Files
cdsl-cad/backend/live-evals/20260831T102709Z/report.json
T
2026-09-01 14:10:23 +08:00

1190 lines
40 KiB
JSON

{
"status": "passed",
"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": 1372
},
"mode": "openai_compatible_tool_calling"
},
"reviewer_capability": {
"schema_hash": "fb899a5de2f869b648df88a32d04a5eee92ae3d193ab23ee641892d6b59cd8e7",
"tool_count": 9,
"supported": true,
"failures": [],
"usage": {
"prompt_tokens": 41789,
"completion_tokens": 543
},
"mode": "openai_compatible_tool_calling"
},
"results": [
{
"scenario": "rectangular_plate",
"repetition": 1,
"success": true,
"started_at": "2026-08-31T10:28:37.832406+00:00",
"finished_at": "2026-08-31T10:29:32.753023+00:00",
"terminal": {
"taskId": "cad_212841a7e248",
"lifecycle": "completed",
"revisionId": "rev_001"
},
"projection": {
"schema_version": "3.0",
"task_id": "cad_212841a7e248",
"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 10:28:45",
"draft_revision": 1,
"event": "requirements_draft_updated",
"item_count": 1
},
{
"sequence": 2,
"at": "2026-08-31 10:28:48",
"draft_revision": 1,
"event": "requirements_review_requested"
},
{
"sequence": 3,
"at": "2026-08-31 10:28:52",
"contract_hash": "891abb525827ad48707b2e7aced262443318e11d71106959adca8e302827910d",
"event": "requirements_contract_frozen",
"requirement_count": 1
},
{
"sequence": 4,
"at": "2026-08-31 10:28:59",
"action_id": "act_004",
"atomic_id": "extrude_add_blind",
"contract_hash": "fa9ffc9fd2141453fa18a389813dde2e2ef3fb4c7ef5d05e61d51d614309e95f",
"event": "proposed",
"expected_change": "Create a single connected solid rectangular plate measuring 80 mm by 50 mm by 8 mm.",
"intent": "Satisfy req_001 by adding the plate as one connected prismatic body in millimetres.",
"requirement_ids": [
"req_001"
],
"working_head": "cad_212841a7e248:root:v3"
},
{
"sequence": 5,
"at": "2026-08-31 10:29:10",
"action_id": "act_004",
"candidate_id": "candidate_c364b4c5678722fb",
"event": "candidate_building",
"fragment_hash": "34b4567ab12fdc5a54b67ab39a791a6a0550daeb157fe6e40abd7f67726046bb"
},
{
"sequence": 6,
"at": "2026-08-31 10:29:11",
"action_id": "act_004",
"candidate_id": "candidate_c364b4c5678722fb",
"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 10:29:20",
"action_id": "act_004",
"actual_atomic_id": "extrude_add_blind",
"candidate_id": "candidate_c364b4c5678722fb",
"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_212841a7e248:root:v3"
},
{
"sequence": 8,
"at": "2026-08-31 10:29:24",
"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 10:29:32",
"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": 72883,
"completion_tokens": 1811,
"context_chars": 51142,
"records": [
{
"completion_tokens": 334,
"context_chars": 878,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 5234,
"provider_id": "openai",
"raw_arguments_hash": "f0d6d9b69102c3fc061fb4cbbaf5b5cfe67da64ee41a8fd30eb6e28baa2b4a23",
"tool": "submit_requirements_draft_batch",
"total_tokens": 5568
},
{
"completion_tokens": 82,
"context_chars": 1700,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 5447,
"provider_id": "openai",
"raw_arguments_hash": "44136fa355b3678a1146ad16f7e8649e94fb4fc21fe77e8310c060f61caaff8a",
"tool": "finalize_requirements_draft",
"total_tokens": 5529
},
{
"completion_tokens": 71,
"context_chars": 1186,
"model_id": "gpt-5.5",
"prompt_tokens": 5010,
"provider_id": "openai",
"raw_arguments_hash": "00996cae282886750e3169d668db3145d5aad3fa73c825b1b9d5d423613bdb52",
"role": "reviewer",
"tool": "review_requirements",
"total_tokens": 5081
},
{
"completion_tokens": 302,
"context_chars": 1282,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 4953,
"provider_id": "openai",
"raw_arguments_hash": "380ec9dea594d2498f7028273894bfdf09c4f6c30024c87aa1b3da4bd5012d33",
"tool": "propose_next_action",
"total_tokens": 5255
},
{
"completion_tokens": 135,
"context_chars": 2254,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 5104,
"provider_id": "openai",
"raw_arguments_hash": "f4def2da351315b270f551588444f6f615a5d4b182249e4dbad7e8ee3e6f2f34",
"tool": "get_cdsl_operation_contract",
