claude-code-workflow 6.1.3 → 6.2.1

package/src/core/core-memory-store.ts

@@ -0,0 +1,1528 @@
+/**
+ * Core Memory Store - Independent storage system for core memories
+ * Provides persistent storage for high-level architectural and strategic context
+ */
+
+import Database from 'better-sqlite3';
+import { existsSync, mkdirSync } from 'fs';
+import { join } from 'path';
+import { StoragePaths, ensureStorageDir } from '../config/storage-paths.js';
+
+// Types
+export interface CoreMemory {
+  id: string; // Format: CMEM-YYYYMMDD-HHMMSS
+  content: string;
+  summary: string;
+  raw_output?: string;
+  created_at: string;
+  updated_at: string;
+  archived: boolean;
+  metadata?: string; // JSON string
+}
+
+export interface SessionCluster {
+  id: string;  // Format: CLST-YYYYMMDD-HHMMSS
+  name: string;
+  description?: string;
+  intent?: string;
+  created_at: string;
+  updated_at: string;
+  status: 'active' | 'archived' | 'merged';
+  metadata?: string;
+}
+
+export interface ClusterMember {
+  cluster_id: string;
+  session_id: string;
+  session_type: 'core_memory' | 'workflow' | 'cli_history' | 'native';
+  sequence_order: number;
+  added_at: string;
+  relevance_score: number;
+}
+
+export interface ClusterRelation {
+  source_cluster_id: string;
+  target_cluster_id: string;
+  relation_type: 'depends_on' | 'extends' | 'conflicts_with' | 'related_to';
+  created_at: string;
+}
+
+export interface SessionMetadataCache {
+  session_id: string;
+  session_type: string;
+  title?: string;
+  summary?: string;
+  keywords?: string[];  // stored as JSON
+  token_estimate?: number;
+  file_patterns?: string[];  // stored as JSON
+  created_at?: string;
+  last_accessed?: string;
+  access_count: number;
+}
+
+export interface MemoryChunk {
+  id?: number;
+  source_id: string;
+  source_type: 'core_memory' | 'workflow' | 'cli_history';
+  chunk_index: number;
+  content: string;
+  embedding?: Buffer;
+  metadata?: string;
+  created_at: string;
+}
+
+export interface ClaudeUpdateRecord {
+  id?: number;
+  file_path: string;
+  file_level: 'user' | 'project' | 'module';
+  module_path?: string;
+  updated_at: string;
+  update_source: 'manual' | 'cli_sync' | 'dashboard' | 'api';
+  git_commit_hash?: string;
+  files_changed_before_update: number;
+  metadata?: string;
+}
+
+/**
+ * Core Memory Store using SQLite
+ */
+export class CoreMemoryStore {
+  private db: Database.Database;
+  private dbPath: string;
+  private projectPath: string;
+
+  constructor(projectPath: string) {
+    this.projectPath = projectPath;
+    // Use centralized storage path
+    const paths = StoragePaths.project(projectPath);
+    const coreMemoryDir = join(paths.root, 'core-memory');
+    ensureStorageDir(coreMemoryDir);
+
+    this.dbPath = join(coreMemoryDir, 'core_memory.db');
+    this.db = new Database(this.dbPath);
+    this.db.pragma('journal_mode = WAL');
+    this.db.pragma('synchronous = NORMAL');
+
+    this.initDatabase();
+  }
+
+  /**
+   * Initialize database schema
+   */
+  private initDatabase(): void {
+    // Migrate old tables
+    this.migrateDatabase();
+
+    this.db.exec(`
+      -- Core memories table
+      CREATE TABLE IF NOT EXISTS memories (
+        id TEXT PRIMARY KEY,
+        content TEXT NOT NULL,
+        summary TEXT,
+        raw_output TEXT,
+        created_at TEXT NOT NULL,
+        updated_at TEXT NOT NULL,
+        archived INTEGER DEFAULT 0,
+        metadata TEXT
+      );
+
+      -- Session clusters table
+      CREATE TABLE IF NOT EXISTS session_clusters (
+        id TEXT PRIMARY KEY,
+        name TEXT NOT NULL,
+        description TEXT,
+        intent TEXT,
+        created_at TEXT NOT NULL,
+        updated_at TEXT NOT NULL,
+        status TEXT DEFAULT 'active',
+        metadata TEXT
+      );
+
+      -- Cluster members table
+      CREATE TABLE IF NOT EXISTS cluster_members (
+        cluster_id TEXT NOT NULL,
+        session_id TEXT NOT NULL,
+        session_type TEXT NOT NULL,
+        sequence_order INTEGER NOT NULL,
+        added_at TEXT NOT NULL,
+        relevance_score REAL DEFAULT 1.0,
+        PRIMARY KEY (cluster_id, session_id),
+        FOREIGN KEY (cluster_id) REFERENCES session_clusters(id) ON DELETE CASCADE
+      );
+
+      -- Cluster relations table
+      CREATE TABLE IF NOT EXISTS cluster_relations (
+        source_cluster_id TEXT NOT NULL,
