Files
cdsl-cad/backend/live-evals/20260831T101800Z/report.json
T
2026-09-01 14:10:23 +08:00

1267 lines
43 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": 1288
},
"mode": "openai_compatible_tool_calling"
},
"reviewer_capability": {
"schema_hash": "fb899a5de2f869b648df88a32d04a5eee92ae3d193ab23ee641892d6b59cd8e7",
"tool_count": 9,
"supported": true,
"failures": [],
"usage": {
"prompt_tokens": 41789,
"completion_tokens": 533
},
"mode": "openai_compatible_tool_calling"
},
"results": [
{
"scenario": "rectangular_plate",
"repetition": 1,
"success": true,
"started_at": "2026-08-31T10:19:19.704593+00:00",
"finished_at": "2026-08-31T10:20:17.376308+00:00",
"terminal": {
"taskId": "cad_2039eb170f23",
"lifecycle": "completed",
"revisionId": "rev_001"
},
"projection": {
"schema_version": "3.0",
"task_id": "cad_2039eb170f23",
"phase": "COMPLETED",
"lifecycle": "completed",
"state_version": 9,
"active_revision": "rev_001",
"current_revision": "rev_001",
"published_revision": "rev_001",
"pending_action": null,
"active_candidate_id": "",
"repair_required": false,
"last_error": "",
"action_ledger_summary": [
{
"sequence": 1,
"at": "2026-08-31 10:19:27",
"draft_revision": 1,
"event": "requirements_draft_updated",
"item_count": 1
},
{
"sequence": 2,
"at": "2026-08-31 10:19:31",
"draft_revision": 1,
"event": "requirements_review_requested"
},
{
"sequence": 3,
"at": "2026-08-31 10:19:37",
"contract_hash": "a65da6add55ca1a2db47b72c8afb329ba61239589f3ec7e5d3875c87f90730ae",
"event": "requirements_contract_frozen",
"requirement_count": 1
},
{
"sequence": 4,
"at": "2026-08-31 10:19:45",
"action_id": "act_004",
"atomic_id": "extrude_add_blind",
"contract_hash": "fa9ffc9fd2141453fa18a389813dde2e2ef3fb4c7ef5d05e61d51d614309e95f",
"event": "proposed",
"expected_change": "Create one connected solid rectangular plate with overall dimensions 80 mm by 50 mm by 8 mm in the current model.",
"intent": "Form the required base plate as a single connected part matching the specified millimetre dimensions.",
"requirement_ids": [
"req_001"
],
"working_head": "cad_2039eb170f23:root:v3"
},
{
"sequence": 5,
"at": "2026-08-31 10:19:57",
"action_id": "act_004",
"candidate_id": "candidate_86a9ab7071766aa8",
"event": "candidate_building",
"fragment_hash": "34b4567ab12fdc5a54b67ab39a791a6a0550daeb157fe6e40abd7f67726046bb"
},
{
"sequence": 6,
"at": "2026-08-31 10:19:58",
"action_id": "act_004",
"candidate_id": "candidate_86a9ab7071766aa8",
"claim_results": [
{
"claim_id": "claim_001",
"claim_kind": "solid_count_equals",
"deterministic": true,
"evidence": {
"solid_count": 1
},
"status": "pass"
},
{
"claim_id": "claim_002",
"claim_kind": "single_connected_body",
"deterministic": true,
"evidence": {
"solid_count": 1
},
"status": "pass"
},
{
"claim_id": "claim_003",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 80.0,
"axis": "x",
"tolerance_mm": 0.0
},
"status": "pass"
},
{
"claim_id": "claim_004",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 50.0,
"axis": "y",
"tolerance_mm": 0.0
},
"status": "pass"
},
{
"claim_id": "claim_005",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 8.0,
"axis": "z",
"tolerance_mm": 0.0
},
"status": "pass"
}
],
"event": "candidate_built"
},
{
"sequence": 7,
"at": "2026-08-31 10:20:06",
"action_id": "act_004",
"actual_atomic_id": "extrude_add_blind",
"candidate_id": "candidate_86a9ab7071766aa8",
"coverage": [
{
"claim_id": "claim_001",
"claim_kind": "solid_count_equals",
"deterministic": true,
"evidence": {
"solid_count": 1
},
"status": "pass"
},
{
"claim_id": "claim_002",
"claim_kind": "single_connected_body",
"deterministic": true,
"evidence": {
"solid_count": 1
},
"status": "pass"
},
{
"claim_id": "claim_003",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 80.0,