"total_tokens": 5239
},
{
"completion_tokens": 248,
"context_chars": 7076,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6759,
"provider_id": "openai",
"raw_arguments_hash": "fdbe1ee1757bcdca738ee8d347cb75ecb0fd5efc9e9066ab47678185ab3b5ab3",
"tool": "submit_cdsl_fragment",
"total_tokens": 7007
},
{
"completion_tokens": 272,
"context_chars": 21896,
"model_id": "gpt-5.5",
"prompt_tokens": 19238,
"provider_id": "openai",
"raw_arguments_hash": "fbd289b4eb30ffa7e584f08d8c51d013513cae871cb041a0141b2be0597e42c6",
"role": "reviewer",
"tool": "review_candidate",
"total_tokens": 19510
},
{
"completion_tokens": 122,
"context_chars": 4054,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6070,
"provider_id": "openai",
"raw_arguments_hash": "44136fa355b3678a1146ad16f7e8649e94fb4fc21fe77e8310c060f61caaff8a",
"tool": "complete_task",
"total_tokens": 6192
},
{
"completion_tokens": 245,
"context_chars": 10816,
"model_id": "gpt-5.5",
"prompt_tokens": 15068,
"provider_id": "openai",
"raw_arguments_hash": "e10e72c02bb0495ebd65b1b2557730b3326c5c30ad1a73b6ed631cf476d8f190",
"role": "reviewer",
"tool": "review_final",
"total_tokens": 15313
}
]
},
"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/20260831T102709Z/artifacts/cad_212841a7e248"
},
{
"scenario": "simple_flange",
"repetition": 1,
"success": true,
"started_at": "2026-08-31T10:29:32.755510+00:00",
"finished_at": "2026-08-31T10:31:20.753138+00:00",
"terminal": {
"taskId": "cad_71d72d14b433",
"lifecycle": "completed",
"revisionId": "rev_002"
},
"projection": {
"schema_version": "3.0",
"task_id": "cad_71d72d14b433",
"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 10:29:46",
"draft_revision": 1,
"event": "requirements_review_requested"
},
{
"sequence": 12,
"at": "2026-08-31 10:29:53",
"contract_hash": "af85f72e5d5efe317ca1f398b9676e21a4be90a83af01f5a6bad893c16a339b6",
"event": "requirements_contract_frozen",
"requirement_count": 1
},
{
"sequence": 13,
"at": "2026-08-31 10:30:02",
"action_id": "act_004",
"atomic_id": "extrude_add_blind",
"contract_hash": "fa9ffc9fd2141453fa18a389813dde2e2ef3fb4c7ef5d05e61d51d614309e95f",
"event": "proposed",
"expected_change": "Create a solid round flange body in millimetres with outer diameter 120 mm and thickness 12 mm.",
"intent": "Establish the main flange geometry as a single connected base solid before adding the centered bore.",
"requirement_ids": [
"req_001"
],
"working_head": "cad_71d72d14b433:root:v3"
},
{
"sequence": 14,
"at": "2026-08-31 10:30:12",
"action_id": "act_004",
"candidate_id": "candidate_241334e701818a16",
"event": "candidate_building",
"fragment_hash": "95bd1e739aed6d4a791a162b2ce05d84d0a52f4d18dc87c11dc7d6e8d7761078"
},
{
"sequence": 15,
"at": "2026-08-31 10:30:14",
"action_id": "act_004",
"candidate_id": "candidate_241334e701818a16",
"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": 0.1
},
"status": "pass"
},
{
"claim_id": "claim_004",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 120.0,
"axis": "y",
"tolerance_mm": 0.1
},
"status": "pass"
},
{
"claim_id": "claim_005",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 12.0,
"axis": "z",
"tolerance_mm": 0.1
},
"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"
}
],
"event": "candidate_built"
},
{
"sequence": 16,
"at": "2026-08-31 10:30:25",
"action_id": "act_004",
"actual_atomic_id": "extrude_add_blind",
"candidate_id": "candidate_241334e701818a16",
"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": 0.1
},
"status": "pass"
},
{
"claim_id": "claim_004",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 120.0,
"axis": "y",
"tolerance_mm": 0.1
},
"status": "pass"
},
{
"claim_id": "claim_005",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 12.0,
"axis": "z",
"tolerance_mm": 0.1
},
"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"
}
],
"event": "accepted",
"fragment_hash": "95bd1e739aed6d4a791a162b2ce05d84d0a52f4d18dc87c11dc7d6e8d7761078",
"review_path": "revisions/rev_001/candidate-review.json",
"revision_id": "rev_001",
"selector_snapshot_id": "",
"working_head_before": "cad_71d72d14b433:root:v3"
},
{
"sequence": 17,