+        target_cluster_id TEXT NOT NULL,
+        relation_type TEXT NOT NULL,
+        created_at TEXT NOT NULL,
+        PRIMARY KEY (source_cluster_id, target_cluster_id),
+        FOREIGN KEY (source_cluster_id) REFERENCES session_clusters(id) ON DELETE CASCADE,
+        FOREIGN KEY (target_cluster_id) REFERENCES session_clusters(id) ON DELETE CASCADE
+      );
+
+      -- Session metadata cache table
+      CREATE TABLE IF NOT EXISTS session_metadata_cache (
+        session_id TEXT PRIMARY KEY,
+        session_type TEXT NOT NULL,
+        title TEXT,
+        summary TEXT,
+        keywords TEXT,
+        token_estimate INTEGER,
+        file_patterns TEXT,
+        created_at TEXT,
+        last_accessed TEXT,
+        access_count INTEGER DEFAULT 0
+      );
+
+      -- Memory chunks table for embeddings
+      CREATE TABLE IF NOT EXISTS memory_chunks (
+        id INTEGER PRIMARY KEY AUTOINCREMENT,
+        source_id TEXT NOT NULL,
+        source_type TEXT NOT NULL,
+        chunk_index INTEGER NOT NULL,
+        content TEXT NOT NULL,
+        embedding BLOB,
+        metadata TEXT,
+        created_at TEXT NOT NULL,
+        UNIQUE(source_id, chunk_index)
+      );
+
+      -- CLAUDE.md update history table
+      CREATE TABLE IF NOT EXISTS claude_update_history (
+        id INTEGER PRIMARY KEY AUTOINCREMENT,
+        file_path TEXT NOT NULL,
+        file_level TEXT NOT NULL CHECK(file_level IN ('user', 'project', 'module')),
+        module_path TEXT,
+        updated_at TEXT NOT NULL,
+        update_source TEXT NOT NULL CHECK(update_source IN ('manual', 'cli_sync', 'dashboard', 'api')),
+        git_commit_hash TEXT,
+        files_changed_before_update INTEGER DEFAULT 0,
+        metadata TEXT,
+        UNIQUE(file_path, updated_at)
+      );
+
+      -- Indexes for efficient queries
+      CREATE INDEX IF NOT EXISTS idx_memories_created ON memories(created_at DESC);
+      CREATE INDEX IF NOT EXISTS idx_memories_updated ON memories(updated_at DESC);
+      CREATE INDEX IF NOT EXISTS idx_memories_archived ON memories(archived);
+      CREATE INDEX IF NOT EXISTS idx_session_clusters_status ON session_clusters(status);
+      CREATE INDEX IF NOT EXISTS idx_cluster_members_cluster ON cluster_members(cluster_id);
+      CREATE INDEX IF NOT EXISTS idx_cluster_members_session ON cluster_members(session_id);
+      CREATE INDEX IF NOT EXISTS idx_session_metadata_type ON session_metadata_cache(session_type);
+      CREATE INDEX IF NOT EXISTS idx_memory_chunks_source ON memory_chunks(source_id, source_type);
+      CREATE INDEX IF NOT EXISTS idx_memory_chunks_embedded ON memory_chunks(embedding IS NOT NULL);
+      CREATE INDEX IF NOT EXISTS idx_claude_history_path ON claude_update_history(file_path);
+      CREATE INDEX IF NOT EXISTS idx_claude_history_updated ON claude_update_history(updated_at DESC);
+      CREATE INDEX IF NOT EXISTS idx_claude_history_module ON claude_update_history(module_path);
+    `);
+  }
+
+  /**
+   * Migrate database by removing old tables, views, and triggers
+   */
+  private migrateDatabase(): void {
+    const oldTables = ['knowledge_graph', 'knowledge_graph_edges', 'evolution_history'];
+
+    try {
+      // Disable foreign key constraints during migration
+      this.db.pragma('foreign_keys = OFF');
+
+      // Drop any triggers that might reference old tables
+      const triggers = this.db.prepare(
+        `SELECT name FROM sqlite_master WHERE type='trigger'`
+      ).all() as { name: string }[];
+
+      for (const trigger of triggers) {
+        try {
+          this.db.exec(`DROP TRIGGER IF EXISTS "${trigger.name}"`);
+        } catch (e) {
+          // Ignore trigger drop errors
+        }
+      }
+
+      // Drop any views that might reference old tables
+      const views = this.db.prepare(
+        `SELECT name FROM sqlite_master WHERE type='view'`
+      ).all() as { name: string }[];
+
+      for (const view of views) {
+        try {
+          this.db.exec(`DROP VIEW IF EXISTS "${view.name}"`);
+        } catch (e) {
+          // Ignore view drop errors
+        }
+      }
+
+      // Now drop the old tables
+      for (const table of oldTables) {
+        try {
+          this.db.exec(`DROP TABLE IF EXISTS "${table}"`);
+        } catch (e) {
+          // Ignore if table doesn't exist
+        }
+      }
+
+      // Re-enable foreign key constraints
+      this.db.pragma('foreign_keys = ON');
