ГЛОБАЛЬНЫЙ РЕФАКТОРИНГ АРХИТЕКТУРЫ - Рефакторинг этапов 3.1 - Добавлен строгий контракт семантического извлечения (валидность, качество, reason-codes, рекомендация применять canonical). Подключен semantic guard в predecompose/runtime: мягкое отклонение плохого canonical: отдельный safe-кейс для сырого полезного raw-фрагмента; fallback теперь выигрывает только когда это оправдано. Подключty semantic arbitration в tool-gate

This commit is contained in:
2026-04-11 16:45:43 +03:00
parent 66402439dc
commit a7ccc92a9c
18 changed files with 750 additions and 26 deletions
@@ -1047,14 +1047,16 @@ describe("assistant address llm pre-decompose candidate preference", () => {
} as any);
expect(response.ok).toBe(true);
expect(response.reply_type).toBe("factual");
expect(calls).toHaveLength(1);
expect(calls[0].message).toBe("заказчики компании svk");
expect(response.debug?.llm_decomposition_attempted).toBe(true);
expect(response.debug?.llm_decomposition_applied).toBe(true);
expect(response.debug?.llm_canonical_candidate_detected).toBe(true);
expect(response.debug?.tool_gate_decision).toBe("run_address_lane");
expect(["llm_canonical_candidate_detected", "llm_canonical_data_signal_detected", "address_mode_classifier_detected"]).toContain(response.debug?.tool_gate_reason);
expect(response.reply_type).toBe("clarification_required");
expect(calls).toHaveLength(0);
expect(response.debug?.address_tool_gate_decision).toBe("skip_address_lane");
expect(
[
"llm_predecompose_semantic_guard_rejected",
"llm_predecompose_unsupported_mode",
"address_signal_unsupported_intent_fallback_to_deep"
]
).toContain(response.debug?.address_tool_gate_reason);
});
it("normalizes short ordinal year like '20й' in noisy docs phrasing", async () => {
@@ -0,0 +1,62 @@
import { describe, expect, it } from "vitest";
import {
buildAddressLlmPredecomposeContractV1,
buildAddressSemanticExtractionContractV1
} from "../src/services/address_runtime/predecomposeContract";
describe("address semantic extraction contract", () => {
it("rejects low-confidence unsupported rewrite without data signal", () => {
const sourceMessage = "yo";
const canonicalMessage = "yoft";
const predecomposeContract = buildAddressLlmPredecomposeContractV1({
sourceMessage,
canonicalMessage
});
const semantic = buildAddressSemanticExtractionContractV1({
sourceMessage,
canonicalMessage,
predecomposeContract
});
expect(semantic.schema_version).toBe("address_semantic_extraction_contract_v1");
expect(semantic.guard_hints.source_data_signal_detected).toBe(false);
expect(semantic.guard_hints.canonical_data_signal_detected).toBe(false);
expect(semantic.guard_hints.unsupported_low_confidence).toBe(true);
expect(semantic.valid).toBe(false);
expect(semantic.apply_canonical_recommended).toBe(false);
expect(semantic.reason_codes).toContain("unsupported_low_confidence_contract");
});
it("flags semantic drift when canonical loses data intent", () => {
const sourceMessage = "покажи документы по договору 12";
const canonicalMessage = "помоги разобраться";
const semantic = buildAddressSemanticExtractionContractV1({
sourceMessage,
canonicalMessage
});
expect(semantic.guard_hints.source_data_signal_detected).toBe(true);
expect(semantic.guard_hints.canonical_data_signal_detected).toBe(false);
expect(semantic.guard_hints.semantic_drift_suspected).toBe(true);
expect(semantic.valid).toBe(false);
expect(semantic.apply_canonical_recommended).toBe(false);
expect(semantic.reason_codes).toContain("semantic_drift_source_vs_canonical");
});
it("keeps canonical rewrite when semantic contract remains coherent", () => {
const sourceMessage = "Покажи незакрытые договоры на 2020-12-31";
const canonicalMessage = "Показать незакрытые договоры по состоянию на конец декабря 2020 года.";
const semantic = buildAddressSemanticExtractionContractV1({
sourceMessage,
canonicalMessage
});
expect(semantic.guard_hints.source_data_signal_detected).toBe(true);
expect(semantic.guard_hints.canonical_data_signal_detected).toBe(true);
expect(semantic.valid).toBe(true);
expect(semantic.apply_canonical_recommended).toBe(true);
expect(["high", "medium"]).toContain(semantic.quality);
});
});