530 lines
25 KiB
JSON
530 lines
25 KiB
JSON
{
|
||
"trace_id": "9zgbcBw6rEvf_w",
|
||
"timestamp": "2026-03-24T08:40:11.349Z",
|
||
"model": "gpt-4o-mini",
|
||
"prompt_version": "normalizer_v2_0_2",
|
||
"schema_version": "v2_0_2",
|
||
"case_id": "BQ-025",
|
||
"user_question_raw": "Есть ли в нашей базе куски, которые выглядят так, будто их кто-то руками “дособирал”, и теперь формально всё стоит, но смысловая логика учёта у этого шатается.",
|
||
"context": {
|
||
"period_hint": null,
|
||
"business_context": null,
|
||
"expected_route": null,
|
||
"case_id": "BQ-025",
|
||
"eval_mode": "single-pass-strict",
|
||
"trace_completeness_issues": []
|
||
},
|
||
"request_payload_redacted": {
|
||
"model": "gpt-4o-mini",
|
||
"baseUrl": "https://api.openai.com/v1",
|
||
"temperature": 0,
|
||
"maxOutputTokens": 900,
|
||
"promptVersion": "normalizer_v2_0_2",
|
||
"systemPrompt": "Ты semantic-normalizer для бухгалтерского ассистента NDC.\nТвоя роль: только нормализация запроса пользователя в строгий JSON-контракт.\n\nЖесткие правила:\n1) Не давай бухгалтерский ответ по сути вопроса.\n2) Возвращай только JSON без markdown и пояснений.\n3) JSON обязан соответствовать переданной schema normalized_query_v1.\n4) Если период не указан, не выдумывай его; отмечай ambiguity.\n5) Для цепочек документов/проводок/оплат поднимай causal и cross-entity признаки.\n6) Для точечного object trace (номер/строка/ref) поднимай needs_exact_object_trace=true.\n7) Используй терминологию NDC.",
|
||
"developerPrompt": "You are semantic-normalizer for accounting assistant NDC.\nReturn strict JSON only, no markdown, no comments.\n\nTarget schema: normalized_query_v2_0_2.\n\nCore behavior (v2.0.2):\n1. Decompose message into semantic fragments.\n2. Classify fragment domain relevance and business scope.\n3. Fill route-critical flags and candidate labels.\n4. For each fragment set execution state fields:\n - execution_readiness\n - clarification_reason\n - soft_assumption_used\n - route_status\n - no_route_reason\n5. Clarification must be rare and justified.\n\nExecution-state policy:\n- Every in-scope fragment must produce a consistent execution state.\n- If a fragment is routable, mark it as executable or executable_with_soft_assumptions.\n- Do not leave routable fragments in unresolved state.\n- If a fragment cannot be routed, set route_status=no_route and provide explicit no_route_reason.\n\nReadiness values:\n- executable\n- executable_with_soft_assumptions\n- needs_clarification\n- no_route\n\nRoute status values:\n- routed\n- no_route\n\nNo-route reason values:\n- out_of_scope\n- insufficient_specificity\n- missing_mapping\n- unsupported_fragment_type\n\nDo not over-require formality:\n- Do not require document IDs, exact periods, or exact object references for scan/review/anomaly/rule-check requests.\n- Colloquial accounting phrases like \"что висит\", \"что подозрительно\", \"что не сходится\", \"что криво\", \"что аукнется\" are executable if accounting area is understandable.