"at": "2026-08-31 10:30:32",
"action_id": "act_008",
"atomic_id": "hole_blind",
"contract_hash": "f62e12603feeca6d7e83acdab651fcb023e08a46c39fd1509e9dcb81d4f0c721",
"event": "proposed",
"expected_change": "Cut a centered cylindrical hole through the 12 mm flange with 40 mm diameter by extending the bore to the full material thickness.",
"intent": "Introduce the required centered through bore while preserving the single connected flange body.",
"requirement_ids": [
"req_001"
],
"working_head": "cad_71d72d14b433:rev_001:v7"
},
{
"sequence": 18,
"at": "2026-08-31 10:30:53",
"action_id": "act_008",
"candidate_id": "candidate_a9836999e50ab5f3",
"event": "candidate_building",
"fragment_hash": "1e77ee8e187829845c362464afb50cf9a5cb2689ec1bcea9cebc073f8fe5a2a1"
},
{
"sequence": 19,
"at": "2026-08-31 10:30:55",
"action_id": "act_008",
"candidate_id": "candidate_a9836999e50ab5f3",
"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": 0.1
},
"status": "pass"
},
{
"claim_id": "claim_004",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 120.00000000000009,
"axis": "y",
"tolerance_mm": 0.1
},
"status": "pass"
},
{
"claim_id": "claim_005",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 12.0,
"axis": "z",
"tolerance_mm": 0.1
},
"status": "pass"
},
{
"claim_id": "claim_006",
"claim_kind": "through_cylindrical_bore",
"deterministic": true,
"evidence": {
"through_bores": [
"body:feature_002:face:2"
]
},
"status": "pass"
}
],
"event": "candidate_built"
},
{
"sequence": 20,
"at": "2026-08-31 10:31:07",
"action_id": "act_008",
"actual_atomic_id": "hole_blind",
"candidate_id": "candidate_a9836999e50ab5f3",
"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": 0.1
},
"status": "pass"
},
{
"claim_id": "claim_004",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 120.00000000000009,
"axis": "y",
"tolerance_mm": 0.1
},
"status": "pass"
},
{
"claim_id": "claim_005",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 12.0,
"axis": "z",
"tolerance_mm": 0.1
},
"status": "pass"
},
{
"claim_id": "claim_006",
"claim_kind": "through_cylindrical_bore",
"deterministic": true,
"evidence": {
"through_bores": [
"body:feature_002:face:2"
]
},
"status": "pass"
}
],
"event": "accepted",
"fragment_hash": "1e77ee8e187829845c362464afb50cf9a5cb2689ec1bcea9cebc073f8fe5a2a1",
"review_path": "revisions/rev_002/candidate-review.json",
"revision_id": "rev_002",
"selector_snapshot_id": "cad_71d72d14b433/candidate_241334e701818a16",
"working_head_before": "cad_71d72d14b433:rev_001:v7"
},
{
"sequence": 21,
"at": "2026-08-31 10:31:11",
"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": 0.1
},
"status": "pass"
},
{
"claim_id": "claim_004",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 120.00000000000009,
"axis": "y",
"tolerance_mm": 0.1
},
"status": "pass"
},
{
"claim_id": "claim_005",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 12.0,
"axis": "z",
"tolerance_mm": 0.1
},
"status": "pass"
},
{
"claim_id": "claim_006",
"claim_kind": "through_cylindrical_bore",
"deterministic": true,
"evidence": {
"through_bores": [
"body:feature_002:face:2"
]
},
"status": "pass"
}
],
"event": "final_validation_requested",
"revision_id": "rev_002"
},
{
"sequence": 22,
"at": "2026-08-31 10:31:20",
"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": 0.1
},
"status": "pass"
},
{
"claim_id": "claim_004",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 120.00000000000009,
"axis": "y",
"tolerance_mm": 0.1
},
"status": "pass"
},
{
"claim_id": "claim_005",
"claim_kind": "bbox_dimension_mm",
"deterministic": true,
"evidence": {
"actual_mm": 12.0,
"axis": "z",
"tolerance_mm": 0.1
},
"status": "pass"
},
{
"claim_id": "claim_006",
"claim_kind": "through_cylindrical_bore",
"deterministic": true,
"evidence": {
"through_bores": [
"body:feature_002:face:2"
]
},
"status": "pass"
}
],
"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": 14,
"prompt_tokens": 120729,
"completion_tokens": 3514,
"context_chars": 105477,
"records": [
{
"completion_tokens": 516,
"context_chars": 934,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 5242,
