@su-record/vibe 3.2.64 → 4.0.1

package/dist/infra/lib/constants.js

@@ -1,97 +0,0 @@
-/**
- * VIBE 전역 상수 정의
- */
-// 타임아웃 설정 (ms)
-export const TIMEOUTS = {
-    RESEARCH: 180000, // 3분
-    PYTHON_PARSE: 30000, // 30초
-    DEFAULT: 120000, // 2분
-};
-// 캐시 설정
-export const CACHE = {
-    MAX_SIZE: 5,
-    TTL: 5 * 60 * 1000, // 5분
-    MAX_TOTAL_MEMORY_MB: 200,
-    INSTRUCTIONS_TTL: 15 * 60 * 1000, // 15분
-};
-// Agent 설정
-export const AGENT = {
-    MAX_TURNS: 10,
-    MAX_CONCURRENCY: 4,
-    DEFAULT_ALLOWED_TOOLS: ['Read', 'Write', 'Edit', 'Glob', 'Grep', 'Bash'],
-};
-// Concurrency 설정
-export const CONCURRENCY = {
-    /** 모델별 동시 실행 제한 — 키는 orchestrator/types.ts ClaudeModel 과 동일해야 한다 */
-    MODEL_LIMITS: {
-        'claude-opus-4-8': 3,
-        'claude-sonnet-5': 5,
-        'claude-haiku-4-5': 8,
-        'default': 5,
-    },
-    /** 프로바이더별 동시 실행 제한 */
-    PROVIDER_LIMITS: {
-        'claude': 10,
-        'gpt': 5,
-        'antigravity': 5,
-        'default': 10,
-    },
-    /** 기본 동시 실행 제한 */
-    DEFAULT: 5,
-    /** 큐 최대 크기 */
-    QUEUE_MAX_SIZE: 100,
-    /** 개별 태스크 타임아웃 (ms) */
-    TASK_TIMEOUT: 300000, // 5분
-    /** 전체 파이프라인 타임아웃 (ms) */
-    PIPELINE_TIMEOUT: 900000, // 15분
-    /** 태스크 실패 시 최대 재시도 횟수 */
-    MAX_RETRIES: 3,
-    /** 재시도 간격 (ms) — 지수 백오프 적용 */
-    RETRY_BASE_DELAY: 2000, // 2초
-    /** 활동 기반 타임아웃 — 무활동 감지 (ms) */
-    ACTIVITY_TIMEOUT: 180_000, // 3분
-    /** Stale 감지 주기 (ms) */
-    STALE_CHECK_INTERVAL: 30_000, // 30초
-    /** 멀티 메시지 배칭 대기 시간 (ms) — Phase 3 */
-    BATCH_WAIT_MS: 2_000, // 2초
-    /** 세션당 최대 인스트럭션 주입 횟수 — Phase 4 */
-    MAX_INJECTION_PER_SESSION: 3,
-    /** 대화 이력 조회 기간 (시간) — Phase 5 */
-    CONVERSATION_HISTORY_HOURS: 24,
-    /** 대화 이력 최대 문자 수 — Phase 5 */
-    CONVERSATION_HISTORY_MAX_CHARS: 8_000,
-    /** 대화 이력 정리 기준 (시간) — Phase 5 */
-    CONVERSATION_CLEANUP_HOURS: 48,
-};
-// 토큰 예산 설정
-export const TOKENS = {
-    PER_CHAR_ESTIMATE: 0.25,
-    DEFAULT_BUDGET: 128_000,
-    DEFAULT_TARGET: 4000,
-};
-// sonolbot-v2 상수
-export const MESSAGING = {
-    /** Phase 1: ProgressReporter 메시지 편집 최소 간격 (Telegram rate limit 방지) */
-    PROGRESS_MIN_INTERVAL_MS: 3_000,
-    /** Phase 2: chatId당 대기 큐 최대 크기 */
-    MAX_PENDING_MESSAGES: 10,
-    /** Phase 2: 대기 메시지 TTL */
-    PENDING_MESSAGE_TTL_MS: 300_000,
-    /** Phase 2: process당 최대 injection 횟수 */
-    MAX_INJECTION_PER_PROCESS: 3,
-    /** Phase 2: injection content 최대 문자수 (LLM context overflow 방지) */
-    MAX_INJECTION_CHARS: 4_000,
-    /** Phase 2: 새 요청 확인 알림 debounce */
-    ACK_DEBOUNCE_MS: 5_000,
-    /** Phase 3: 무활동 감지 타임아웃 */
-    ACTIVITY_TIMEOUT_MS: 600_000,
-    /** Phase 3: Stale 감지 주기 */
-    STALE_CHECK_INTERVAL_MS: 60_000,
-    /** Phase 3: stale 후 자동 재처리 최대 횟수 (poison pill 방지) */
-    MAX_STALE_RETRY: 1,
-    /** Phase 3: DevSessionManager lock 대기 최대 시간 */
-    LOCK_WAIT_TIMEOUT_MS: 30_000,
-    /** Phase 4: 외부 채널 기본 포맷 */
-    DEFAULT_EXTERNAL_FORMAT: 'text',
-};
-//# sourceMappingURL=constants.js.map

