compact-full-digest",
|
|
"input": "compact",
|
|
"output": "full",
|
|
"import_ms": 1760.5044,
|
|
"digest_ms": 0.08051999999997861,
|
|
"digest_cpu_us": {
|
|
"user": 76,
|
|
"system": 14
|
|
},
|
|
"raw_sha256": "3ebc2dac6cff195b54918e5afaedc0b73fe9f1a63ef12ebb253d9a1aca683c8c"
|
|
},
|
|
{
|
|
"recording": "candidate-1-1-compact-off-digest",
|
|
"input": "compact",
|
|
"output": "off",
|
|
"import_ms": 1922.20413,
|
|
"digest_ms": 0.07567999999992026,
|
|
"digest_cpu_us": {
|
|
"user": 75,
|
|
"system": 11
|
|
},
|
|
"raw_sha256": "d5916e1e5760f9c7ce5a42ee8b7635477464a5a7cb7d379309666357b6fbe35f"
|
|
},
|
|
{
|
|
"recording": "candidate-1-1-full-full-digest",
|
|
"input": "full",
|
|
"output": "full",
|
|
"import_ms": 1777.54332,
|
|
"digest_ms": 0.08724099999994905,
|
|
"digest_cpu_us": {
|
|
"user": 93,
|
|
"system": 13
|
|
},
|
|
"raw_sha256": "ed68a8c592c9556b7a279a90b7324d5549f9dc197fe593d27c55c5de24827481"
|
|
},
|
|
{
|
|
"recording": "candidate-1-1-full-off-digest",
|
|
"input": "full",
|
|
"output": "off",
|
|
"import_ms": 2062.896171,
|
|
"digest_ms": 0.06988900000033027,
|
|
"digest_cpu_us": {
|
|
"user": 69,
|
|
"system": 12
|
|
},
|
|
"raw_sha256": "166dcdaa064c334dcc882429c2434c0353b0f6d2ccdea986138a2149ccccf10a"
|
|
},
|
|
{
|
|
"recording": "candidate-1-2-compact-full-digest",
|
|
"input": "compact",
|
|
"output": "full",
|
|
"import_ms": 1813.35421,
|
|
"digest_ms": 0.06941099999994549,
|
|
"digest_cpu_us": {
|
|
"user": 73,
|
|
"system": 10
|
|
},
|
|
"raw_sha256": "86497ee4962fd03fcad4f2bcc31e95f5f57d9c4270cc0578d4b43a076ec01b53"
|
|
},
|
|
{
|
|
"recording": "candidate-1-2-compact-off-digest",
|
|
"input": "compact",
|
|
"output": "off",
|
|
"import_ms": 2024.0317209999998,
|
|
"digest_ms": 0.08607000000029075,
|
|
"digest_cpu_us": {
|
|
"user": 89,
|
|
"system": 13
|
|
},
|
|
"raw_sha256": "82b47ed9bc203f1417f5f76de808d493c171f95598934fc9ebbd6f63be43b938"
|
|
},
|
|
{
|
|
"recording": "candidate-1-2-full-full-digest",
|
|
"input": "full",
|
|
"output": "full",
|
|
"import_ms": 1917.6121309999999,
|
|
"digest_ms": 0.07732000000009975,
|
|
"digest_cpu_us": {
|
|
"user": 77,
|
|
"system": 11
|
|
},
|
|
"raw_sha256": "410395499363c55e53fe4d8488b5e83c54170786fe1814389fd3bf4ec317c6ac"
|
|
},
|
|
{
|
|
"recording": "candidate-1-2-full-off-digest",
|
|
"input": "full",
|
|
"output": "off",
|
|
"import_ms": 1715.96008,
|
|
"digest_ms": 0.08907999999996719,
|
|
"digest_cpu_us": {
|
|
"user": 104,
|
|
"system": 15
|
|
},
|
|
"raw_sha256": "f6ea0fffc2974e02bdab1348736533c8dbd677c2311e339a945d760bd9337f9a"
|
|
}
|
|
],
|
|
"build_digest_phase": [
|
|
{
|
|
"phase": "before",
|
|
"operation": "createCatalogDigests",
|
|
"import_ms": 2103.978981,
|
|
"compute_ms": 1640.0212299999998,
|
|
"compute_cpu_us": {
|
|
"user": 1487510,
|
|
"system": 148234
|
|
},
|
|
"raw_sha256": "3a4bfe5e4297e95cf0ab3b832bb34c81edeca02e7f884db7dd5a6944c2ee1674"
|
|
},
|
|
{
|
|
"phase": "after",
|
|
"operation": "createCatalogDigestProfiles",
|
|
"import_ms": 1752.475121,
|
|
"compute_ms": 2379.7143199999996,
|
|
"compute_cpu_us": {
|
|
"user": 2232836,
|
|
"system": 140061
|
|
},
|
|
"raw_sha256": "b25120373c7433457e30ff00925ca2771c0d61ba96828fe89e63be651b4f6e4a"
|
|
}
|
|
],
|
|
"failed_attempts": [
|
|
{
|
|
"recording": "codex-host-full-1",
|
|