"axis": "x",
"tolerance_mm": 0.0
},
"status": "pass"
},
{
"claim_id": "claim_004",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 50.0,
"axis": "y",
"tolerance_mm": 0.0
},
"status": "pass"
},
{
"claim_id": "claim_005",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 8.0,
"axis": "z",
"tolerance_mm": 0.0
},
"status": "pass"
}
],
"event": "accepted",
"fragment_hash": "34b4567ab12fdc5a54b67ab39a791a6a0550daeb157fe6e40abd7f67726046bb",
"review_path": "revisions/rev_001/candidate-review.json",
"revision_id": "rev_001",
"selector_snapshot_id": "",
"working_head_before": "cad_2039eb170f23:root:v3"
},
{
"sequence": 8,
"at": "2026-08-31 10:20:09",
"claim_results": [
{
"claim_id": "claim_001",
"claim_kind": "solid_count_equals",
"deterministic": true,
"evidence": {
"solid_count": 1
},
"status": "pass"
},
{
"claim_id": "claim_002",
"claim_kind": "single_connected_body",
"deterministic": true,
"evidence": {
"solid_count": 1
},
"status": "pass"
},
{
"claim_id": "claim_003",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 80.0,
"axis": "x",
"tolerance_mm": 0.0
},
"status": "pass"
},
{
"claim_id": "claim_004",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 50.0,
"axis": "y",
"tolerance_mm": 0.0
},
"status": "pass"
},
{
"claim_id": "claim_005",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 8.0,
"axis": "z",
"tolerance_mm": 0.0
},
"status": "pass"
}
],
"event": "final_validation_requested",
"revision_id": "rev_001"
},
{
"sequence": 9,
"at": "2026-08-31 10:20:17",
"claim_results": [
{
"claim_id": "claim_001",
"claim_kind": "solid_count_equals",
"deterministic": true,
"evidence": {
"solid_count": 1
},
"status": "pass"
},
{
"claim_id": "claim_002",
"claim_kind": "single_connected_body",
"deterministic": true,
"evidence": {
"solid_count": 1
},
"status": "pass"
},
{
"claim_id": "claim_003",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 80.0,
"axis": "x",
"tolerance_mm": 0.0
},
"status": "pass"
},
{
"claim_id": "claim_004",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 50.0,
"axis": "y",
"tolerance_mm": 0.0
},
"status": "pass"
},
{
"claim_id": "claim_005",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 8.0,
"axis": "z",
"tolerance_mm": 0.0
},
"status": "pass"
}
],
"event": "completed",
"final_review_path": "revisions/rev_001/final-review.json",
"revision_id": "rev_001"
}
],
"revisions": [
{
"revision_id": "rev_001",
"status": "success",
"visibility": "final",
"cdsl_path": "revisions/rev_001/model.cdsl.json",
"step_path": "revisions/rev_001/model.step",
"glb_path": "revisions/rev_001/model.glb",
"report_path": "revisions/rev_001/rebuild-report.json",
"candidate_review_path": "revisions/rev_001/candidate-review.json"
}
]
},
"usage": {
"calls": 9,
"prompt_tokens": 72947,
"completion_tokens": 1889,
"context_chars": 51372,
"records": [
{
"completion_tokens": 334,
"context_chars": 878,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 5234,
"provider_id": "openai",
"raw_arguments_hash": "7eb460d7a0240064a62a1238b05a1654d3c3e6b0cc99c08ced5a29f3f8a6dc28",
"tool": "submit_requirements_draft_batch",
"total_tokens": 5568
},
{
"completion_tokens": 79,
"context_chars": 1684,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 5442,
"provider_id": "openai",
"raw_arguments_hash": "44136fa355b3678a1146ad16f7e8649e94fb4fc21fe77e8310c060f61caaff8a",
"tool": "finalize_requirements_draft",
"total_tokens": 5521
},
{
"completion_tokens": 71,
"context_chars": 1172,
"model_id": "gpt-5.5",
"prompt_tokens": 5005,
"provider_id": "openai",
"raw_arguments_hash": "00996cae282886750e3169d668db3145d5aad3fa73c825b1b9d5d423613bdb52",
"role": "reviewer",
"tool": "review_requirements",
"total_tokens": 5076
},
{
"completion_tokens": 402,
"context_chars": 1282,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 4961,