package/dist/infra/lib/constants.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"constants.js","sourceRoot":"","sources":["../../../src/infra/lib/constants.ts"],"names":[],"mappings":"AAAA;;GAEG;AAEH,eAAe;AACf,MAAM,CAAC,MAAM,QAAQ,GAAG;IACtB,QAAQ,EAAE,MAAM,EAAO,KAAK;IAC5B,YAAY,EAAE,KAAK,EAAI,MAAM;IAC7B,OAAO,EAAE,MAAM,EAAQ,KAAK;CACpB,CAAC;AAEX,QAAQ;AACR,MAAM,CAAC,MAAM,KAAK,GAAG;IACnB,QAAQ,EAAE,CAAC;IACX,GAAG,EAAE,CAAC,GAAG,EAAE,GAAG,IAAI,EAAY,KAAK;IACnC,mBAAmB,EAAE,GAAG;IACxB,gBAAgB,EAAE,EAAE,GAAG,EAAE,GAAG,IAAI,EAAG,MAAM;CACjC,CAAC;AAEX,WAAW;AACX,MAAM,CAAC,MAAM,KAAK,GAAG;IACnB,SAAS,EAAE,EAAE;IACb,eAAe,EAAE,CAAC;IAClB,qBAAqB,EAAE,CAAC,MAAM,EAAE,OAAO,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,CAAa;CACrF,CAAC;AAEF,iBAAiB;AACjB,MAAM,CAAC,MAAM,WAAW,GAAG;IACzB,oEAAoE;IACpE,YAAY,EAAE;QACZ,iBAAiB,EAAE,CAAC;QACpB,iBAAiB,EAAE,CAAC;QACpB,kBAAkB,EAAE,CAAC;QACrB,SAAS,EAAE,CAAC;KACa;IAC3B,sBAAsB;IACtB,eAAe,EAAE;QACf,QAAQ,EAAE,EAAE;QACZ,KAAK,EAAE,CAAC;QACR,aAAa,EAAE,CAAC;QAChB,SAAS,EAAE,EAAE;KACY;IAC3B,kBAAkB;IAClB,OAAO,EAAE,CAAC;IACV,cAAc;IACd,cAAc,EAAE,GAAG;IACnB,uBAAuB;IACvB,YAAY,EAAE,MAAM,EAAE,KAAK;IAC3B,yBAAyB;IACzB,gBAAgB,EAAE,MAAM,EAAE,MAAM;IAChC,yBAAyB;IACzB,WAAW,EAAE,CAAC;IACd,8BAA8B;IAC9B,gBAAgB,EAAE,IAAI,EAAE,KAAK;IAC7B,+BAA+B;IAC/B,gBAAgB,EAAE,OAAO,EAAE,KAAK;IAChC,uBAAuB;IACvB,oBAAoB,EAAE,MAAM,EAAE,MAAM;IACpC,qCAAqC;IACrC,aAAa,EAAE,KAAK,EAAE,KAAK;IAC3B,mCAAmC;IACnC,yBAAyB,EAAE,CAAC;IAC5B,iCAAiC;IACjC,0BAA0B,EAAE,EAAE;IAC9B,8BAA8B;IAC9B,8BAA8B,EAAE,KAAK;IACrC,iCAAiC;IACjC,0BAA0B,EAAE,EAAE;CACtB,CAAC;AAEX,WAAW;AACX,MAAM,CAAC,MAAM,MAAM,GAAG;IACpB,iBAAiB,EAAE,IAAI;IACvB,cAAc,EAAE,OAAO;IACvB,cAAc,EAAE,IAAI;CACZ,CAAC;AAEX,iBAAiB;AACjB,MAAM,CAAC,MAAM,SAAS,GAAG;IACvB,sEAAsE;IACtE,wBAAwB,EAAE,KAAK;IAC/B,kCAAkC;IAClC,oBAAoB,EAAE,EAAE;IACxB,0BAA0B;IAC1B,sBAAsB,EAAE,OAAO;IAC/B,wCAAwC;IACxC,yBAAyB,EAAE,CAAC;IAC5B,kEAAkE;IAClE,mBAAmB,EAAE,KAAK;IAC1B,mCAAmC;IACnC,eAAe,EAAE,KAAK;IACtB,2BAA2B;IAC3B,mBAAmB,EAAE,OAAO;IAC5B,2BAA2B;IAC3B,uBAAuB,EAAE,MAAM;IAC/B,qDAAqD;IACrD,eAAe,EAAE,CAAC;IAClB,+CAA+C;IAC/C,oBAAoB,EAAE,MAAM;IAC5B,2BAA2B;IAC3B,uBAAuB,EAAE,MAAe;CAChC,CAAC"}

package/dist/infra/lib/embedding/EmbeddingProvider.d.ts

@@ -1,20 +0,0 @@
-/**
- * EmbeddingProvider — GPT 임베딩 API 클라이언트
- *
- * API 키 없으면 isAvailable() = false → graceful degradation.
- */
-import type { EmbeddingResponse, EmbeddingProviderConfig } from './types.js';
-export declare class EmbeddingProvider {
-    private config;
-    constructor(config?: Partial<EmbeddingProviderConfig>);
-    /**
-     * API 키가 설정되어 있는지 확인
-     */
-    isAvailable(): boolean;
-    /**
-     * 텍스트 배열을 임베딩 벡터로 변환
-     */
-    embed(texts: string[]): Promise<EmbeddingResponse>;
-    private callGptEmbedding;
-}
-//# sourceMappingURL=EmbeddingProvider.d.ts.map

package/dist/infra/lib/embedding/EmbeddingProvider.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"EmbeddingProvider.d.ts","sourceRoot":"","sources":["../../../../src/infra/lib/embedding/EmbeddingProvider.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AAEH,OAAO,KAAK,EACV,iBAAiB,EACjB,uBAAuB,EACxB,MAAM,YAAY,CAAC;AASpB,qBAAa,iBAAiB;IAC5B,OAAO,CAAC,MAAM,CAA0B;gBAE5B,MAAM,CAAC,EAAE,OAAO,CAAC,uBAAuB,CAAC;IAOrD;;OAEG;IACI,WAAW,IAAI,OAAO;IAI7B;;OAEG;IACU,KAAK,CAAC,KAAK,EAAE,MAAM,EAAE,GAAG,OAAO,CAAC,iBAAiB,CAAC;YAcjD,gBAAgB;CAuD/B"}