+    } catch (e) {
+      // If migration fails, continue - tables may not exist
+      try {
+        this.db.pragma('foreign_keys = ON');
+      } catch (_) {
+        // Ignore
+      }
+    }
+  }
+
+  /**
+   * Generate timestamp-based ID for core memory
+   */
+  private generateId(): string {
+    const now = new Date();
+    const year = now.getFullYear();
+    const month = String(now.getMonth() + 1).padStart(2, '0');
+    const day = String(now.getDate()).padStart(2, '0');
+    const hours = String(now.getHours()).padStart(2, '0');
+    const minutes = String(now.getMinutes()).padStart(2, '0');
+    const seconds = String(now.getSeconds()).padStart(2, '0');
+    return `CMEM-${year}${month}${day}-${hours}${minutes}${seconds}`;
+  }
+
+  /**
+   * Generate cluster ID
+   */
+  generateClusterId(): string {
+    const now = new Date();
+    const year = now.getFullYear();
+    const month = String(now.getMonth() + 1).padStart(2, '0');
+    const day = String(now.getDate()).padStart(2, '0');
+    const hours = String(now.getHours()).padStart(2, '0');
+    const minutes = String(now.getMinutes()).padStart(2, '0');
+    const seconds = String(now.getSeconds()).padStart(2, '0');
+    const ms = String(now.getMilliseconds()).padStart(3, '0');
+    // Add random 4-digit suffix to ensure uniqueness (10000 combinations)
+    const random = String(Math.floor(Math.random() * 10000)).padStart(4, '0');
+    return `CLST-${year}${month}${day}-${hours}${minutes}${seconds}${ms}${random}`;
+  }
+
+  /**
+   * Upsert a core memory
+   */
+  upsertMemory(memory: Partial<CoreMemory> & { content: string }): CoreMemory {
+    const now = new Date().toISOString();
+    const id = memory.id || this.generateId();
+
+    // Check if memory exists
+    const existingMemory = this.getMemory(id);
+
+    if (existingMemory) {
+      // Update existing memory
+      const stmt = this.db.prepare(`
+        UPDATE memories
+        SET content = ?, summary = ?, raw_output = ?, updated_at = ?, archived = ?, metadata = ?
+        WHERE id = ?
+      `);
+
+      stmt.run(
+        memory.content,
+        memory.summary || existingMemory.summary,
+        memory.raw_output || existingMemory.raw_output,
+        now,
+        memory.archived !== undefined ? (memory.archived ? 1 : 0) : existingMemory.archived ? 1 : 0,
+        memory.metadata || existingMemory.metadata,
+        id
+      );
+
+      return this.getMemory(id)!;
+    } else {
+      // Insert new memory
+      const stmt = this.db.prepare(`
+        INSERT INTO memories (id, content, summary, raw_output, created_at, updated_at, archived, metadata)
+        VALUES (?, ?, ?, ?, ?, ?, ?, ?)
+      `);
+
+      stmt.run(
+        id,
+        memory.content,
+        memory.summary || '',
+        memory.raw_output || null,
+        now,
+        now,
+        memory.archived ? 1 : 0,
+        memory.metadata || null
+      );
+
+      return this.getMemory(id)!;
+    }
+  }
+
+  /**
+   * Get memory by ID
+   */
+  getMemory(id: string): CoreMemory | null {
+    const stmt = this.db.prepare(`SELECT * FROM memories WHERE id = ?`);
+    const row = stmt.get(id) as any;
+    if (!row) return null;
+
+    return {
+      id: row.id,
+      content: row.content,
+      summary: row.summary,
+      raw_output: row.raw_output,
+      created_at: row.created_at,
+      updated_at: row.updated_at,
+      archived: Boolean(row.archived),
+      metadata: row.metadata
+    };
+  }
+
+  /**
+   * Get all memories
+   */
+  getMemories(options: { archived?: boolean; limit?: number; offset?: number } = {}): CoreMemory[] {
+    const { archived = false, limit = 50, offset = 0 } = options;
+
+    const stmt = this.db.prepare(`
+      SELECT * FROM memories
+      WHERE archived = ?
+      ORDER BY updated_at DESC
+      LIMIT ? OFFSET ?
+    `);
+
+    const rows = stmt.all(archived ? 1 : 0, limit, offset) as any[];
+    return rows.map(row => ({
+      id: row.id,
+      content: row.content,
+      summary: row.summary,
+      raw_output: row.raw_output,
+      created_at: row.created_at,
+      updated_at: row.updated_at,
+      archived: Boolean(row.archived),
+      metadata: row.metadata
+    }));
+  }
+
+  /**
+   * Archive a memory
+   */
+  archiveMemory(id: string): void {
+    const stmt = this.db.prepare(`
+      UPDATE memories
+      SET archived = 1, updated_at = ?