"provider_id": "openai",
"raw_arguments_hash": "fbc8fc3885260b3df2ea39c080a46cc7469e153ba405c80d6c5ae926eedbd4c5",
"tool": "propose_next_action",
"total_tokens": 5363
},
{
"completion_tokens": 213,
"context_chars": 2350,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 5142,
"provider_id": "openai",
"raw_arguments_hash": "267689e4fda03af2c5ca2311999445bcaae217f3ea424045ef748e2400285d14",
"tool": "get_cdsl_operation_contract",
"total_tokens": 5355
},
{
"completion_tokens": 262,
"context_chars": 7172,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6771,
"provider_id": "openai",
"raw_arguments_hash": "fdbe1ee1757bcdca738ee8d347cb75ecb0fd5efc9e9066ab47678185ab3b5ab3",
"tool": "submit_cdsl_fragment",
"total_tokens": 7033
},
{
"completion_tokens": 239,
"context_chars": 21930,
"model_id": "gpt-5.5",
"prompt_tokens": 19246,
"provider_id": "openai",
"raw_arguments_hash": "e8b0a855c46120aef598aa14e37a7d6ce99d38c654a40ac10767f2810e69556d",
"role": "reviewer",
"tool": "review_candidate",
"total_tokens": 19485
},
{
"completion_tokens": 50,
"context_chars": 4102,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6075,
"provider_id": "openai",
"raw_arguments_hash": "44136fa355b3678a1146ad16f7e8649e94fb4fc21fe77e8310c060f61caaff8a",
"tool": "complete_task",
"total_tokens": 6125
},
{
"completion_tokens": 239,
"context_chars": 10802,
"model_id": "gpt-5.5",
"prompt_tokens": 15071,
"provider_id": "openai",
"raw_arguments_hash": "ceba073ff2dae05b9c08ee83af5d12d9b538505e6f778cbec4cb681b410a6396",
"role": "reviewer",
"tool": "review_final",
"total_tokens": 15310
}
]
},
"checks": {
"expected_terminal_phase": true,
"token_budget": true,
"call_budget": true,
"author_turn_budget": true,
"reviewer_turn_budget": true,
"required_claims": true,
"required_operations": true,
"required_operation_alternative": true,
"immutable_artifacts": true,
"unique_revisions": true,
"raw_argument_audit": true,
"no_schema_or_decision_rejections": true
},
"rejection_codes": [],
"artifact_root": "/Users/lk/Projects/cdsl-cad/backend/live-evals/20260831T101800Z/artifacts/cad_2039eb170f23"
},
{
"scenario": "simple_flange",
"repetition": 1,
"success": false,
"started_at": "2026-08-31T10:20:17.378094+00:00",
"finished_at": "2026-08-31T10:22:33.804876+00:00",
"terminal": {
"taskId": "cad_ef8202e48077",
"lifecycle": "completed",
"revisionId": "rev_002"
},
"projection": {
"schema_version": "3.0",
"task_id": "cad_ef8202e48077",
"phase": "COMPLETED",
"lifecycle": "completed",
"state_version": 13,
"active_revision": "rev_002",
"current_revision": "rev_002",
"published_revision": "rev_002",
"pending_action": null,
"active_candidate_id": "",
"repair_required": false,
"last_error": "",
"action_ledger_summary": [
{
"sequence": 11,
"at": "2026-08-31 10:20:34",
"draft_revision": 1,
"event": "requirements_review_requested"
},
{
"sequence": 12,
"at": "2026-08-31 10:20:41",
"contract_hash": "5b6ba3dbd137e308cf1b0ac6a37726d3a7ea5a94a37e47e9fef0395115aa4912",
"event": "requirements_contract_frozen",
"requirement_count": 1
},
{
"sequence": 13,
"at": "2026-08-31 10:20:48",
"action_id": "act_004",
"atomic_id": "extrude_add_blind",
"contract_hash": "fa9ffc9fd2141453fa18a389813dde2e2ef3fb4c7ef5d05e61d51d614309e95f",
"event": "proposed",
"expected_change": "Create a 120 mm diameter cylindrical flange body 12 mm thick from the root sketch/profile, establishing the main connected solid in millimetres.",
"intent": "Build the base flange solid before adding the centered through bore.",
"requirement_ids": [
"req_001"
],
"working_head": "cad_ef8202e48077:root:v3"
},
{
"sequence": 14,
"at": "2026-08-31 10:21:07",
"action_id": "act_004",
"candidate_id": "candidate_644388a67554ac7d",
"event": "candidate_building",