\n\nFragment required fields:\n- fragment_id\n- raw_fragment_text\n- normalized_fragment_text\n- domain_relevance\n- business_scope\n- entity_hints\n- account_hints\n- document_hints\n- register_hints\n- time_scope\n- flags\n- candidate_labels\n- confidence\n- execution_readiness\n- clarification_reason\n- soft_assumption_used\n- route_status\n- no_route_reason\n\nSoft assumptions (`soft_assumption_used`) allowed values:\n- period_from_session_context\n- company_scope_defaulted\n- problem_scan_mode_enabled\n\nGlobal notes:\n- global_notes.needs_clarification should be true only when execution is truly blocked for all in-scope fragments.\n- global_notes.clarification_reason must explain the blocker.\n\nSchema version must be:\n- \"schema_version\": \"normalized_query_v2_0_2\"",
|
||
"domainPrompt": "Контекст домена: бухгалтерия 1С/NDC.\n\nКлючевые счета:\n- 01, 02, 10, 41, 51, 60, 62, 68.02, 90, 97.\n\nТиповые сущности:\n- контрагент, договор, документ реализации, документ поступления, оплата, проводка, регистр, закрывающий документ.\n\nЛексика causal и сверки (сильные сигналы для cross_entity):\n- \"не бьется\", \"не сходится\", \"не видно\", \"не собралось\", \"повисло\", \"хвост\";\n- \"разложи по документам/оплатам/закрывающим\";\n- \"чем подтверждается\", \"где ошибка в цепочке\", \"что пошло криво\".\n\nЛексика точечного drilldown:\n- \"документ №...\", \"ref\", \"строка проводки\", \"покажи конкретную операцию\", \"точный source-of-record\".\n\nЛексика rule-based контроля:\n- \"проверь настройки\", \"ошибка срока/даты\", \"контроль 97/10/ОС\", \"нарушение правила учета\".\n\nЛексика обзорной аналитики:\n- \"рейтинг\", \"топ рисков\", \"в целом по компании\", \"перед закрытием периода\", \"приоритизация проверок\".\n\nВажное правило:\nЕсли в одном вопросе есть и риск-лексика, и цепочка document/payment/posting, не понижать задачу до чистого `store_feature_risk`.\nПриоритет у causal cross-entity семантики.",
|
||
"fewShotExamples": "Q: По каким поставщикам висят хвосты по 60, что подозрительно по цепочке оплат?\nA: in_scope fragment, execution_readiness=executable_with_soft_assumptions, route_status=routed, no_route_reason=null.\n\nQ: Покажи записи по 97, которые повисли и могут аукнуться.\nA: in_scope fragment, execution_readiness=executable_with_soft_assumptions, route_status=routed, no_route_reason=null.\n\nQ: Чекни, что у нас не так.\nA: unclear/in_scope fragment, execution_readiness=needs_clarification, route_status=no_route, no_route_reason=insufficient_specificity.\n\nQ: Как вообще по ФСБУ правильно?\nA: out_of_scope fragment, execution_readiness=no_route, route_status=no_route, no_route_reason=out_of_scope.",
|
||
"userQuestion": "Есть ли в нашей базе куски, которые выглядят так, будто их кто-то руками “дособирал”, и теперь формально всё стоит, но смысловая логика учёта у этого шатается.",
|
||
"context": {
|
||
"eval_label": "eval-uQiZ1L_ga6",
|
||
"case_id": "BQ-025",
|
||
"eval_mode": "single-pass-strict"
|
||
},
|
||
"retryPolicy": "single-pass-strict",
|
||
"useMock": false
|
||
},
|
||
"raw_model_response": {
|
||
"id": "resp_0b56b5b2450f04760069c24de0f7c881a19e42f69bbe3432ec",
|
||
"object": "response",
|
||
"created_at": 1774341600,
|
||
"status": "completed",