+      WHERE id = ?
+    `);
+    stmt.run(new Date().toISOString(), id);
+  }
+
+  /**
+   * Delete a memory
+   */
+  deleteMemory(id: string): void {
+    const stmt = this.db.prepare(`DELETE FROM memories WHERE id = ?`);
+    stmt.run(id);
+  }
+
+  /**
+   * Generate summary for a memory using CLI tool
+   */
+  async generateSummary(memoryId: string, tool: 'gemini' | 'qwen' = 'gemini'): Promise<string> {
+    const memory = this.getMemory(memoryId);
+    if (!memory) throw new Error('Memory not found');
+
+    // Import CLI executor
+    const { executeCliTool } = await import('../tools/cli-executor.js');
+
+    const prompt = `
+PURPOSE: Generate a concise summary (2-3 sentences) of the following core memory content
+TASK: Extract key architectural decisions, strategic insights, and important context
+MODE: analysis
+EXPECTED: Plain text summary without markdown or formatting
+RULES: Be concise. Focus on high-level understanding. No technical jargon unless essential.
+
+CONTENT:
+${memory.content}
+`;
+
+    const result = await executeCliTool({
+      tool,
+      prompt,
+      mode: 'analysis',
+      timeout: 60000,
+      cd: this.projectPath,
+      category: 'internal'
+    });
+
+    const summary = result.stdout?.trim() || 'Failed to generate summary';
+
+    // Update memory with summary
+    const stmt = this.db.prepare(`
+      UPDATE memories
+      SET summary = ?, updated_at = ?
+      WHERE id = ?
+    `);
+    stmt.run(summary, new Date().toISOString(), memoryId);
+
+    return summary;
+  }
+
+  /**
+   * Create a new session cluster
+   */
+  createCluster(cluster: Partial<SessionCluster> & { name: string }): SessionCluster {
+    const now = new Date().toISOString();
+    const id = cluster.id || this.generateClusterId();
+
+    const stmt = this.db.prepare(`
+      INSERT INTO session_clusters (id, name, description, intent, created_at, updated_at, status, metadata)
+      VALUES (?, ?, ?, ?, ?, ?, ?, ?)
+    `);
+
+    stmt.run(
+      id,
+      cluster.name,
+      cluster.description || null,
+      cluster.intent || null,
+      now,
+      now,
+      cluster.status || 'active',
+      cluster.metadata || null
+    );
+
+    return this.getCluster(id)!;
+  }
+
+  /**
+   * Get cluster by ID
+   */
+  getCluster(id: string): SessionCluster | null {
+    const stmt = this.db.prepare(`SELECT * FROM session_clusters WHERE id = ?`);
+    const row = stmt.get(id) as any;
+    if (!row) return null;
+
+    return {
+      id: row.id,
+      name: row.name,
+      description: row.description,
+      intent: row.intent,
+      created_at: row.created_at,
+      updated_at: row.updated_at,
+      status: row.status,
+      metadata: row.metadata
+    };
+  }
+
+  /**
+   * List all clusters
+   */
+  listClusters(status?: string): SessionCluster[] {
+    let query = 'SELECT * FROM session_clusters';
+    const params: any[] = [];
+
+    if (status) {
+      query += ' WHERE status = ?';
+      params.push(status);
+    }
+
+    query += ' ORDER BY updated_at DESC';
+
+    const stmt = this.db.prepare(query);
+    const rows = stmt.all(...params) as any[];
+
+    return rows.map(row => ({
+      id: row.id,
+      name: row.name,
+      description: row.description,
+      intent: row.intent,
+      created_at: row.created_at,
+      updated_at: row.updated_at,
+      status: row.status,
+      metadata: row.metadata
+    }));
+  }
+
+  /**
+   * Update cluster
+   */
+  updateCluster(id: string, updates: Partial<SessionCluster>): SessionCluster | null {
+    const existing = this.getCluster(id);
+    if (!existing) return null;
+
+    const now = new Date().toISOString();
+    const stmt = this.db.prepare(`
+      UPDATE session_clusters
+      SET name = ?, description = ?, intent = ?, updated_at = ?, status = ?, metadata = ?
+      WHERE id = ?