"fragment_hash": "95bd1e739aed6d4a791a162b2ce05d84d0a52f4d18dc87c11dc7d6e8d7761078"
},
{
"sequence": 15,
"at": "2026-08-31 10:21:09",
"action_id": "act_004",
"candidate_id": "candidate_644388a67554ac7d",
"claim_results": [
{
"claim_id": "claim_001",
"claim_kind": "solid_count_equals",
"deterministic": true,
"evidence": {
"solid_count": 1
},
"status": "pass"
},
{
"claim_id": "claim_002",
"claim_kind": "single_connected_body",
"deterministic": true,
"evidence": {
"solid_count": 1
},
"status": "pass"
},
{
"claim_id": "claim_003",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 120.0,
"axis": "x",
"tolerance_mm": 0.01
},
"status": "pass"
},
{
"claim_id": "claim_004",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 120.0,
"axis": "y",
"tolerance_mm": 0.01
},
"status": "pass"
},
{
"claim_id": "claim_005",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 12.0,
"axis": "z",
"tolerance_mm": 0.01
},
"status": "pass"
},
{
"claim_id": "claim_006",
"claim_kind": "through_cylindrical_bore",
"deterministic": true,
"evidence": {
"reason": "the target bore has not been introduced at this checkpoint"
},
"status": "pending"
},
{
"claim_id": "claim_007",
"claim_kind": "visual",
"deterministic": false,
"evidence": {
"reason": "requires independent visual review"
},
"status": "pending"
}
],
"event": "candidate_built"
},
{
"sequence": 16,
"at": "2026-08-31 10:21:19",
"action_id": "act_004",
"actual_atomic_id": "extrude_add_blind",
"candidate_id": "candidate_644388a67554ac7d",
"coverage": [
{
"claim_id": "claim_001",
"claim_kind": "solid_count_equals",
"deterministic": true,
"evidence": {
"solid_count": 1
},
"status": "pass"
},
{
"claim_id": "claim_002",
"claim_kind": "single_connected_body",
"deterministic": true,
"evidence": {
"solid_count": 1
},
"status": "pass"
},
{
"claim_id": "claim_003",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 120.0,
"axis": "x",
"tolerance_mm": 0.01
},
"status": "pass"
},
{
"claim_id": "claim_004",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 120.0,
"axis": "y",
"tolerance_mm": 0.01
},
"status": "pass"
},
{
"claim_id": "claim_005",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 12.0,
"axis": "z",
"tolerance_mm": 0.01
},
"status": "pass"
},
{
"claim_id": "claim_006",
"claim_kind": "through_cylindrical_bore",
"deterministic": true,
"evidence": {
"reason": "the target bore has not been introduced at this checkpoint"
},
"status": "pending"
},
{
"claim_id": "claim_007",
"claim_kind": "visual",
"deterministic": false,
"evidence": {
"reason": "requires independent visual review"
},
"status": "pending"
}
],
"event": "accepted",
"fragment_hash": "95bd1e739aed6d4a791a162b2ce05d84d0a52f4d18dc87c11dc7d6e8d7761078",
"review_path": "revisions/rev_001/candidate-review.json",
"revision_id": "rev_001",
"selector_snapshot_id": "",
"working_head_before": "cad_ef8202e48077:root:v3"
},
{
"sequence": 17,
"at": "2026-08-31 10:21:29",
"action_id": "act_008",
"atomic_id": "hole_blind",
"contract_hash": "f62e12603feeca6d7e83acdab651fcb023e08a46c39fd1509e9dcb81d4f0c721",
"event": "proposed",
"expected_change": "Add one centered 40 mm diameter cylindrical hole through the 12 mm flange thickness, removing material from the existing solid so the bore passes fully through the part.",
"intent": "Create the required centered through bore in the finished flange body.",
"requirement_ids": [
"req_001"
],
"working_head": "cad_ef8202e48077:rev_001:v7"
},
{
"sequence": 18,
"at": "2026-08-31 10:22:00",
"action_id": "act_008",
"candidate_id": "candidate_a6ed08c266993269",
"event": "candidate_building",
"fragment_hash": "3eab7f4f6742d6ebc32b5982299b1f4d0c009e306cc43df00e69a10bfafa0998"
},
{
"sequence": 19,
"at": "2026-08-31 10:22:02",
"action_id": "act_008",