|
||
"background": false,
|
||
"billing": {
|
||
"payer": "developer"
|
||
},
|
||
"completed_at": 1774341610,
|
||
"error": null,
|
||
"frequency_penalty": 0,
|
||
"incomplete_details": null,
|
||
"instructions": null,
|
||
"max_output_tokens": 900,
|
||
"max_tool_calls": null,
|
||
"model": "gpt-4o-mini-2024-07-18",
|
||
"output": [
|
||
{
|
||
"id": "msg_0b56b5b2450f04760069c24de1661481a1b7688aa3a7627af2",
|
||
"type": "message",
|
||
"status": "completed",
|
||
"content": [
|
||
{
|
||
"type": "output_text",
|
||
"annotations": [],
|
||
"logprobs": [],
|
||
"text": "{\n \"schema_version\": \"normalized_query_v2_0_2\",\n \"user_message_raw\": \"Есть ли в нашей базе куски, которые выглядят так, будто их кто-то руками “дособирал”, и теперь формально всё стоит, но смысловая логика учёта у этого шатается.\",\n \"message_in_scope\": true,\n \"scope_confidence\": \"high\",\n \"contains_multiple_tasks\": false,\n \"fragments\": [\n {\n \"fragment_id\": \"1\",\n \"raw_fragment_text\": \"куски, которые выглядят так, будто их кто-то руками “дособирал”\",\n \"normalized_fragment_text\": \"не бьется, куски, которые выглядят так, будто их кто-то руками дособирал\",\n \"domain_relevance\": \"in_scope\",\n \"business_scope\": \"generic_accounting\",\n \"entity_hints\": [],\n \"account_hints\": [],\n \"document_hints\": [],\n \"register_hints\": [],\n \"time_scope\": {\n \"type\": \"missing\",\n \"value\": null,\n \"confidence\": \"low\"\n },\n \"flags\": {\n \"has_multi_entity_scope\": true,\n \"asks_for_chain_explanation\": true\n ,\n \"asks_for_ranking_or_top\": false,\n \"asks_for_period_summary\": false,\n \"asks_for_rule_check\": false,\n \"asks_for_anomaly_scan\": false,\n \"asks_for_exact_object_trace\": false,\n \"asks_for_evidence\": false,\n \"mentions_period_close_context\": false\n },\n \"candidate_labels\": [\n \"cross_entity\",\n \"anomaly_probe\"\n ],\n \"confidence\": \"high\",\n \"execution_readiness\": \"executable\",\n \"clarification_reason\": null,\n \"soft_assumption_used\": [],\n \"route_status\": \"routed\",\n \"no_route_reason\": null\n }\n ],\n \"discarded_fragments\": [],\n \"global_notes\": {\n \"needs_clarification\": false,\n \"clarification_reason\": null\n }\n}"
|
||
}
|
||
],
|
||
"role": "assistant"
|
||
}
|
||
],
|
||
"parallel_tool_calls": true,
|
||
"presence_penalty": 0,
|
||
"previous_response_id": null,
|
||
"prompt_cache_key": null,
|
||
"prompt_cache_retention": null,
|
||
"reasoning": {
|
||
"effort": null,
|
||
"summary": null
|
||
},
|
||
"safety_identifier": null,
|
||
"service_tier": "default",
|
||
"store": true,
|
||
"temperature": 0,
|
||
"text": {
|
||
"format": {
|
||
"type": "json_schema",
|
||
"description": null,
|
||
"name": "normalized_query_v2_0_2",
|
||
"schema": {
|
||
"$id": "normalized_query_v2_0_2",
|
||
"title": "Normalized Query V2.0.2",
|
||
"type": "object",
|
||
"additionalProperties": false,
|
||
"required": [
|
||
"schema_version",
|
||
"user_message_raw",
|
||
"message_in_scope",
|
||
"scope_confidence",
|
||
"contains_multiple_tasks",
|
||
"fragments",
|
||
"discarded_fragments",
|
||
"global_notes"