+    `);
+
+    stmt.run(
+      updates.name || existing.name,
+      updates.description !== undefined ? updates.description : existing.description,
+      updates.intent !== undefined ? updates.intent : existing.intent,
+      now,
+      updates.status || existing.status,
+      updates.metadata !== undefined ? updates.metadata : existing.metadata,
+      id
+    );
+
+    return this.getCluster(id);
+  }
+
+  /**
+   * Delete cluster
+   */
+  deleteCluster(id: string): boolean {
+    const stmt = this.db.prepare(`DELETE FROM session_clusters WHERE id = ?`);
+    const result = stmt.run(id);
+    return result.changes > 0;
+  }
+
+  /**
+   * Merge multiple clusters into one
+   * Keeps the first cluster and moves all members from others into it
+   * @param targetClusterId The cluster to keep
+   * @param sourceClusterIds The clusters to merge into target (will be deleted)
+   * @returns Number of members moved
+   */
+  mergeClusters(targetClusterId: string, sourceClusterIds: string[]): number {
+    const targetCluster = this.getCluster(targetClusterId);
+    if (!targetCluster) {
+      throw new Error(`Target cluster not found: ${targetClusterId}`);
+    }
+
+    let membersMoved = 0;
+    const existingMembers = new Set(
+      this.getClusterMembers(targetClusterId).map(m => m.session_id)
+    );
+
+    for (const sourceId of sourceClusterIds) {
+      if (sourceId === targetClusterId) continue;
+
+      const sourceMembers = this.getClusterMembers(sourceId);
+      const maxOrder = this.getClusterMembers(targetClusterId).length;
+
+      for (const member of sourceMembers) {
+        // Skip if already exists in target
+        if (existingMembers.has(member.session_id)) continue;
+
+        // Move member to target cluster
+        this.addClusterMember({
+          cluster_id: targetClusterId,
+          session_id: member.session_id,
+          session_type: member.session_type,
+          sequence_order: maxOrder + membersMoved + 1,
+          relevance_score: member.relevance_score
+        });
+
+        existingMembers.add(member.session_id);
+        membersMoved++;
+      }
+
+      // Delete source cluster
+      this.deleteCluster(sourceId);
+    }
+
+    // Update target cluster description
+    const finalMembers = this.getClusterMembers(targetClusterId);
+    this.updateCluster(targetClusterId, {
+      description: `Merged cluster with ${finalMembers.length} sessions`
+    });
+
+    return membersMoved;
+  }
+
+  /**
+   * Add member to cluster
+   */
+  addClusterMember(member: Omit<ClusterMember, 'added_at'>): ClusterMember {
+    const now = new Date().toISOString();
+
+    const stmt = this.db.prepare(`
+      INSERT INTO cluster_members (cluster_id, session_id, session_type, sequence_order, added_at, relevance_score)
+      VALUES (?, ?, ?, ?, ?, ?)
+    `);
+
+    stmt.run(
+      member.cluster_id,
+      member.session_id,
+      member.session_type,
+      member.sequence_order,
+      now,
+      member.relevance_score
+    );
+
+    return {
+      ...member,
+      added_at: now
+    };
+  }
+
+  /**
+   * Remove member from cluster
+   */
+  removeClusterMember(clusterId: string, sessionId: string): boolean {
+    const stmt = this.db.prepare(`
+      DELETE FROM cluster_members
+      WHERE cluster_id = ? AND session_id = ?
+    `);
+    const result = stmt.run(clusterId, sessionId);
+    return result.changes > 0;
+  }
+
+  /**
+   * Get all members of a cluster
+   */
+  getClusterMembers(clusterId: string): ClusterMember[] {
+    const stmt = this.db.prepare(`
+      SELECT * FROM cluster_members
+      WHERE cluster_id = ?
+      ORDER BY sequence_order ASC
+    `);
+
+    const rows = stmt.all(clusterId) as any[];
+    return rows.map(row => ({
+      cluster_id: row.cluster_id,
+      session_id: row.session_id,
+      session_type: row.session_type,
+      sequence_order: row.sequence_order,
+      added_at: row.added_at,
+      relevance_score: row.relevance_score
+    }));
+  }
+
+  /**
+   * Get all clusters that contain a session
+   */
+  getSessionClusters(sessionId: string): SessionCluster[] {
+    const stmt = this.db.prepare(`
+      SELECT sc.*
+      FROM session_clusters sc
+      INNER JOIN cluster_members cm ON sc.id = cm.cluster_id
+      WHERE cm.session_id = ?
+      ORDER BY sc.updated_at DESC
+    `);
+
+    const rows = stmt.all(sessionId) as any[];
+    return rows.map(row => ({
+      id: row.id,
+      name: row.name,
+      description: row.description,
+      intent: row.intent,
+      created_at: row.created_at,
+      updated_at: row.updated_at,
+      status: row.status,
+      metadata: row.metadata
+    }));
+  }
+
+  /**
+   * Add relation between clusters
+   */
+  addClusterRelation(relation: Omit<ClusterRelation, 'created_at'>): ClusterRelation {
+    const now = new Date().toISOString();
+
+    const stmt = this.db.prepare(`
+      INSERT INTO cluster_relations (source_cluster_id, target_cluster_id, relation_type, created_at)
+      VALUES (?, ?, ?, ?)
+    `);
+
+    stmt.run(
+      relation.source_cluster_id,
+      relation.target_cluster_id,
+      relation.relation_type,
+      now
+    );
+
+    return {
+      ...relation,
+      created_at: now
+    };
+  }
+
+  /**
+   * Remove relation between clusters
+   */
+  removeClusterRelation(sourceId: string, targetId: string): boolean {
+    const stmt = this.db.prepare(`
+      DELETE FROM cluster_relations
+      WHERE source_cluster_id = ? AND target_cluster_id = ?