package/dist/infra/lib/embedding/EmbeddingProvider.js

@@ -1,83 +0,0 @@
-/**
- * EmbeddingProvider — GPT 임베딩 API 클라이언트
- *
- * API 키 없으면 isAvailable() = false → graceful degradation.
- */
-import { DEFAULT_EMBEDDING_DIMENSIONS, EMBEDDING_API_TIMEOUT_MS, GPT_EMBEDDING_BASE_URL, GPT_EMBEDDING_MODEL, } from './types.js';
-import { getGptApiKey } from '../config/GlobalConfigManager.js';
-export class EmbeddingProvider {
-    config;
-    constructor(config) {
-        this.config = {
-            priority: config?.priority ?? ['gpt'],
-            dimensions: config?.dimensions ?? DEFAULT_EMBEDDING_DIMENSIONS,
-        };
-    }
-    /**
-     * API 키가 설정되어 있는지 확인
-     */
-    isAvailable() {
-        return Boolean(getGptApiKey() || process.env.OPENAI_API_KEY);
-    }
-    /**
-     * 텍스트 배열을 임베딩 벡터로 변환
-     */
-    async embed(texts) {
-        if (texts.length === 0) {
-            return { embeddings: [], model: '' };
-        }
-        if (!this.isAvailable()) {
-            throw new Error('All embedding providers failed: OPENAI_API_KEY not set');
-        }
-        return this.callGptEmbedding(texts);
-    }
-    async callGptEmbedding(texts) {
-        const apiKey = getGptApiKey() || process.env.OPENAI_API_KEY;
-        if (!apiKey)
-            throw new Error('OPENAI_API_KEY not set');
-        const controller = new AbortController();
-        const timeout = setTimeout(() => controller.abort(), EMBEDDING_API_TIMEOUT_MS);
-        try {
-            const response = await fetch(`${GPT_EMBEDDING_BASE_URL}/embeddings`, {
-                method: 'POST',
-                headers: {
-                    'Authorization': `Bearer ${apiKey}`,
-                    'Content-Type': 'application/json',
-                },
-                body: JSON.stringify({
-                    input: texts,
-                    model: GPT_EMBEDDING_MODEL,
-                    dimensions: this.config.dimensions,
-                }),
-                signal: controller.signal,
-            });
-            if (!response.ok) {
-                const errorText = await response.text();
-                let errorMessage = `GPT Embedding API error (${response.status})`;
-                try {
-                    const errorJson = JSON.parse(errorText);
-                    if (errorJson.error?.message) {
-                        errorMessage = errorJson.error.message;
-                    }
-                }
-                catch { /* ignore parse error */ }
-                throw new Error(errorMessage);
-            }
-            const result = await response.json();
-            return {
-                embeddings: result.data.map(d => d.embedding),
-                model: result.model,
-            };
-        }
-        catch (error) {
-            if (error instanceof Error && error.name === 'AbortError') {
-                throw new Error(`GPT Embedding API timeout (${EMBEDDING_API_TIMEOUT_MS / 1000}s)`);
-            }
-            throw error;
-        }
-        finally {
-            clearTimeout(timeout);
-        }
-    }
-}
-//# sourceMappingURL=EmbeddingProvider.js.map

package/dist/infra/lib/embedding/EmbeddingProvider.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"EmbeddingProvider.js","sourceRoot":"","sources":["../../../../src/infra/lib/embedding/EmbeddingProvider.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AAMH,OAAO,EACL,4BAA4B,EAC5B,wBAAwB,EACxB,sBAAsB,EACtB,mBAAmB,GACpB,MAAM,YAAY,CAAC;AACpB,OAAO,EAAE,YAAY,EAAE,MAAM,kCAAkC,CAAC;AAEhE,MAAM,OAAO,iBAAiB;IACpB,MAAM,CAA0B;IAExC,YAAY,MAAyC;QACnD,IAAI,CAAC,MAAM,GAAG;YACZ,QAAQ,EAAE,MAAM,EAAE,QAAQ,IAAI,CAAC,KAAK,CAAC;YACrC,UAAU,EAAE,MAAM,EAAE,UAAU,IAAI,4BAA4B;SAC/D,CAAC;IACJ,CAAC;IAED;;OAEG;IACI,WAAW;QAChB,OAAO,OAAO,CAAC,YAAY,EAAE,IAAI,OAAO,CAAC,GAAG,CAAC,cAAc,CAAC,CAAC;IAC/D,CAAC;IAED;;OAEG;IACI,KAAK,CAAC,KAAK,CAAC,KAAe;QAChC,IAAI,KAAK,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;YACvB,OAAO,EAAE,UAAU,EAAE,EAAE,EAAE,KAAK,EAAE,EAAE,EAAE,CAAC;QACvC,CAAC;QAED,IAAI,CAAC,IAAI,CAAC,WAAW,EAAE,EAAE,CAAC;YACxB,MAAM,IAAI,KAAK,CACb,wDAAwD,CACzD,CAAC;QACJ,CAAC;QAED,OAAO,IAAI,CAAC,gBAAgB,CAAC,KAAK,CAAC,CAAC;IACtC,CAAC;IAEO,KAAK,CAAC,gBAAgB,CAAC,KAAe;QAC5C,MAAM,MAAM,GAAG,YAAY,EAAE,IAAI,OAAO,CAAC,GAAG,CAAC,cAAc,CAAC;QAC5D,IAAI,CAAC,MAAM;YAAE,MAAM,IAAI,KAAK,CAAC,wBAAwB,CAAC,CAAC;QAEvD,MAAM,UAAU,GAAG,IAAI,eAAe,EAAE,CAAC;QACzC,MAAM,OAAO,GAAG,UAAU,CACxB,GAAG,EAAE,CAAC,UAAU,CAAC,KAAK,EAAE,EACxB,wBAAwB,CACzB,CAAC;QAEF,IAAI,CAAC;YACH,MAAM,QAAQ,GAAG,MAAM,KAAK,CAAC,GAAG,sBAAsB,aAAa,EAAE;gBACnE,MAAM,EAAE,MAAM;gBACd,OAAO,EAAE;oBACP,eAAe,EAAE,UAAU,MAAM,EAAE;oBACnC,cAAc,EAAE,kBAAkB;iBACnC;gBACD,IAAI,EAAE,IAAI,CAAC,SAAS,CAAC;oBACnB,KAAK,EAAE,KAAK;oBACZ,KAAK,EAAE,mBAAmB;oBAC1B,UAAU,EAAE,IAAI,CAAC,MAAM,CAAC,UAAU;iBACnC,CAAC;gBACF,MAAM,EAAE,UAAU,CAAC,MAAM;aAC1B,CAAC,CAAC;YAEH,IAAI,CAAC,QAAQ,CAAC,EAAE,EAAE,CAAC;gBACjB,MAAM,SAAS,GAAG,MAAM,QAAQ,CAAC,IAAI,EAAE,CAAC;gBACxC,IAAI,YAAY,GAAG,4BAA4B,QAAQ,CAAC,MAAM,GAAG,CAAC;gBAClE,IAAI,CAAC;oBACH,MAAM,SAAS,GAAG,IAAI,CAAC,KAAK,CAAC,SAAS,CAAqC,CAAC;oBAC5E,IAAI,SAAS,CAAC,KAAK,EAAE,OAAO,EAAE,CAAC;wBAC7B,YAAY,GAAG,SAAS,CAAC,KAAK,CAAC,OAAO,CAAC;oBACzC,CAAC;gBACH,CAAC;gBAAC,MAAM,CAAC,CAAC,wBAAwB,CAAC,CAAC;gBACpC,MAAM,IAAI,KAAK,CAAC,YAAY,CAAC,CAAC;YAChC,CAAC;YAED,MAAM,MAAM,GAAG,MAAM,QAAQ,CAAC,IAAI,EAGjC,CAAC;YAEF,OAAO;gBACL,UAAU,EAAE,MAAM,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,SAAS,CAAC;gBAC7C,KAAK,EAAE,MAAM,CAAC,KAAK;aACpB,CAAC;QACJ,CAAC;QAAC,OAAO,KAAK,EAAE,CAAC;YACf,IAAI,KAAK,YAAY,KAAK,IAAI,KAAK,CAAC,IAAI,KAAK,YAAY,EAAE,CAAC;gBAC1D,MAAM,IAAI,KAAK,CAAC,8BAA8B,wBAAwB,GAAG,IAAI,IAAI,CAAC,CAAC;YACrF,CAAC;YACD,MAAM,KAAK,CAAC;QACd,CAAC;gBAAS,CAAC;YACT,YAAY,CAAC,OAAO,CAAC,CAAC;QACxB,CAAC;IACH,CAAC;CACF"}