"candidate_id": "candidate_a6ed08c266993269",
"claim_results": [
{
"claim_id": "claim_001",
"claim_kind": "solid_count_equals",
"deterministic": true,
"evidence": {
"solid_count": 1
},
"status": "pass"
},
{
"claim_id": "claim_002",
"claim_kind": "single_connected_body",
"deterministic": true,
"evidence": {
"solid_count": 1
},
"status": "pass"
},
{
"claim_id": "claim_003",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 120.00000000000014,
"axis": "x",
"tolerance_mm": 0.01
},
"status": "pass"
},
{
"claim_id": "claim_004",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 120.00000000000009,
"axis": "y",
"tolerance_mm": 0.01
},
"status": "pass"
},
{
"claim_id": "claim_005",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 12.0,
"axis": "z",
"tolerance_mm": 0.01
},
"status": "pass"
},
{
"claim_id": "claim_006",
"claim_kind": "through_cylindrical_bore",
"deterministic": true,
"evidence": {
"through_bores": [
"body:feature_002:face:2"
]
},
"status": "pass"
},
{
"claim_id": "claim_007",
"claim_kind": "visual",
"deterministic": false,
"evidence": {
"reason": "requires independent visual review"
},
"status": "pending"
}
],
"event": "candidate_built"
},
{
"sequence": 20,
"at": "2026-08-31 10:22:12",
"action_id": "act_008",
"actual_atomic_id": "hole_blind",
"candidate_id": "candidate_a6ed08c266993269",
"coverage": [
{
"claim_id": "claim_001",
"claim_kind": "solid_count_equals",
"deterministic": true,
"evidence": {
"solid_count": 1
},
"status": "pass"
},
{
"claim_id": "claim_002",
"claim_kind": "single_connected_body",
"deterministic": true,
"evidence": {
"solid_count": 1
},
"status": "pass"
},
{
"claim_id": "claim_003",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 120.00000000000014,
"axis": "x",
"tolerance_mm": 0.01
},
"status": "pass"
},
{
"claim_id": "claim_004",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 120.00000000000009,
"axis": "y",
"tolerance_mm": 0.01
},
"status": "pass"
},
{
"claim_id": "claim_005",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 12.0,
"axis": "z",
"tolerance_mm": 0.01
},
"status": "pass"
},
{
"claim_id": "claim_006",
"claim_kind": "through_cylindrical_bore",
"deterministic": true,
"evidence": {
"through_bores": [
"body:feature_002:face:2"
]
},
"status": "pass"
},
{
"claim_id": "claim_007",
"claim_kind": "visual",
"deterministic": false,
"evidence": {
"reason": "requires independent visual review"
},
"status": "pending"
}
],
"event": "accepted",
"fragment_hash": "3eab7f4f6742d6ebc32b5982299b1f4d0c009e306cc43df00e69a10bfafa0998",
"review_path": "revisions/rev_002/candidate-review.json",
"revision_id": "rev_002",
"selector_snapshot_id": "cad_ef8202e48077/candidate_644388a67554ac7d",
"working_head_before": "cad_ef8202e48077:rev_001:v7"
},
{
"sequence": 21,
"at": "2026-08-31 10:22:22",
"claim_results": [
{
"claim_id": "claim_001",
"claim_kind": "solid_count_equals",
"deterministic": true,
"evidence": {
"solid_count": 1
},
"status": "pass"
},
{
"claim_id": "claim_002",
"claim_kind": "single_connected_body",
"deterministic": true,
"evidence": {
"solid_count": 1
},
"status": "pass"
},
{
"claim_id": "claim_003",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 120.00000000000014,
"axis": "x",
"tolerance_mm": 0.01
},
"status": "pass"
},
{
"claim_id": "claim_004",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 120.00000000000009,
"axis": "y",
"tolerance_mm": 0.01
},
"status": "pass"
},
{
"claim_id": "claim_005",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 12.0,
"axis": "z",
"tolerance_mm": 0.01
},
"status": "pass"
},
{
"claim_id": "claim_006",
"claim_kind": "through_cylindrical_bore",
"deterministic": true,
"evidence": {
"through_bores": [
"body:feature_002:face:2"
]
},
"status": "pass"
},
{
"claim_id": "claim_007",