|
||
],
|
||
"properties": {
|
||
"schema_version": {
|
||
"type": "string",
|
||
"const": "normalized_query_v2_0_2"
|
||
},
|
||
"user_message_raw": {
|
||
"type": "string",
|
||
"minLength": 1
|
||
},
|
||
"message_in_scope": {
|
||
"type": "boolean"
|
||
},
|
||
"scope_confidence": {
|
||
"type": "string",
|
||
"enum": [
|
||
"high",
|
||
"medium",
|
||
"low"
|
||
]
|
||
},
|
||
"contains_multiple_tasks": {
|
||
"type": "boolean"
|
||
},
|
||
"fragments": {
|
||
"type": "array",
|
||
"items": {
|
||
"type": "object",
|
||
"additionalProperties": false,
|
||
"required": [
|
||
"fragment_id",
|
||
"raw_fragment_text",
|
||
"normalized_fragment_text",
|
||
"domain_relevance",
|
||
"business_scope",
|
||
"entity_hints",
|
||
"account_hints",
|
||
"document_hints",
|
||
"register_hints",
|
||
"time_scope",
|
||
"flags",
|
||
"candidate_labels",
|
||
"confidence",
|
||
"execution_readiness",
|
||
"clarification_reason",
|
||
"soft_assumption_used",
|
||
"route_status",
|
||
"no_route_reason"
|
||
],
|
||
"properties": {
|
||
"fragment_id": {
|
||
"type": "string",
|
||
"minLength": 1
|
||
},
|
||
"raw_fragment_text": {
|
||
"type": "string",
|
||
"minLength": 1
|
||
},
|
||
"normalized_fragment_text": {
|
||
"type": "string",
|
||
"minLength": 1
|
||
},
|
||
"domain_relevance": {
|
||
"type": "string",
|
||
"enum": [
|
||
"in_scope",
|
||
"out_of_scope",
|
||
"unclear"
|
||
]
|
||
},
|
||
"business_scope": {
|
||
"type": "string",
|
||
"enum": [
|
||
"company_specific_accounting",
|
||
"generic_accounting",
|
||
"offtopic",
|
||
"unclear"
|
||
]
|
||
},
|
||
"entity_hints": {
|
||
"type": "array",
|
||
"items": {
|
||
"type": "string"
|
||
}
|
||
},
|
||
"account_hints": {
|
||
"type": "array",
|
||
"items": {
|
||
"type": "string"
|
||
}
|
||
},
|
||
"document_hints": {
|
||
"type": "array",
|
||
"items": {
|
||
"type": "string"
|
||
}
|
||
},
|
||
"register_hints": {
|
||
"type": "array",
|
||
"items": {
|
||
"type": "string"
|
||
}
|
||
},
|
||
"time_scope": {
|
||
"type": "object",
|
||
"additionalProperties": false,
|
||
"required": [
|
||
"type",
|
||
"value",
|
||
"confidence"
|
||
],
|
||
"properties": {
|
||
"type": {
|
||
"type": "string",
|
||
"enum": [
|
||
"explicit",
|
||
"inferred",
|
||
"missing"
|
||
]
|
||
},
|
||
"value": {
|
||
"type": [
|
||
"string",
|
||
"null"
|
||
]
|
||
},
|
||
"confidence": {
|
||
"type": "string",
|
||
"enum": [
|
||
"high",
|
||
"medium",
|
||
"low"
|
||
]
|
||
}
|
||
}
|
||
},
|
||
"flags": {
|
||
"type": "object",
|
||
"additionalProperties": false,
|
||
"required": [
|
||
"has_multi_entity_scope",
|
||
"asks_for_chain_explanation",
|
||
"asks_for_ranking_or_top",
|
||
"asks_for_period_summary",
|
||
"asks_for_rule_check",
|
||
"asks_for_anomaly_scan",
|
||
"asks_for_exact_object_trace",
|
||
"asks_for_evidence",
|
||
"mentions_period_close_context"
|
||
],
|
||
"properties": {
|
||
"has_multi_entity_scope": {
|
||
"type": "boolean"
|
||
},
|
||
"asks_for_chain_explanation": {
|
||
"type": "boolean"
|
||
},
|
||
"asks_for_ranking_or_top": {
|
||
"type": "boolean"
|
||
},
|
||
"asks_for_period_summary": {
|
||