package/dist/infra/lib/embedding/VectorStore.d.ts

@@ -1,54 +0,0 @@
-/**
- * VectorStore — SQLite 기반 벡터 저장소
- *
- * 메모리/세션 엔티티의 임베딩 벡터를 BLOB으로 저장.
- * 검색: 행 단위 iterate → 코사인 유사도 → 스트리밍 top-k 선택.
- * (메모리 수 천~만 건 수준이므로 인덱스 없는 선형 스캔으로 충분하나,
- *  전량 적재+전체 정렬은 피해 메모리 O(limit)·정렬 비용 제거)
- */
-import Database from 'better-sqlite3';
-import type { VectorSearchResult, SessionVectorSearchResult } from './types.js';
-export declare class VectorStore {
-    private db;
-    constructor(db: Database.Database);
-    private initializeTables;
-    /**
-     * 메모리 벡터 저장 (upsert)
-     */
-    saveMemoryVector(key: string, embedding: number[]): void;
-    /**
-     * 세션 엔티티 벡터 저장 (upsert)
-     */
-    saveSessionVector(entityType: string, entityId: number, embedding: number[]): void;
-    /**
-     * 메모리 벡터 삭제
-     */
-    deleteMemoryVector(key: string): void;
-    /**
-     * 세션 엔티티 벡터 삭제
-     */
-    deleteSessionVector(entityType: string, entityId: number): void;
-    /**
-     * 스트리밍 top-k 선택 — 전 행을 배열로 적재해 전체 정렬(O(V log V) + O(V) 메모리)
-     * 하는 대신, iterate()로 한 행씩 스코어링하며 크기 limit의 정렬 배열만 유지한다.
-     * limit은 통상 ≤20이므로 선형 삽입으로 충분 (O(V·limit), 메모리 O(limit)).
-     */
-    private static topKBySimilarity;
-    /**
-     * 메모리 벡터 코사인 유사도 검색
-     */
-    searchMemoryVectors(queryVec: number[], limit?: number): VectorSearchResult[];
-    /**
-     * 세션 엔티티 벡터 코사인 유사도 검색
-     */
-    searchSessionVectors(entityType: string, queryVec: number[], limit?: number): SessionVectorSearchResult[];
-    /**
-     * 저장된 메모리 벡터 수
-     */
-    getMemoryVectorCount(): number;
-    /**
-     * 저장된 세션 벡터 수
-     */
-    getSessionVectorCount(entityType?: string): number;
-}
-//# sourceMappingURL=VectorStore.d.ts.map

package/dist/infra/lib/embedding/VectorStore.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"VectorStore.d.ts","sourceRoot":"","sources":["../../../../src/infra/lib/embedding/VectorStore.ts"],"names":[],"mappings":"AAAA;;;;;;;GAOG;AAEH,OAAO,QAAQ,MAAM,gBAAgB,CAAC;AAMtC,OAAO,KAAK,EAAE,kBAAkB,EAAE,yBAAyB,EAAE,MAAM,YAAY,CAAC;AAEhF,qBAAa,WAAW;IACtB,OAAO,CAAC,EAAE,CAAoB;gBAElB,EAAE,EAAE,QAAQ,CAAC,QAAQ;IAKjC,OAAO,CAAC,gBAAgB;IAoBxB;;OAEG;IACI,gBAAgB,CAAC,GAAG,EAAE,MAAM,EAAE,SAAS,EAAE,MAAM,EAAE,GAAG,IAAI;IAU/D;;OAEG;IACI,iBAAiB,CACtB,UAAU,EAAE,MAAM,EAClB,QAAQ,EAAE,MAAM,EAChB,SAAS,EAAE,MAAM,EAAE,GAClB,IAAI;IAUP;;OAEG;IACI,kBAAkB,CAAC,GAAG,EAAE,MAAM,GAAG,IAAI;IAI5C;;OAEG;IACI,mBAAmB,CAAC,UAAU,EAAE,MAAM,EAAE,QAAQ,EAAE,MAAM,GAAG,IAAI;IAMtE;;;;OAIG;IACH,OAAO,CAAC,MAAM,CAAC,gBAAgB;IAqB/B;;OAEG;IACI,mBAAmB,CACxB,QAAQ,EAAE,MAAM,EAAE,EAClB,KAAK,GAAE,MAAW,GACjB,kBAAkB,EAAE;IAgBvB;;OAEG;IACI,oBAAoB,CACzB,UAAU,EAAE,MAAM,EAClB,QAAQ,EAAE,MAAM,EAAE,EAClB,KAAK,GAAE,MAAW,GACjB,yBAAyB,EAAE;IAiB9B;;OAEG;IACI,oBAAoB,IAAI,MAAM;IAOrC;;OAEG;IACI,qBAAqB,CAAC,UAAU,CAAC,EAAE,MAAM,GAAG,MAAM;CAY1D"}