"claim_kind": "visual",
"deterministic": false,
"evidence": {
"reason": "requires independent visual review"
},
"status": "pending"
}
],
"event": "final_validation_requested",
"revision_id": "rev_002"
},
{
"sequence": 22,
"at": "2026-08-31 10:22:33",
"claim_results": [
{
"claim_id": "claim_001",
"claim_kind": "solid_count_equals",
"deterministic": true,
"evidence": {
"solid_count": 1
},
"status": "pass"
},
{
"claim_id": "claim_002",
"claim_kind": "single_connected_body",
"deterministic": true,
"evidence": {
"solid_count": 1
},
"status": "pass"
},
{
"claim_id": "claim_003",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 120.00000000000014,
"axis": "x",
"tolerance_mm": 0.01
},
"status": "pass"
},
{
"claim_id": "claim_004",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 120.00000000000009,
"axis": "y",
"tolerance_mm": 0.01
},
"status": "pass"
},
{
"claim_id": "claim_005",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 12.0,
"axis": "z",
"tolerance_mm": 0.01
},
"status": "pass"
},
{
"claim_id": "claim_006",
"claim_kind": "through_cylindrical_bore",
"deterministic": true,
"evidence": {
"through_bores": [
"body:feature_002:face:2"
]
},
"status": "pass"
},
{
"claim_id": "claim_007",
"claim_kind": "visual",
"deterministic": false,
"evidence": {
"reason": "requires independent visual review"
},
"status": "pending"
}
],
"event": "completed",
"final_review_path": "revisions/rev_002/final-review.json",
"revision_id": "rev_002"
}
],
"revisions": [
{
"revision_id": "rev_001",
"status": "success",
"visibility": "checkpoint",
"cdsl_path": "revisions/rev_001/model.cdsl.json",
"step_path": "revisions/rev_001/model.step",
"glb_path": "revisions/rev_001/model.glb",
"report_path": "revisions/rev_001/rebuild-report.json",
"candidate_review_path": "revisions/rev_001/candidate-review.json"
},
{
"revision_id": "rev_002",
"status": "success",
"visibility": "final",
"cdsl_path": "revisions/rev_002/model.cdsl.json",
"step_path": "revisions/rev_002/model.step",
"glb_path": "revisions/rev_002/model.glb",
"report_path": "revisions/rev_002/rebuild-report.json",
"candidate_review_path": "revisions/rev_002/candidate-review.json"
}
]
},
"usage": {
"calls": 16,
"prompt_tokens": 133213,
"completion_tokens": 4905,
"context_chars": 118349,
"records": [
{
"completion_tokens": 736,
"context_chars": 934,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 5242,
"provider_id": "openai",
"raw_arguments_hash": "ddbd5623b25d8561fa8c0804534deb5f39b1d5abe0290f8478357398e88e2903",
"tool": "submit_requirements_draft_batch",
"total_tokens": 5978
},
{
"completion_tokens": 68,
"context_chars": 2073,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 5541,
"provider_id": "openai",
"raw_arguments_hash": "44136fa355b3678a1146ad16f7e8649e94fb4fc21fe77e8310c060f61caaff8a",
"tool": "finalize_requirements_draft",
"total_tokens": 5609
},
{
"completion_tokens": 71,
"context_chars": 1509,
"model_id": "gpt-5.5",
"prompt_tokens": 5100,
"provider_id": "openai",
"raw_arguments_hash": "00996cae282886750e3169d668db3145d5aad3fa73c825b1b9d5d423613bdb52",
"role": "reviewer",
"tool": "review_requirements",
"total_tokens": 5171
},
{
"completion_tokens": 288,
"context_chars": 1355,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 4985,
"provider_id": "openai",
"raw_arguments_hash": "ae240facd59e8e4741fb47fb699b532def00adcc8b14f5a432721a44e801cc99",
"tool": "propose_next_action",
"total_tokens": 5273
},
{
"completion_tokens": 206,
"context_chars": 2420,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 5162,
"provider_id": "openai",
"raw_arguments_hash": "0104a85097dfd4dee5706a118dcf463106e3d43d89481d0861c4e5b9a2f6c281",
"tool": "get_cdsl_operation_contract",
"total_tokens": 5368
},
{
"completion_tokens": 144,