"type": "boolean"
|
||
},
|
||
"asks_for_rule_check": {
|
||
"type": "boolean"
|
||
},
|
||
"asks_for_anomaly_scan": {
|
||
"type": "boolean"
|
||
},
|
||
"asks_for_exact_object_trace": {
|
||
"type": "boolean"
|
||
},
|
||
"asks_for_evidence": {
|
||
"type": "boolean"
|
||
},
|
||
"mentions_period_close_context": {
|
||
"type": "boolean"
|
||
}
|
||
}
|
||
},
|
||
"candidate_labels": {
|
||
"type": "array",
|
||
"items": {
|
||
"type": "string",
|
||
"enum": [
|
||
"heavy_analytical",
|
||
"cross_entity",
|
||
"drilldown_explain",
|
||
"rule_based_account_control",
|
||
"anomaly_probe",
|
||
"period_close_risk",
|
||
"ambiguous_human_query",
|
||
"simple_factual"
|
||
]
|
||
}
|
||
},
|
||
"confidence": {
|
||
"type": "string",
|
||
"enum": [
|
||
"high",
|
||
"medium",
|
||
"low"
|
||
]
|
||
},
|
||
"execution_readiness": {
|
||
"type": "string",
|
||
"enum": [
|
||
"executable",
|
||
"executable_with_soft_assumptions",
|
||
"needs_clarification",
|
||
"no_route"
|
||
]
|
||
},
|
||
"clarification_reason": {
|
||
"type": [
|
||
"string",
|
||
"null"
|
||
]
|
||
},
|
||
"soft_assumption_used": {
|
||
"type": "array",
|
||
"items": {
|
||
"type": "string",
|
||
"enum": [
|
||
"period_from_session_context",
|
||
"company_scope_defaulted",
|
||
"problem_scan_mode_enabled"
|
||
]
|
||
}
|
||
},
|
||
"route_status": {
|
||
"type": "string",
|
||
"enum": [
|
||
"routed",
|
||
"no_route"
|
||
]
|
||
},
|
||
"no_route_reason": {
|
||
"type": [
|
||
"string",
|
||
"null"
|
||
],
|
||
"enum": [
|
||
"out_of_scope",
|
||
"insufficient_specificity",
|
||
"missing_mapping",
|
||
"unsupported_fragment_type",
|
||
null
|
||
]
|
||
}
|
||
}
|
||
}
|
||
},
|
||
"discarded_fragments": {
|
||
"type": "array",
|
||
"items": {
|
||
"type": "object",
|
||
"additionalProperties": false,
|
||
"required": [
|
||
"raw_fragment_text",
|
||
"reason"
|
||
],
|
||
"properties": {
|
||
"raw_fragment_text": {
|
||
"type": "string",
|
||
"minLength": 1
|
||
},
|
||
"reason": {
|
||
"type": "string",
|
||
"minLength": 1
|
||
}
|
||
}
|
||
}
|
||
},
|
||
"global_notes": {
|
||
"type": "object",
|
||
"additionalProperties": false,
|
||
"required": [
|
||
"needs_clarification",
|
||
"clarification_reason"
|
||
],
|
||
"properties": {
|
||
"needs_clarification": {
|
||
"type": "boolean"
|
||
},
|
||
"clarification_reason": {
|
||
"type": [
|
||
"string",
|
||
"null"
|
||
]
|
||
}
|
||
}
|
||
}
|
||
}
|
||
},
|
||
"strict": true
|
||
},
|
||
"verbosity": "medium"
|
||
},
|
||
"tool_choice": "auto",
|
||
"tools": [],
|
||
"top_logprobs": 0,
|
||
"top_p": 1,
|
||
"truncation": "disabled",
|
||
"usage": {
|
||
"input_tokens": 1949,
|
||
"input_tokens_details": {
|
||
"cached_tokens": 0
|
||
},
|
||
"output_tokens": 445,
|
||
"output_tokens_details": {
|
||
"reasoning_tokens": 0
|
||
},
|
||
"total_tokens": 2394
|
||
},
|
||
"user": null,
|
||
"metadata": {}
|
||
},
|
||
"parsed_normalized_json": {
|
||
"schema_version": "normalized_query_v2_0_2",
|
||
"user_message_raw": "Есть ли в нашей базе куски, которые выглядят так, будто их кто-то руками “дособирал”, и теперь формально всё стоит, но смысловая логика учёта у этого шатается.",