"provider_id": "openai",
"raw_arguments_hash": "f106c56ab0ea27466f7e159e2f0f8941b1ef392709960088d9349e5c818cfa11",
"tool": "submit_requirements_draft_batch",
"total_tokens": 5758
},
{
"completion_tokens": 117,
"context_chars": 1905,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 5507,
"provider_id": "openai",
"raw_arguments_hash": "44136fa355b3678a1146ad16f7e8649e94fb4fc21fe77e8310c060f61caaff8a",
"tool": "finalize_requirements_draft",
"total_tokens": 5624
},
{
"completion_tokens": 71,
"context_chars": 1353,
"model_id": "gpt-5.5",
"prompt_tokens": 5066,
"provider_id": "openai",
"raw_arguments_hash": "00996cae282886750e3169d668db3145d5aad3fa73c825b1b9d5d423613bdb52",
"role": "reviewer",
"tool": "review_requirements",
"total_tokens": 5137
},
{
"completion_tokens": 434,
"context_chars": 1352,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 4982,
"provider_id": "openai",
"raw_arguments_hash": "32c39ab63762970c2aa6be8196f469052916f43b2793550b9033a97fcd959297",
"tool": "propose_next_action",
"total_tokens": 5416
},
{
"completion_tokens": 145,
"context_chars": 2383,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 5133,
"provider_id": "openai",
"raw_arguments_hash": "7fb08e885b4c06efee8b4c34fdfd5b8cf4b40f5bb5fad90974318c47d08c74e1",
"tool": "get_cdsl_operation_contract",
"total_tokens": 5278
},
{
"completion_tokens": 204,
"context_chars": 7205,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6788,
"provider_id": "openai",
"raw_arguments_hash": "af5dd7a546cf394eab4affaf0e3486f75fedc950354cbac3345229497aaa0a90",
"tool": "submit_cdsl_fragment",
"total_tokens": 6992
},
{
"completion_tokens": 319,
"context_chars": 22297,
"model_id": "gpt-5.5",
"prompt_tokens": 19257,
"provider_id": "openai",
"raw_arguments_hash": "748f9f858ad36266bab0b65a10b649758be386669461c6b8022a20756c021c0a",
"role": "reviewer",
"tool": "review_candidate",
"total_tokens": 19576
},
{
"completion_tokens": 311,
"context_chars": 4598,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6194,
"provider_id": "openai",
"raw_arguments_hash": "ea671a4ef923dd91d9a66b03e9ce6e6c8ac6fbc78c5f9e643397f31dd6fb9cf7",
"tool": "propose_next_action",
"total_tokens": 6505
},
{
"completion_tokens": 203,
"context_chars": 5185,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 5947,
"provider_id": "openai",
"raw_arguments_hash": "1375c42503256d751af6b845ffc59d8cdf37fe498c467e8aa75754de5c06df14",
"tool": "get_cdsl_operation_contract",
"total_tokens": 6150
},
{
"completion_tokens": 243,
"context_chars": 8582,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 7041,
"provider_id": "openai",
"raw_arguments_hash": "9e3c866c98c64757af25324c376141c17bffd3fdc676360e83f429c5ed61fde4",
"tool": "inspect_topology",
"total_tokens": 7284
},
{
"completion_tokens": 260,
"context_chars": 10645,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 8291,
"provider_id": "openai",
"raw_arguments_hash": "6ed475989c3ce0338e53ec1302d94dc0f0473493ea3e3d9fdabe5886b14408ff",
"tool": "submit_cdsl_fragment",
"total_tokens": 8551
},
{
"completion_tokens": 317,
"context_chars": 22935,
"model_id": "gpt-5.5",
"prompt_tokens": 19692,
"provider_id": "openai",
"raw_arguments_hash": "d46b81f9b006e8049f65319a80368020a4de5cba0664896341afd30de5722d6f",
"role": "reviewer",
"tool": "review_candidate",
"total_tokens": 20009
},
{
"completion_tokens": 70,
"context_chars": 4658,
"model_id": "gpt-5.4-mini",
"prompt_tokens": 6244,
"provider_id": "openai",
"raw_arguments_hash": "44136fa355b3678a1146ad16f7e8649e94fb4fc21fe77e8310c060f61caaff8a",
"tool": "complete_task",
"total_tokens": 6314
},
{
"completion_tokens": 304,
"context_chars": 11445,
"model_id": "gpt-5.5",
"prompt_tokens": 15345,
"provider_id": "openai",
"raw_arguments_hash": "5c317600b3e1d40d78df0e3362d74e8d73fad3d281e4e96fbc333d6bb9bf9255",
"role": "reviewer",
"tool": "review_final",
"total_tokens": 15649
}
]
},
"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/20260831T102709Z/artifacts/cad_71d72d14b433"
}
],
"suite": "smoke",
"require_live": true,
"report_root": "/Users/lk/Projects/cdsl-cad/backend/live-evals/20260831T102709Z"
}