+    `);
+    const result = stmt.run(sourceId, targetId);
+    return result.changes > 0;
+  }
+
+  /**
+   * Get all relations for a cluster
+   */
+  getClusterRelations(clusterId: string): ClusterRelation[] {
+    const stmt = this.db.prepare(`
+      SELECT * FROM cluster_relations
+      WHERE source_cluster_id = ? OR target_cluster_id = ?
+      ORDER BY created_at DESC
+    `);
+
+    const rows = stmt.all(clusterId, clusterId) as any[];
+    return rows.map(row => ({
+      source_cluster_id: row.source_cluster_id,
+      target_cluster_id: row.target_cluster_id,
+      relation_type: row.relation_type,
+      created_at: row.created_at
+    }));
+  }
+
+  /**
+   * Upsert session metadata
+   */
+  upsertSessionMetadata(metadata: SessionMetadataCache): SessionMetadataCache {
+    const now = new Date().toISOString();
+
+    const existing = this.getSessionMetadata(metadata.session_id);
+
+    if (existing) {
+      const stmt = this.db.prepare(`
+        UPDATE session_metadata_cache
+        SET session_type = ?, title = ?, summary = ?, keywords = ?, token_estimate = ?,
+            file_patterns = ?, last_accessed = ?, access_count = ?
+        WHERE session_id = ?
+      `);
+
+      stmt.run(
+        metadata.session_type,
+        metadata.title || null,
+        metadata.summary || null,
+        metadata.keywords ? JSON.stringify(metadata.keywords) : null,
+        metadata.token_estimate || null,
+        metadata.file_patterns ? JSON.stringify(metadata.file_patterns) : null,
+        now,
+        existing.access_count + 1,
+        metadata.session_id
+      );
+    } else {
+      const stmt = this.db.prepare(`
+        INSERT INTO session_metadata_cache
+        (session_id, session_type, title, summary, keywords, token_estimate, file_patterns, created_at, last_accessed, access_count)
+        VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?)
+      `);
+
+      stmt.run(
+        metadata.session_id,
+        metadata.session_type,
+        metadata.title || null,
+        metadata.summary || null,
+        metadata.keywords ? JSON.stringify(metadata.keywords) : null,
+        metadata.token_estimate || null,
+        metadata.file_patterns ? JSON.stringify(metadata.file_patterns) : null,
+        metadata.created_at || now,
+        now,
+        metadata.access_count || 1
+      );
+    }
+
+    return this.getSessionMetadata(metadata.session_id)!;
+  }
+
+  /**
+   * Get session metadata
+   */
+  getSessionMetadata(sessionId: string): SessionMetadataCache | null {
+    const stmt = this.db.prepare(`SELECT * FROM session_metadata_cache WHERE session_id = ?`);
+    const row = stmt.get(sessionId) as any;
+    if (!row) return null;
+
+    return {
+      session_id: row.session_id,
+      session_type: row.session_type,
+      title: row.title,
+      summary: row.summary,
+      keywords: row.keywords ? JSON.parse(row.keywords) : undefined,
+      token_estimate: row.token_estimate,
+      file_patterns: row.file_patterns ? JSON.parse(row.file_patterns) : undefined,
+      created_at: row.created_at,
+      last_accessed: row.last_accessed,
+      access_count: row.access_count
+    };
+  }
+
+  /**
+   * Search sessions by keyword
+   */
+  searchSessionsByKeyword(keyword: string): SessionMetadataCache[] {
+    const stmt = this.db.prepare(`
+      SELECT * FROM session_metadata_cache
+      WHERE title LIKE ? OR summary LIKE ? OR keywords LIKE ?