package/dist/infra/lib/embedding/VectorStore.js

@@ -1,139 +0,0 @@
-/**
- * VectorStore — SQLite 기반 벡터 저장소
- *
- * 메모리/세션 엔티티의 임베딩 벡터를 BLOB으로 저장.
- * 검색: 행 단위 iterate → 코사인 유사도 → 스트리밍 top-k 선택.
- * (메모리 수 천~만 건 수준이므로 인덱스 없는 선형 스캔으로 충분하나,
- *  전량 적재+전체 정렬은 피해 메모리 O(limit)·정렬 비용 제거)
- */
-import { cosineSimilarity, serializeVector, deserializeVector, } from './cosine.js';
-export class VectorStore {
-    db;
-    constructor(db) {
-        this.db = db;
-        this.initializeTables();
-    }
-    initializeTables() {
-        this.db.exec(`
-      CREATE TABLE IF NOT EXISTS memory_vectors (
-        key TEXT PRIMARY KEY,
-        embedding BLOB NOT NULL,
-        dimension INTEGER NOT NULL,
-        updatedAt TEXT NOT NULL
-      );
-
-      CREATE TABLE IF NOT EXISTS session_vectors (
-        entityType TEXT NOT NULL,
-        entityId INTEGER NOT NULL,
-        embedding BLOB NOT NULL,
-        dimension INTEGER NOT NULL,
-        updatedAt TEXT NOT NULL,
-        PRIMARY KEY (entityType, entityId)
-      );
-    `);
-    }
-    /**
-     * 메모리 벡터 저장 (upsert)
-     */
-    saveMemoryVector(key, embedding) {
-        const blob = serializeVector(embedding);
-        const now = new Date().toISOString();
-        this.db.prepare(`
-      INSERT OR REPLACE INTO memory_vectors (key, embedding, dimension, updatedAt)
-      VALUES (?, ?, ?, ?)
-    `).run(key, blob, embedding.length, now);
-    }
-    /**
-     * 세션 엔티티 벡터 저장 (upsert)
-     */
-    saveSessionVector(entityType, entityId, embedding) {
-        const blob = serializeVector(embedding);
-        const now = new Date().toISOString();
-        this.db.prepare(`
-      INSERT OR REPLACE INTO session_vectors (entityType, entityId, embedding, dimension, updatedAt)
-      VALUES (?, ?, ?, ?, ?)
-    `).run(entityType, entityId, blob, embedding.length, now);
-    }
-    /**
-     * 메모리 벡터 삭제
-     */
-    deleteMemoryVector(key) {
-        this.db.prepare('DELETE FROM memory_vectors WHERE key = ?').run(key);
-    }
-    /**
-     * 세션 엔티티 벡터 삭제
-     */
-    deleteSessionVector(entityType, entityId) {
-        this.db.prepare('DELETE FROM session_vectors WHERE entityType = ? AND entityId = ?').run(entityType, entityId);
-    }
-    /**
-     * 스트리밍 top-k 선택 — 전 행을 배열로 적재해 전체 정렬(O(V log V) + O(V) 메모리)
-     * 하는 대신, iterate()로 한 행씩 스코어링하며 크기 limit의 정렬 배열만 유지한다.
-     * limit은 통상 ≤20이므로 선형 삽입으로 충분 (O(V·limit), 메모리 O(limit)).
-     */
-    static topKBySimilarity(rows, limit, score) {
-        const best = [];
-        for (const row of rows) {
-            const item = score(row);
-            if (best.length >= limit && item.similarity <= best[best.length - 1].similarity) {
-                continue;
-            }
-            let insertAt = best.length;
-            while (insertAt > 0 && best[insertAt - 1].similarity < item.similarity) {
-                insertAt--;
-            }
-            best.splice(insertAt, 0, item);
-            if (best.length > limit)
-                best.pop();
-        }
-        return best;
-    }
-    /**
-     * 메모리 벡터 코사인 유사도 검색
-     */
-    searchMemoryVectors(queryVec, limit = 20) {
-        const rows = this.db.prepare('SELECT key, embedding FROM memory_vectors').iterate();
-        const queryFloat = new Float32Array(queryVec);
-        return VectorStore.topKBySimilarity(rows, limit, (raw) => {
-            const row = raw;
-            return {
-                key: row.key,
-                similarity: cosineSimilarity(queryFloat, deserializeVector(row.embedding)),
-            };
-        });
-    }
-    /**
-     * 세션 엔티티 벡터 코사인 유사도 검색
-     */
-    searchSessionVectors(entityType, queryVec, limit = 20) {
-        const rows = this.db.prepare('SELECT entityId, embedding FROM session_vectors WHERE entityType = ?').iterate(entityType);
-        const queryFloat = new Float32Array(queryVec);
-        return VectorStore.topKBySimilarity(rows, limit, (raw) => {
-            const row = raw;
-            return {
-                entityType,
-                entityId: row.entityId,
-                similarity: cosineSimilarity(queryFloat, deserializeVector(row.embedding)),
-            };
-        });
-    }
-    /**
-     * 저장된 메모리 벡터 수
-     */
-    getMemoryVectorCount() {
-        const row = this.db.prepare('SELECT COUNT(*) as cnt FROM memory_vectors').get();
-        return row.cnt;
-    }
-    /**
-     * 저장된 세션 벡터 수
-     */
-    getSessionVectorCount(entityType) {
-        if (entityType) {
-            const row = this.db.prepare('SELECT COUNT(*) as cnt FROM session_vectors WHERE entityType = ?').get(entityType);
-            return row.cnt;
-        }
-        const row = this.db.prepare('SELECT COUNT(*) as cnt FROM session_vectors').get();
-        return row.cnt;
-    }
-}
-//# sourceMappingURL=VectorStore.js.map

