Files
cdsl-cad/backend/live-evals/20260831T085924Z/report.json
T
2026-09-01 14:10:23 +08:00

386 lines
13 KiB
JSON

{
"status": "failed",
"author": {
"provider": "openai",
"model": "gpt-5.4-mini"
},
"reviewer": {
"provider": "openai",
"model": "gpt-5.5"
},
"author_capability": {
"schema_hash": "fb899a5de2f869b648df88a32d04a5eee92ae3d193ab23ee641892d6b59cd8e7",
"tool_count": 9,
"supported": true,
"failures": [],
"usage": {
"prompt_tokens": 41789,
"completion_tokens": 1247
},
"mode": "openai_compatible_tool_calling"
},
"reviewer_capability": {
"schema_hash": "fb899a5de2f869b648df88a32d04a5eee92ae3d193ab23ee641892d6b59cd8e7",
"tool_count": 9,
"supported": true,
"failures": [],
"usage": {
"prompt_tokens": 41789,
"completion_tokens": 538
},
"mode": "openai_compatible_tool_calling"
},
"results": [
{
"scenario": "rectangular_plate",
"repetition": 1,
"success": false,
"started_at": "2026-08-31T09:00:56.612675+00:00",
"finished_at": "2026-08-31T09:02:11.406524+00:00",
"terminal": {
"taskId": "cad_ccc4ec062035",
"lifecycle": "failed",
"code": "FAILED_AUTHOR_FORMAT",
"message": "Author repeatedly failed the same canonical schema."
},
"projection": {
"schema_version": "3.0",
"task_id": "cad_ccc4ec062035",
"phase": "FAILED",
"lifecycle": "failed",
"state_version": 4,
"active_revision": "",
"current_revision": "",
"published_revision": "",
"pending_action": null,
"active_candidate_id": "",
"repair_required": false,
"last_error": "FAILED_AUTHOR_FORMAT",
"action_ledger_summary": [
{
"sequence": 1,
"at": "2026-08-31 09:01:04",
"draft_revision": 1,
"event": "requirements_draft_updated",
"item_count": 1
},
{
"sequence": 2,
"at": "2026-08-31 09:01:30",
"draft_revision": 2,
"event": "requirements_draft_patched",
"item_count": 1
},
{
"sequence": 3,
"at": "2026-08-31 09:01:53",
"draft_revision": 3,
"event": "requirements_draft_patched",
"item_count": 1
},
{
"sequence": 4,
"at": "2026-08-31 09:02:11",
"event": "failed_author_format",
"field_errors": [
{
"message": "Field required",
"path": "/patches/0/target_draft_id"
},
{
"message": "Extra inputs are not permitted",
"path": "/target_draft_id"
}
],
"tool": "patch_requirements_draft"
}
],
"revisions": []
},
"usage": {
"calls": 9,
"prompt_tokens": 55142,
"completion_tokens": 2943,
"context_chars": 15133,
"records": [
{
"completion_tokens": 385,
"context_chars": 836,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 5910,
"provider_id": "openai",
"raw_arguments_hash": "c901d4c9931742387dfdd444508e5988a292648298da2b185857baea5349f213",
"tool": "submit_requirements_draft_batch",
"total_tokens": 6295
},
{
"completion_tokens": 278,
"context_chars": 1518,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6098,
"provider_id": "openai",
"raw_arguments_hash": "54f21fc272547493351d211b86a1e4aeb420946bf7b65a56433f9096b9cf9f5c",
"tool": "submit_requirements_draft_batch",
"total_tokens": 6376
},
{
"completion_tokens": 303,
"context_chars": 1919,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6178,
"provider_id": "openai",
"raw_arguments_hash": "23f239dab50941a8e43c1d717b7791115146edab9b9718d1478c9ae012fb1782",
"tool": "patch_requirements_draft",
"total_tokens": 6481
},
{
"completion_tokens": 252,
"context_chars": 1993,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6202,
"provider_id": "openai",
"raw_arguments_hash": "33da3f40fbd2ce0cb9dad052f8d4f140dea658fbb0fdb475d3625c347621f4db",
"tool": "patch_requirements_draft",
"total_tokens": 6454
},
{
"completion_tokens": 291,
"context_chars": 1518,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6098,
"provider_id": "openai",
"raw_arguments_hash": "6642d29cd838aaa3f6709e8f2a5edc1920e8c3caf9239e433ed66c321bb3af2b",
"tool": "submit_requirements_draft_batch",
"total_tokens": 6389
},
{
"completion_tokens": 303,
"context_chars": 1919,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6178,
"provider_id": "openai",
"raw_arguments_hash": "232052bc3c6da6582f57a81f4061433188d4838c3f938ad1c7fff7c66049c653",
"tool": "patch_requirements_draft",
"total_tokens": 6481
},
{
"completion_tokens": 334,
"context_chars": 1993,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6202,
"provider_id": "openai",
"raw_arguments_hash": "05d75ea8566bc896be811b0b2801ec3665ded0db0271a053e0ac6bb4be6cfd59",
"tool": "patch_requirements_draft",
"total_tokens": 6536
},
{
"completion_tokens": 398,
"context_chars": 1518,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6098,
"provider_id": "openai",
"raw_arguments_hash": "7d6b3e0925d86ca6fbec794861924a8df0f4b51f56d074d558b864a43a2bc67e",
"tool": "submit_requirements_draft_batch",
"total_tokens": 6496
},
{
"completion_tokens": 399,
"context_chars": 1919,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6178,
"provider_id": "openai",
"raw_arguments_hash": "1d13a8e2e056ed9b1b49885a9ab05599f58f51e4cecbd22a0640f3dc974b4872",
"tool": "patch_requirements_draft",
"total_tokens": 6577
}
]
},
"checks": {
"expected_terminal_phase": false,
"token_budget": true,
"call_budget": true,
"author_turn_budget": true,
"reviewer_turn_budget": true,
"required_claims": false,
"required_operations": false,
"required_operation_alternative": true,
"immutable_artifacts": false,
"unique_revisions": true,
"raw_argument_audit": true,
"no_schema_or_decision_rejections": false
},
"rejection_codes": [
"AUTHOR_FORMAT_INVALID",
"AUTHOR_FORMAT_INVALID",
"AUTHOR_FORMAT_INVALID",
"AUTHOR_FORMAT_INVALID",
"AUTHOR_FORMAT_INVALID",
"AUTHOR_FORMAT_INVALID"
],
"artifact_root": "/Users/lk/Projects/cdsl-cad/backend/live-evals/20260831T085924Z/artifacts/cad_ccc4ec062035"
},
{
"scenario": "simple_flange",
"repetition": 1,
"success": false,
"started_at": "2026-08-31T09:02:11.409883+00:00",
"finished_at": "2026-08-31T09:03:23.606881+00:00",
"terminal": {
"taskId": "cad_1a1c78be107f",
"lifecycle": "failed",
"code": "FAILED_AUTHOR_FORMAT",
"message": "Author repeatedly failed the same canonical schema."