|
||
"message_in_scope": true,
|
||
"scope_confidence": "high",
|
||
"contains_multiple_tasks": false,
|
||
"fragments": [
|
||
{
|
||
"fragment_id": "1",
|
||
"raw_fragment_text": "куски, которые выглядят так, будто их кто-то руками “дособирал”",
|
||
"normalized_fragment_text": "не бьется, куски, которые выглядят так, будто их кто-то руками дособирал",
|
||
"domain_relevance": "in_scope",
|
||
"business_scope": "generic_accounting",
|
||
"entity_hints": [],
|
||
"account_hints": [],
|
||
"document_hints": [],
|
||
"register_hints": [],
|
||
"time_scope": {
|
||
"type": "missing",
|
||
"value": null,
|
||
"confidence": "low"
|
||
},
|
||
"flags": {
|
||
"has_multi_entity_scope": true,
|
||
"asks_for_chain_explanation": true,
|
||
"asks_for_ranking_or_top": false,
|
||
"asks_for_period_summary": false,
|
||
"asks_for_rule_check": false,
|
||
"asks_for_anomaly_scan": false,
|
||
"asks_for_exact_object_trace": false,
|
||
"asks_for_evidence": false,
|
||
"mentions_period_close_context": false
|
||
},
|
||
"candidate_labels": [
|
||
"cross_entity",
|
||
"anomaly_probe"
|
||
],
|
||
"confidence": "high",
|
||
"execution_readiness": "executable",
|
||
"clarification_reason": null,
|
||
"soft_assumption_used": [],
|
||
"route_status": "routed",
|
||
"no_route_reason": null
|
||
}
|
||
],
|
||
"discarded_fragments": [],
|
||
"global_notes": {
|
||
"needs_clarification": false,
|
||
"clarification_reason": null
|
||
}
|
||
},
|
||
"validation_result": {
|
||
"passed": true,
|
||
"errors": []
|
||
},
|
||
"route_hint_summary": {
|
||
"mode": "deterministic_v2",
|
||
"message_in_scope": true,
|
||
"scope_confidence": "high",
|
||
"planner": {
|
||
"total_fragments": 1,
|
||
"in_scope_fragments": 1,
|
||
"out_of_scope_fragments": 0,
|
||
"discarded_fragments": 0,
|
||
"contains_multiple_tasks": false
|
||
},
|
||
"decisions": [
|
||
{
|
||
"fragment_id": "1",
|
||
"domain_relevance": "in_scope",
|
||
"business_scope": "generic_accounting",
|
||
"candidate_labels": [
|
||
"cross_entity",
|
||
"anomaly_probe"
|
||
],
|
||
"decision_flags": {
|
||
"has_multi_entity_scope": true,
|
||
"asks_for_chain_explanation": true,
|
||
"asks_for_ranking_or_top": false,
|
||
"asks_for_period_summary": false,
|
||
"asks_for_rule_check": false,
|
||
"asks_for_anomaly_scan": false,
|
||
"asks_for_exact_object_trace": false,
|
||
"asks_for_evidence": false,
|
||
"mentions_period_close_context": false
|
||
},
|
||
"execution_readiness": "executable",
|
||
"clarification_reason": null,
|
||
"soft_assumption_used": [],
|
||
"route_status": "routed",
|
||
"no_route_reason": null,
|
||
"route": "hybrid_store_plus_live",
|
||
"reason": "Multi-entity causal chain requested."
|
||
}
|
||
],
|
||
"fallback": {
|
||
"type": "none",
|
||
"message": null
|
||
}
|
||
},
|
||
"route_hint": "hybrid_store_plus_live",
|
||
"confidence": "high",
|
||
"usage": {
|
||
"input_tokens": 1949,
|
||
"output_tokens": 445,
|
||
"total_tokens": 2394
|
||
},
|
||
"latency_ms": 11894,
|
||
"eval_label": "eval-uQiZ1L_ga6",
|
||
"eval_mode": "single-pass-strict",
|
||
"request_count_for_case": 1
|
||
} |