package/dist/infra/lib/embedding/VectorStore.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"VectorStore.js","sourceRoot":"","sources":["../../../../src/infra/lib/embedding/VectorStore.ts"],"names":[],"mappings":"AAAA;;;;;;;GAOG;AAGH,OAAO,EACL,gBAAgB,EAChB,eAAe,EACf,iBAAiB,GAClB,MAAM,aAAa,CAAC;AAGrB,MAAM,OAAO,WAAW;IACd,EAAE,CAAoB;IAE9B,YAAY,EAAqB;QAC/B,IAAI,CAAC,EAAE,GAAG,EAAE,CAAC;QACb,IAAI,CAAC,gBAAgB,EAAE,CAAC;IAC1B,CAAC;IAEO,gBAAgB;QACtB,IAAI,CAAC,EAAE,CAAC,IAAI,CAAC;;;;;;;;;;;;;;;;KAgBZ,CAAC,CAAC;IACL,CAAC;IAED;;OAEG;IACI,gBAAgB,CAAC,GAAW,EAAE,SAAmB;QACtD,MAAM,IAAI,GAAG,eAAe,CAAC,SAAS,CAAC,CAAC;QACxC,MAAM,GAAG,GAAG,IAAI,IAAI,EAAE,CAAC,WAAW,EAAE,CAAC;QAErC,IAAI,CAAC,EAAE,CAAC,OAAO,CAAC;;;KAGf,CAAC,CAAC,GAAG,CAAC,GAAG,EAAE,IAAI,EAAE,SAAS,CAAC,MAAM,EAAE,GAAG,CAAC,CAAC;IAC3C,CAAC;IAED;;OAEG;IACI,iBAAiB,CACtB,UAAkB,EAClB,QAAgB,EAChB,SAAmB;QAEnB,MAAM,IAAI,GAAG,eAAe,CAAC,SAAS,CAAC,CAAC;QACxC,MAAM,GAAG,GAAG,IAAI,IAAI,EAAE,CAAC,WAAW,EAAE,CAAC;QAErC,IAAI,CAAC,EAAE,CAAC,OAAO,CAAC;;;KAGf,CAAC,CAAC,GAAG,CAAC,UAAU,EAAE,QAAQ,EAAE,IAAI,EAAE,SAAS,CAAC,MAAM,EAAE,GAAG,CAAC,CAAC;IAC5D,CAAC;IAED;;OAEG;IACI,kBAAkB,CAAC,GAAW;QACnC,IAAI,CAAC,EAAE,CAAC,OAAO,CAAC,0CAA0C,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC;IACvE,CAAC;IAED;;OAEG;IACI,mBAAmB,CAAC,UAAkB,EAAE,QAAgB;QAC7D,IAAI,CAAC,EAAE,CAAC,OAAO,CACb,mEAAmE,CACpE,CAAC,GAAG,CAAC,UAAU,EAAE,QAAQ,CAAC,CAAC;IAC9B,CAAC;IAED;;;;OAIG;IACK,MAAM,CAAC,gBAAgB,CAC7B,IAA+B,EAC/B,KAAa,EACb,KAA0B;QAE1B,MAAM,IAAI,GAAQ,EAAE,CAAC;QACrB,KAAK,MAAM,GAAG,IAAI,IAAI,EAAE,CAAC;YACvB,MAAM,IAAI,GAAG,KAAK,CAAC,GAAG,CAAC,CAAC;YACxB,IAAI,IAAI,CAAC,MAAM,IAAI,KAAK,IAAI,IAAI,CAAC,UAAU,IAAI,IAAI,CAAC,IAAI,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,UAAU,EAAE,CAAC;gBAChF,SAAS;YACX,CAAC;YACD,IAAI,QAAQ,GAAG,IAAI,CAAC,MAAM,CAAC;YAC3B,OAAO,QAAQ,GAAG,CAAC,IAAI,IAAI,CAAC,QAAQ,GAAG,CAAC,CAAC,CAAC,UAAU,GAAG,IAAI,CAAC,UAAU,EAAE,CAAC;gBACvE,QAAQ,EAAE,CAAC;YACb,CAAC;YACD,IAAI,CAAC,MAAM,CAAC,QAAQ,EAAE,CAAC,EAAE,IAAI,CAAC,CAAC;YAC/B,IAAI,IAAI,CAAC,MAAM,GAAG,KAAK;gBAAE,IAAI,CAAC,GAAG,EAAE,CAAC;QACtC,CAAC;QACD,OAAO,IAAI,CAAC;IACd,CAAC;IAED;;OAEG;IACI,mBAAmB,CACxB,QAAkB,EAClB,QAAgB,EAAE;QAElB,MAAM,IAAI,GAAG,IAAI,CAAC,EAAE,CAAC,OAAO,CAC1B,2CAA2C,CAC5C,CAAC,OAAO,EAAE,CAAC;QAEZ,MAAM,UAAU,GAAG,IAAI,YAAY,CAAC,QAAQ,CAAC,CAAC;QAE9C,OAAO,WAAW,CAAC,gBAAgB,CAAC,IAAI,EAAE,KAAK,EAAE,CAAC,GAAG,EAAE,EAAE;YACvD,MAAM,GAAG,GAAG,GAAyC,CAAC;YACtD,OAAO;gBACL,GAAG,EAAE,GAAG,CAAC,GAAG;gBACZ,UAAU,EAAE,gBAAgB,CAAC,UAAU,EAAE,iBAAiB,CAAC,GAAG,CAAC,SAAS,CAAC,CAAC;aAC3E,CAAC;QACJ,CAAC,CAAC,CAAC;IACL,CAAC;IAED;;OAEG;IACI,oBAAoB,CACzB,UAAkB,EAClB,QAAkB,EAClB,QAAgB,EAAE;QAElB,MAAM,IAAI,GAAG,IAAI,CAAC,EAAE,CAAC,OAAO,CAC1B,sEAAsE,CACvE,CAAC,OAAO,CAAC,UAAU,CAAC,CAAC;QAEtB,MAAM,UAAU,GAAG,IAAI,YAAY,CAAC,QAAQ,CAAC,CAAC;QAE9C,OAAO,WAAW,CAAC,gBAAgB,CAAC,IAAI,EAAE,KAAK,EAAE,CAAC,GAAG,EAAE,EAAE;YACvD,MAAM,GAAG,GAAG,GAA8C,CAAC;YAC3D,OAAO;gBACL,UAAU;gBACV,QAAQ,EAAE,GAAG,CAAC,QAAQ;gBACtB,UAAU,EAAE,gBAAgB,CAAC,UAAU,EAAE,iBAAiB,CAAC,GAAG,CAAC,SAAS,CAAC,CAAC;aAC3E,CAAC;QACJ,CAAC,CAAC,CAAC;IACL,CAAC;IAED;;OAEG;IACI,oBAAoB;QACzB,MAAM,GAAG,GAAG,IAAI,CAAC,EAAE,CAAC,OAAO,CACzB,4CAA4C,CAC7C,CAAC,GAAG,EAAqB,CAAC;QAC3B,OAAO,GAAG,CAAC,GAAG,CAAC;IACjB,CAAC;IAED;;OAEG;IACI,qBAAqB,CAAC,UAAmB;QAC9C,IAAI,UAAU,EAAE,CAAC;YACf,MAAM,GAAG,GAAG,IAAI,CAAC,EAAE,CAAC,OAAO,CACzB,kEAAkE,CACnE,CAAC,GAAG,CAAC,UAAU,CAAoB,CAAC;YACrC,OAAO,GAAG,CAAC,GAAG,CAAC;QACjB,CAAC;QACD,MAAM,GAAG,GAAG,IAAI,CAAC,EAAE,CAAC,OAAO,CACzB,6CAA6C,CAC9C,CAAC,GAAG,EAAqB,CAAC;QAC3B,OAAO,GAAG,CAAC,GAAG,CAAC;IACjB,CAAC;CACF"}

package/dist/infra/lib/embedding/cosine.d.ts

@@ -1,19 +0,0 @@
-/**
- * 코사인 유사도 계산 + 벡터 직렬화
- *
- * Float32Array 사용: 256차원 * 4바이트 = 1KB/벡터
- */
-/**
- * 두 벡터의 코사인 유사도 계산 (0~1)
- * 동일 차원이어야 함. 빈 벡터는 0 반환.
- */
-export declare function cosineSimilarity(a: Float32Array, b: Float32Array): number;
-/**
- * number[] → Buffer (Float32Array 직렬화)
- */
-export declare function serializeVector(vec: number[]): Buffer;
-/**
- * Buffer → Float32Array (역직렬화)
- */
-export declare function deserializeVector(buf: Buffer): Float32Array;
-//# sourceMappingURL=cosine.d.ts.map