"context_chars": 2786,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 5259,
"provider_id": "openai",
"raw_arguments_hash": "0e4c1562f9031cfb60f2c36487c44eee29c82c778133d4e138041a0d4d8cf6de",
"tool": "get_cdsl_operation_contract",
"total_tokens": 5403
},
{
"completion_tokens": 283,
"context_chars": 7608,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6888,
"provider_id": "openai",
"raw_arguments_hash": "31c65cdd0f45421d4fc38984c788e521d6ed99c8fb8fa58ab2ba62bc87526b1e",
"tool": "submit_cdsl_fragment",
"total_tokens": 7171
},
{
"completion_tokens": 305,
"context_chars": 22656,
"model_id": "gpt-5.5",
"prompt_tokens": 19473,
"provider_id": "openai",
"raw_arguments_hash": "8c74bce21c96269b82130cb7cb35f2d318e9714599599ba9e583ed3cf306aa08",
"role": "reviewer",
"tool": "review_candidate",
"total_tokens": 19778
},
{
"completion_tokens": 456,
"context_chars": 4980,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6281,
"provider_id": "openai",
"raw_arguments_hash": "cd87015d21863d5a301c5ed21a12322360c851cad3204d7fa6047627e6327624",
"tool": "propose_next_action",
"total_tokens": 6737
},
{
"completion_tokens": 376,
"context_chars": 5578,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6066,
"provider_id": "openai",
"raw_arguments_hash": "89abd3374b1f6eb20bc93717469a243f0596f4619f3540fe65d20025f657f1b6",
"tool": "get_cdsl_operation_contract",
"total_tokens": 6442
},
{
"completion_tokens": 283,
"context_chars": 5950,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6167,
"provider_id": "openai",
"raw_arguments_hash": "f63a33f3d06df5403ee42eae6e3d3be1797c73ade7175743eac89dc1253982f5",
"tool": "get_cdsl_operation_contract",
"total_tokens": 6450
},
{
"completion_tokens": 194,
"context_chars": 9347,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 7246,
"provider_id": "openai",
"raw_arguments_hash": "4c513ec5bbdbc82a26d93652e12cbb9f7e5b2fa11c90f3963e5a8899d28cf3e7",
"tool": "inspect_topology",
"total_tokens": 7440
},
{
"completion_tokens": 321,
"context_chars": 11038,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 8406,
"provider_id": "openai",
"raw_arguments_hash": "6d796108b832bc3d33155fc09a27913120112d7d863343548f45231e2ee31ce8",
"tool": "submit_cdsl_fragment",
"total_tokens": 8727
},
{
"completion_tokens": 336,
"context_chars": 23291,
"model_id": "gpt-5.5",
"prompt_tokens": 19661,
"provider_id": "openai",
"raw_arguments_hash": "7d94d745a72edef2f342e0470e32e25283b54939634b6a7bdc779186a8955bf3",
"role": "reviewer",
"tool": "review_candidate",
"total_tokens": 19997
},
{
"completion_tokens": 515,
"context_chars": 5037,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6331,
"provider_id": "openai",
"raw_arguments_hash": "44136fa355b3678a1146ad16f7e8649e94fb4fc21fe77e8310c060f61caaff8a",
"tool": "complete_task",
"total_tokens": 6846
},
{
"completion_tokens": 323,
"context_chars": 11787,
"model_id": "gpt-5.5",
"prompt_tokens": 15405,
"provider_id": "openai",
"raw_arguments_hash": "f7dbfe4000ef981bc0266908c05d1b3c3e9a7785d9e35fd6376ac90b28b092b6",
"role": "reviewer",
"tool": "review_final",
"total_tokens": 15728
}
]
},
"checks": {
"expected_terminal_phase": true,
"token_budget": true,
"call_budget": true,
"author_turn_budget": true,
"reviewer_turn_budget": true,
"required_claims": true,
"required_operations": true,
"required_operation_alternative": true,
"immutable_artifacts": true,
"unique_revisions": true,
"raw_argument_audit": true,
"no_schema_or_decision_rejections": false
},
"rejection_codes": [
"STALE_WORKING_HEAD",
"STALE_WORKING_HEAD"
],
"artifact_root": "/Users/lk/Projects/cdsl-cad/backend/live-evals/20260831T101800Z/artifacts/cad_ef8202e48077"
}
],
"suite": "smoke",
"require_live": true,
"report_root": "/Users/lk/Projects/cdsl-cad/backend/live-evals/20260831T101800Z"
}