"outcome": "Harness expected counts.tools rather than canonical counts.mcp_tools; corrected oracle, not a completed task.",
|
|
"raw_sha256": {
|
|
"report.json": "ac3d531a9dad31e31cb9a5957ed1f5e6415534fda1e52c64bd93a07a36ef7d0f",
|
|
"codex-host-full-1.log": "7c5f13e2976efef59306b9a2fabd2e2031277d3c5aed2b74f4ff2ad80c25a61c"
|
|
}
|
|
},
|
|
{
|
|
"recording": "codex-host-matrix-compact-off-1",
|
|
"outcome": "Functional checks passed but MCP child remained at payload exit. Excluded from qualified cohort. Controller now waits for allocated scope to drain without targeting a PID.",
|
|
"raw_sha256": {
|
|
"report.json": "1cbc9df5aff972538629ed695271f0c3a66fcfe546208b392d09b3b7cf72f97c",
|
|
"codex-host-matrix-compact-off-1.log": "997e13a9ea7b3c5247d5037279fbef1d0dfd238eed0781748cc2ddf3afc706e7"
|
|
}
|
|
}
|
|
],
|
|
"unverified": [
|
|
"Model selection/injection and actual model tokens",
|
|
"Provider analysis, target execution and end-to-end reverse-engineering application quality",
|
|
"Source-mode speedup: absent build records still use dynamic fallback",
|
|
"Other MCP clients/OS compatibility not established by this Codex experiment"
|
|
],
|
|
"validation": [
|
|
{
|
|
"check": "check-profiles",
|
|
"payload_exit_code": 1,
|
|
"owned_cleanup_verified": true,
|
|
"raw_sha256": "04dd1ad8142e61e9c2960ec55a3a7f0421f4c22f5cb5b9702761a0d7f29d3cc4",
|
|
"source_head": "pre-rebase candidate",
|
|
"outcome": "Unused build-only export identified; existing output-policy coverage extended to the profile builder."
|
|
},
|
|
{
|
|
"check": "build-latest",
|
|
"payload_exit_code": 0,
|
|
"owned_cleanup_verified": true,
|
|
"raw_sha256": "f430e38ecb205aa7bde17b48ff7ad97805b18511dbbfee8bdbab605f4f93b86e",
|
|
"source_head": "7ad06628df042a7c8034155a98fc51c35fa0704c",
|
|
"outcome": "passed"
|
|
},
|
|
{
|
|
"check": "tests-latest",
|
|
"payload_exit_code": 0,
|
|
"owned_cleanup_verified": true,
|
|
"raw_sha256": "58b06f5be6ae5c0cb871cba6788b3f84581289a47ffe26aaad8897b606c94e2d",
|
|
"source_head": "7ad06628df042a7c8034155a98fc51c35fa0704c",
|
|
"outcome": "passed"
|
|
},
|
|
{
|
|
"check": "check-latest",
|
|
"payload_exit_code": 0,
|
|
"owned_cleanup_verified": true,
|
|
"raw_sha256": "2373fec55a74515f6e3036d46becd4ed76149d9a70731a64c5dddacaeeb920cc",
|
|
"source_head": "7ad06628df042a7c8034155a98fc51c35fa0704c",
|
|
"outcome": "passed"
|
|
},
|
|
{
|
|
"check": "docs-latest",
|
|
"payload_exit_code": 0,
|
|
"owned_cleanup_verified": true,
|
|
"raw_sha256": "23ef004935364b7a1ad2634610b67c1d739bde4f4c455552a473a05b346920ee",
|
|
"source_head": "7ad06628df042a7c8034155a98fc51c35fa0704c",
|
|
"outcome": "passed"
|
|
}
|
|
],
|
|
"post_rebase_qualification": {
|
|
"head": "7ad06628df042a7c8034155a98fc51c35fa0704c",
|
|
"base": "e84468c0b",
|
|
"client": "Codex app-server 0.163.0",
|
|
"profiles": ["compact", "off"],
|
|
"outcome": "Complete discovery and first/warm status facts pass; allocated scope drains; no model turns. Qualification only; not relabeled as the pinned before/after measurements.",
|
|
"raw_sha256": {
|
|
"final-latest-compact-off/report.json": "ba6f549fe087f55dc48011daf92bd18f4bfc2a32e6185081ab2b7436bf39b1b8",
|
|
"final-latest-compact-off/transcript.json": "461a25723c2a4bb4c18675ffa613e98a8592a4e933054fa8643f1da73fc4686f",
|
|
"final-latest-compact-off.log": "4efc56e69b8f0d1111345bce10d0d0792b510e18901f52d10e538374257e1d61"
|
|
}
|
|
}
|
|
}
|