package/dist/infra/lib/embedding/cosine.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"cosine.d.ts","sourceRoot":"","sources":["../../../../src/infra/lib/embedding/cosine.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AAEH;;;GAGG;AACH,wBAAgB,gBAAgB,CAAC,CAAC,EAAE,YAAY,EAAE,CAAC,EAAE,YAAY,GAAG,MAAM,CAiBzE;AAED;;GAEG;AACH,wBAAgB,eAAe,CAAC,GAAG,EAAE,MAAM,EAAE,GAAG,MAAM,CAGrD;AAED;;GAEG;AACH,wBAAgB,iBAAiB,CAAC,GAAG,EAAE,MAAM,GAAG,YAAY,CAM3D"}

package/dist/infra/lib/embedding/cosine.js

@@ -1,40 +0,0 @@
-/**
- * 코사인 유사도 계산 + 벡터 직렬화
- *
- * Float32Array 사용: 256차원 * 4바이트 = 1KB/벡터
- */
-/**
- * 두 벡터의 코사인 유사도 계산 (0~1)
- * 동일 차원이어야 함. 빈 벡터는 0 반환.
- */
-export function cosineSimilarity(a, b) {
-    if (a.length !== b.length || a.length === 0)
-        return 0;
-    let dotProduct = 0;
-    let normA = 0;
-    let normB = 0;
-    for (let i = 0; i < a.length; i++) {
-        dotProduct += a[i] * b[i];
-        normA += a[i] * a[i];
-        normB += b[i] * b[i];
-    }
-    const denominator = Math.sqrt(normA) * Math.sqrt(normB);
-    if (denominator === 0)
-        return 0;
-    return dotProduct / denominator;
-}
-/**
- * number[] → Buffer (Float32Array 직렬화)
- */
-export function serializeVector(vec) {
-    const float32 = new Float32Array(vec);
-    return Buffer.from(float32.buffer);
-}
-/**
- * Buffer → Float32Array (역직렬화)
- */
-export function deserializeVector(buf) {
-    const arrayBuffer = buf.buffer.slice(buf.byteOffset, buf.byteOffset + buf.byteLength);
-    return new Float32Array(arrayBuffer);
-}
-//# sourceMappingURL=cosine.js.map

package/dist/infra/lib/embedding/cosine.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"cosine.js","sourceRoot":"","sources":["../../../../src/infra/lib/embedding/cosine.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AAEH;;;GAGG;AACH,MAAM,UAAU,gBAAgB,CAAC,CAAe,EAAE,CAAe;IAC/D,IAAI,CAAC,CAAC,MAAM,KAAK,CAAC,CAAC,MAAM,IAAI,CAAC,CAAC,MAAM,KAAK,CAAC;QAAE,OAAO,CAAC,CAAC;IAEtD,IAAI,UAAU,GAAG,CAAC,CAAC;IACnB,IAAI,KAAK,GAAG,CAAC,CAAC;IACd,IAAI,KAAK,GAAG,CAAC,CAAC;IAEd,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QAClC,UAAU,IAAI,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC;QAC1B,KAAK,IAAI,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC;QACrB,KAAK,IAAI,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC;IACvB,CAAC;IAED,MAAM,WAAW,GAAG,IAAI,CAAC,IAAI,CAAC,KAAK,CAAC,GAAG,IAAI,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC;IACxD,IAAI,WAAW,KAAK,CAAC;QAAE,OAAO,CAAC,CAAC;IAEhC,OAAO,UAAU,GAAG,WAAW,CAAC;AAClC,CAAC;AAED;;GAEG;AACH,MAAM,UAAU,eAAe,CAAC,GAAa;IAC3C,MAAM,OAAO,GAAG,IAAI,YAAY,CAAC,GAAG,CAAC,CAAC;IACtC,OAAO,MAAM,CAAC,IAAI,CAAC,OAAO,CAAC,MAAM,CAAC,CAAC;AACrC,CAAC;AAED;;GAEG;AACH,MAAM,UAAU,iBAAiB,CAAC,GAAW;IAC3C,MAAM,WAAW,GAAG,GAAG,CAAC,MAAM,CAAC,KAAK,CAClC,GAAG,CAAC,UAAU,EACd,GAAG,CAAC,UAAU,GAAG,GAAG,CAAC,UAAU,CAChC,CAAC;IACF,OAAO,IAAI,YAAY,CAAC,WAAW,CAAC,CAAC;AACvC,CAAC"}

package/dist/infra/lib/embedding/index.d.ts

@@ -1,9 +0,0 @@
-/**
- * Embedding 모듈 — 벡터 검색 인프라
- */
-export { EmbeddingProvider } from './EmbeddingProvider.js';
-export { VectorStore } from './VectorStore.js';
-export { cosineSimilarity, serializeVector, deserializeVector } from './cosine.js';
-export type { EmbeddingResponse, EmbeddingProviderType, EmbeddingProviderConfig, VectorSearchResult, SessionVectorSearchResult, } from './types.js';
-export { DEFAULT_EMBEDDING_DIMENSIONS } from './types.js';
-//# sourceMappingURL=index.d.ts.map

package/dist/infra/lib/embedding/index.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../../../src/infra/lib/embedding/index.ts"],"names":[],"mappings":"AAAA;;GAEG;AAEH,OAAO,EAAE,iBAAiB,EAAE,MAAM,wBAAwB,CAAC;AAC3D,OAAO,EAAE,WAAW,EAAE,MAAM,kBAAkB,CAAC;AAC/C,OAAO,EAAE,gBAAgB,EAAE,eAAe,EAAE,iBAAiB,EAAE,MAAM,aAAa,CAAC;AACnF,YAAY,EACV,iBAAiB,EACjB,qBAAqB,EACrB,uBAAuB,EACvB,kBAAkB,EAClB,yBAAyB,GAC1B,MAAM,YAAY,CAAC;AACpB,OAAO,EAAE,4BAA4B,EAAE,MAAM,YAAY,CAAC"}

package/dist/infra/lib/embedding/index.js

@@ -1,8 +0,0 @@
-/**
- * Embedding 모듈 — 벡터 검색 인프라
- */
-export { EmbeddingProvider } from './EmbeddingProvider.js';
-export { VectorStore } from './VectorStore.js';
-export { cosineSimilarity, serializeVector, deserializeVector } from './cosine.js';
-export { DEFAULT_EMBEDDING_DIMENSIONS } from './types.js';
-//# sourceMappingURL=index.js.map