},
"projection": {
"schema_version": "3.0",
"task_id": "cad_1a1c78be107f",
"phase": "FAILED",
"lifecycle": "failed",
"state_version": 3,
"active_revision": "",
"current_revision": "",
"published_revision": "",
"pending_action": null,
"active_candidate_id": "",
"repair_required": false,
"last_error": "FAILED_AUTHOR_FORMAT",
"action_ledger_summary": [
{
"sequence": 5,
"at": "2026-08-31 09:02:22",
"draft_revision": 1,
"event": "requirements_draft_updated",
"item_count": 1
},
{
"sequence": 6,
"at": "2026-08-31 09:03:07",
"draft_revision": 2,
"event": "requirements_draft_patched",
"item_count": 1
},
{
"sequence": 7,
"at": "2026-08-31 09:03:23",
"event": "failed_author_format",
"field_errors": [
{
"message": "Field required",
"path": "/patches/0/target_draft_id"
},
{
"message": "Extra inputs are not permitted",
"path": "/target_draft_id"
}
],
"tool": "patch_requirements_draft"
}
],
"revisions": []
},
"usage": {
"calls": 7,
"prompt_tokens": 43222,
"completion_tokens": 3167,
"context_chars": 13018,
"records": [
{
"completion_tokens": 412,
"context_chars": 892,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 5918,
"provider_id": "openai",
"raw_arguments_hash": "5030e50bc35c2569181a2866da0ab2f10a7b23a88d6ef6e134c92f9976b6d949",
"tool": "submit_requirements_draft_batch",
"total_tokens": 6330
},
{
"completion_tokens": 366,
"context_chars": 1729,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6156,
"provider_id": "openai",
"raw_arguments_hash": "29b6b77e6f42a319a373f7050e291647a60f4fd20940775f09411ac63140170d",
"tool": "submit_requirements_draft_batch",
"total_tokens": 6522
},
{
"completion_tokens": 329,
"context_chars": 2130,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6236,
"provider_id": "openai",
"raw_arguments_hash": "be0d9112723c27fbae5139e29d93a5a88b0dd6b7a666e041816db9e515e40a86",
"tool": "patch_requirements_draft",
"total_tokens": 6565
},
{
"completion_tokens": 751,
"context_chars": 2204,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6260,
"provider_id": "openai",
"raw_arguments_hash": "83f8db722fbbfebf45ad2a39588e9ea3c5d9281803287cf4eef4c80c1070c52c",
"tool": "patch_requirements_draft",
"total_tokens": 7011
},
{
"completion_tokens": 638,
"context_chars": 2204,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6260,
"provider_id": "openai",
"raw_arguments_hash": "7dffc30397b3bb660a206bc55945db9a643e2221c0c83e1fb06de2fe880163fa",
"tool": "patch_requirements_draft",
"total_tokens": 6898
},
{
"completion_tokens": 340,
"context_chars": 1729,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6156,
"provider_id": "openai",
"raw_arguments_hash": "29b6b77e6f42a319a373f7050e291647a60f4fd20940775f09411ac63140170d",
"tool": "submit_requirements_draft_batch",
"total_tokens": 6496
},
{
"completion_tokens": 331,
"context_chars": 2130,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6236,
"provider_id": "openai",
"raw_arguments_hash": "6660f5b34c43201605c0b9b2d948bd928770f3cc185991687703193c61566ec9",
"tool": "patch_requirements_draft",
"total_tokens": 6567
}
]
},
"checks": {
"expected_terminal_phase": false,
"token_budget": true,
"call_budget": true,
"author_turn_budget": true,
"reviewer_turn_budget": true,
"required_claims": false,
"required_operations": false,
"required_operation_alternative": true,
"immutable_artifacts": false,
"unique_revisions": true,
"raw_argument_audit": true,
"no_schema_or_decision_rejections": false
},
"rejection_codes": [
"AUTHOR_FORMAT_INVALID",
"AUTHOR_FORMAT_INVALID",
"AUTHOR_FORMAT_INVALID",
"AUTHOR_FORMAT_INVALID",
"AUTHOR_FORMAT_INVALID"
],
"artifact_root": "/Users/lk/Projects/cdsl-cad/backend/live-evals/20260831T085924Z/artifacts/cad_1a1c78be107f"
}
],
"suite": "smoke",
"require_live": true,
"report_root": "/Users/lk/Projects/cdsl-cad/backend/live-evals/20260831T085924Z"
}