+      ORDER BY access_count DESC, last_accessed DESC
+    `);
+
+    const pattern = `%${keyword}%`;
+    const rows = stmt.all(pattern, pattern, pattern) as any[];
+
+    return rows.map(row => ({
+      session_id: row.session_id,
+      session_type: row.session_type,
+      title: row.title,
+      summary: row.summary,
+      keywords: row.keywords ? JSON.parse(row.keywords) : undefined,
+      token_estimate: row.token_estimate,
+      file_patterns: row.file_patterns ? JSON.parse(row.file_patterns) : undefined,
+      created_at: row.created_at,
+      last_accessed: row.last_accessed,
+      access_count: row.access_count
+    }));
+  }
+
+  // ============================================================================
+  // Memory Chunks CRUD Operations
+  // ============================================================================
+
+  /**
+   * Chunk content into smaller pieces for embedding
+   * @param content Content to chunk
+   * @param sourceId Source identifier (e.g., memory ID)
+   * @param sourceType Type of source
+   * @returns Array of chunk content strings
+   */
+  chunkContent(content: string, sourceId: string, sourceType: string): string[] {
+    const CHUNK_SIZE = 1500;
+    const OVERLAP = 200;
+    const chunks: string[] = [];
+
+    // Split by paragraph boundaries first
+    const paragraphs = content.split(/\n\n+/);
+    let currentChunk = '';
+
+    for (const paragraph of paragraphs) {
+      // If adding this paragraph would exceed chunk size
+      if (currentChunk.length + paragraph.length > CHUNK_SIZE && currentChunk.length > 0) {
+        // Save current chunk
+        chunks.push(currentChunk.trim());
+
+        // Start new chunk with overlap
+        const overlapText = currentChunk.slice(-OVERLAP);
+        currentChunk = overlapText + '\n\n' + paragraph;
+      } else {
+        // Add paragraph to current chunk
+        currentChunk += (currentChunk ? '\n\n' : '') + paragraph;
+      }
+    }
+
+    // Add remaining chunk
+    if (currentChunk.trim()) {
+      chunks.push(currentChunk.trim());
+    }
+
+    // If no paragraphs or chunks are still too large, split by sentences
+    const finalChunks: string[] = [];
+    for (const chunk of chunks) {
+      if (chunk.length <= CHUNK_SIZE) {
+        finalChunks.push(chunk);
+      } else {
+        // Split by sentence boundaries
+        const sentences = chunk.split(/\. +/);
+        let sentenceChunk = '';
+
+        for (const sentence of sentences) {
+          const sentenceWithPeriod = sentence + '. ';
+          if (sentenceChunk.length + sentenceWithPeriod.length > CHUNK_SIZE && sentenceChunk.length > 0) {
+            finalChunks.push(sentenceChunk.trim());
+            const overlapText = sentenceChunk.slice(-OVERLAP);
+            sentenceChunk = overlapText + sentenceWithPeriod;
+          } else {
+            sentenceChunk += sentenceWithPeriod;
+          }
+        }
+
+        if (sentenceChunk.trim()) {
+          finalChunks.push(sentenceChunk.trim());
+        }
+      }
+    }
+
+    return finalChunks.length > 0 ? finalChunks : [content];
+  }
+
+  /**
+   * Insert a single chunk
+   */
+  insertChunk(chunk: Omit<MemoryChunk, 'id'>): number {
+    const now = new Date().toISOString();
+
+    const stmt = this.db.prepare(`
+      INSERT INTO memory_chunks (source_id, source_type, chunk_index, content, embedding, metadata, created_at)
+      VALUES (?, ?, ?, ?, ?, ?, ?)
+    `);
+
+    const result = stmt.run(
+      chunk.source_id,
+      chunk.source_type,
+      chunk.chunk_index,
+      chunk.content,
+      chunk.embedding || null,
+      chunk.metadata || null,
+      chunk.created_at || now
+    );
+
+    return result.lastInsertRowid as number;
+  }
+
+  /**
+   * Insert multiple chunks in a batch
+   */
+  insertChunksBatch(chunks: Omit<MemoryChunk, 'id'>[]): void {
+    const now = new Date().toISOString();
+    const insert = this.db.prepare(`
+      INSERT INTO memory_chunks (source_id, source_type, chunk_index, content, embedding, metadata, created_at)
+      VALUES (?, ?, ?, ?, ?, ?, ?)
+    `);
+
+    const transaction = this.db.transaction((chunks: Omit<MemoryChunk, 'id'>[]) => {
+      for (const chunk of chunks) {
+        insert.run(
+          chunk.source_id,
+          chunk.source_type,
+          chunk.chunk_index,
+          chunk.content,
+          chunk.embedding || null,
+          chunk.metadata || null,
+          chunk.created_at || now
+        );
+      }
+    });
+
+    transaction(chunks);
+  }
+
+  /**
+   * Get all chunks for a source
+   */
+  getChunks(sourceId: string): MemoryChunk[] {
+    const stmt = this.db.prepare(`
+      SELECT * FROM memory_chunks
+      WHERE source_id = ?
+      ORDER BY chunk_index ASC
+    `);
+
+    const rows = stmt.all(sourceId) as any[];
+    return rows.map(row => ({
+      id: row.id,
+      source_id: row.source_id,
+      source_type: row.source_type,
+      chunk_index: row.chunk_index,
+      content: row.content,
+      embedding: row.embedding,
+      metadata: row.metadata,
+      created_at: row.created_at
+    }));
+  }
+
+  /**
+   * Get chunks by source type
+   */
+  getChunksByType(sourceType: string): MemoryChunk[] {
+    const stmt = this.db.prepare(`
+      SELECT * FROM memory_chunks
+      WHERE source_type = ?