package/dist/infra/lib/embedding/index.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"index.js","sourceRoot":"","sources":["../../../../src/infra/lib/embedding/index.ts"],"names":[],"mappings":"AAAA;;GAEG;AAEH,OAAO,EAAE,iBAAiB,EAAE,MAAM,wBAAwB,CAAC;AAC3D,OAAO,EAAE,WAAW,EAAE,MAAM,kBAAkB,CAAC;AAC/C,OAAO,EAAE,gBAAgB,EAAE,eAAe,EAAE,iBAAiB,EAAE,MAAM,aAAa,CAAC;AAQnF,OAAO,EAAE,4BAA4B,EAAE,MAAM,YAAY,CAAC"}

package/dist/infra/lib/embedding/types.d.ts

@@ -1,30 +0,0 @@
-/**
- * Embedding 타입 정의
- * 벡터 검색 + 하이브리드 스코어링에 사용
- */
-export interface EmbeddingResponse {
-    embeddings: number[][];
-    model: string;
-}
-export type EmbeddingProviderType = 'gpt';
-export interface EmbeddingProviderConfig {
-    priority: EmbeddingProviderType[];
-    dimensions: number;
-}
-export interface VectorSearchResult {
-    key: string;
-    similarity: number;
-}
-export interface SessionVectorSearchResult {
-    entityType: string;
-    entityId: number;
-    similarity: number;
-}
-/** 256차원 축소 (text-embedding-3-large 기본 3072 → 256) */
-export declare const DEFAULT_EMBEDDING_DIMENSIONS = 256;
-/** 임베딩 API 타임아웃 (ms) */
-export declare const EMBEDDING_API_TIMEOUT_MS = 30000;
-/** GPT Embedding 엔드포인트 */
-export declare const GPT_EMBEDDING_BASE_URL = "https://api.openai.com/v1";
-export declare const GPT_EMBEDDING_MODEL = "text-embedding-3-large";
-//# sourceMappingURL=types.d.ts.map

package/dist/infra/lib/embedding/types.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"types.d.ts","sourceRoot":"","sources":["../../../../src/infra/lib/embedding/types.ts"],"names":[],"mappings":"AAAA;;;GAGG;AAEH,MAAM,WAAW,iBAAiB;IAChC,UAAU,EAAE,MAAM,EAAE,EAAE,CAAC;IACvB,KAAK,EAAE,MAAM,CAAC;CACf;AAED,MAAM,MAAM,qBAAqB,GAAG,KAAK,CAAC;AAE1C,MAAM,WAAW,uBAAuB;IACtC,QAAQ,EAAE,qBAAqB,EAAE,CAAC;IAClC,UAAU,EAAE,MAAM,CAAC;CACpB;AAED,MAAM,WAAW,kBAAkB;IACjC,GAAG,EAAE,MAAM,CAAC;IACZ,UAAU,EAAE,MAAM,CAAC;CACpB;AAED,MAAM,WAAW,yBAAyB;IACxC,UAAU,EAAE,MAAM,CAAC;IACnB,QAAQ,EAAE,MAAM,CAAC;IACjB,UAAU,EAAE,MAAM,CAAC;CACpB;AAED,sDAAsD;AACtD,eAAO,MAAM,4BAA4B,MAAM,CAAC;AAEhD,wBAAwB;AACxB,eAAO,MAAM,wBAAwB,QAAS,CAAC;AAE/C,0BAA0B;AAC1B,eAAO,MAAM,sBAAsB,8BAA8B,CAAC;AAClE,eAAO,MAAM,mBAAmB,2BAA2B,CAAC"}

package/dist/infra/lib/embedding/types.js

@@ -1,12 +0,0 @@
-/**
- * Embedding 타입 정의
- * 벡터 검색 + 하이브리드 스코어링에 사용
- */
-/** 256차원 축소 (text-embedding-3-large 기본 3072 → 256) */
-export const DEFAULT_EMBEDDING_DIMENSIONS = 256;
-/** 임베딩 API 타임아웃 (ms) */
-export const EMBEDDING_API_TIMEOUT_MS = 30_000;
-/** GPT Embedding 엔드포인트 */
-export const GPT_EMBEDDING_BASE_URL = 'https://api.openai.com/v1';
-export const GPT_EMBEDDING_MODEL = 'text-embedding-3-large';
-//# sourceMappingURL=types.js.map

package/dist/infra/lib/embedding/types.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"types.js","sourceRoot":"","sources":["../../../../src/infra/lib/embedding/types.ts"],"names":[],"mappings":"AAAA;;;GAGG;AAyBH,sDAAsD;AACtD,MAAM,CAAC,MAAM,4BAA4B,GAAG,GAAG,CAAC;AAEhD,wBAAwB;AACxB,MAAM,CAAC,MAAM,wBAAwB,GAAG,MAAM,CAAC;AAE/C,0BAA0B;AAC1B,MAAM,CAAC,MAAM,sBAAsB,GAAG,2BAA2B,CAAC;AAClE,MAAM,CAAC,MAAM,mBAAmB,GAAG,wBAAwB,CAAC"}

package/dist/infra/lib/evolution/AgentAnalyzer.d.ts

@@ -1,15 +0,0 @@
-import { InsightStore } from './InsightStore.js';
-export declare class AgentAnalyzer {
-    private insightStore;
-    private agentDb;
-    constructor(insightStore: InsightStore, agentRegistryDbPath?: string);
-    /**
-     * Analyze agent execution stats and generate optimization insights
-     */
-    analyze(): {
-        newInsights: string[];
-        agentsAnalyzed: number;
-    };
-    close(): void;
-}
-//# sourceMappingURL=AgentAnalyzer.d.ts.map

package/dist/infra/lib/evolution/AgentAnalyzer.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"AgentAnalyzer.d.ts","sourceRoot":"","sources":["../../../../src/infra/lib/evolution/AgentAnalyzer.ts"],"names":[],"mappings":"AAIA,OAAO,EAAE,YAAY,EAAE,MAAM,mBAAmB,CAAC;AAUjD,qBAAa,aAAa;IACxB,OAAO,CAAC,YAAY,CAAe;IACnC,OAAO,CAAC,OAAO,CAAkC;gBAErC,YAAY,EAAE,YAAY,EAAE,mBAAmB,CAAC,EAAE,MAAM;IAWpE;;OAEG;IACI,OAAO,IAAI;QAAE,WAAW,EAAE,MAAM,EAAE,CAAC;QAAC,cAAc,EAAE,MAAM,CAAA;KAAE;IAoE5D,KAAK,IAAI,IAAI;CAMrB"}