+      ORDER BY source_id, chunk_index ASC
+    `);
+
+    const rows = stmt.all(sourceType) as any[];
+    return rows.map(row => ({
+      id: row.id,
+      source_id: row.source_id,
+      source_type: row.source_type,
+      chunk_index: row.chunk_index,
+      content: row.content,
+      embedding: row.embedding,
+      metadata: row.metadata,
+      created_at: row.created_at
+    }));
+  }
+
+  /**
+   * Get chunks without embeddings
+   */
+  getUnembeddedChunks(limit?: number): MemoryChunk[] {
+    const query = `
+      SELECT * FROM memory_chunks
+      WHERE embedding IS NULL
+      ORDER BY created_at ASC
+      ${limit ? 'LIMIT ?' : ''}
+    `;
+
+    const stmt = this.db.prepare(query);
+    const rows = (limit ? stmt.all(limit) : stmt.all()) as any[];
+
+    return rows.map(row => ({
+      id: row.id,
+      source_id: row.source_id,
+      source_type: row.source_type,
+      chunk_index: row.chunk_index,
+      content: row.content,
+      embedding: row.embedding,
+      metadata: row.metadata,
+      created_at: row.created_at
+    }));
+  }
+
+  /**
+   * Update embedding for a chunk
+   */
+  updateChunkEmbedding(chunkId: number, embedding: Buffer): void {
+    const stmt = this.db.prepare(`
+      UPDATE memory_chunks
+      SET embedding = ?
+      WHERE id = ?
+    `);
+
+    stmt.run(embedding, chunkId);
+  }
+
+  /**
+   * Update embeddings for multiple chunks in a batch
+   */
+  updateChunkEmbeddingsBatch(updates: { id: number; embedding: Buffer }[]): void {
+    const update = this.db.prepare(`
+      UPDATE memory_chunks
+      SET embedding = ?
+      WHERE id = ?
+    `);
+
+    const transaction = this.db.transaction((updates: { id: number; embedding: Buffer }[]) => {
+      for (const { id, embedding } of updates) {
+        update.run(embedding, id);
+      }
+    });
+
+    transaction(updates);
+  }
+
+  /**
+   * Delete all chunks for a source
+   */
+  deleteChunks(sourceId: string): void {
+    const stmt = this.db.prepare(`
+      DELETE FROM memory_chunks
+      WHERE source_id = ?
+    `);
+
+    stmt.run(sourceId);
+  }
+
+  // ============================================================================
+  // CLAUDE.md Update History CRUD Operations
+  // ============================================================================
+
+  /**
+   * Insert a CLAUDE.md update record
+   */
+  insertClaudeUpdateRecord(record: Omit<ClaudeUpdateRecord, 'id'>): ClaudeUpdateRecord {
+    const stmt = this.db.prepare(`
+      INSERT INTO claude_update_history
+      (file_path, file_level, module_path, updated_at, update_source, git_commit_hash, files_changed_before_update, metadata)
+      VALUES (?, ?, ?, ?, ?, ?, ?, ?)
+    `);
+
+    const result = stmt.run(
+      record.file_path,
+      record.file_level,
+      record.module_path || null,
+      record.updated_at,
+      record.update_source,
+      record.git_commit_hash || null,
+      record.files_changed_before_update,
+      record.metadata || null
+    );
+
+    return {
+      id: result.lastInsertRowid as number,
+      ...record
+    };
+  }
+
+  /**
+   * Get the last update record for a file
+   */
+  getLastClaudeUpdate(filePath: string): ClaudeUpdateRecord | null {
+    const stmt = this.db.prepare(`
+      SELECT * FROM claude_update_history
+      WHERE file_path = ?
+      ORDER BY updated_at DESC
+      LIMIT 1
+    `);
+
+    const row = stmt.get(filePath) as any;
+    if (!row) return null;
+
+    return {
+      id: row.id,
+      file_path: row.file_path,
+      file_level: row.file_level,
+      module_path: row.module_path,
+      updated_at: row.updated_at,
+      update_source: row.update_source,
+      git_commit_hash: row.git_commit_hash,
+      files_changed_before_update: row.files_changed_before_update,
+      metadata: row.metadata
+    };
+  }
+
+  /**
+   * Get update history for a file
+   */
+  getClaudeUpdateHistory(filePath: string, limit: number = 50): ClaudeUpdateRecord[] {
+    const stmt = this.db.prepare(`
+      SELECT * FROM claude_update_history
+      WHERE file_path = ?
+      ORDER BY updated_at DESC
+      LIMIT ?
+    `);
+
+    const rows = stmt.all(filePath, limit) as any[];
+    return rows.map(row => ({
+      id: row.id,
+      file_path: row.file_path,
+      file_level: row.file_level,
+      module_path: row.module_path,
+      updated_at: row.updated_at,
+      update_source: row.update_source,
+      git_commit_hash: row.git_commit_hash,
+      files_changed_before_update: row.files_changed_before_update,
+      metadata: row.metadata
+    }));
+  }
+
+  /**
+   * Get all CLAUDE.md update records for freshness calculation
+   */
+  getAllClaudeUpdateRecords(): ClaudeUpdateRecord[] {
+    const stmt = this.db.prepare(`
+      SELECT * FROM claude_update_history
+      WHERE id IN (
+        SELECT MAX(id) FROM claude_update_history
+        GROUP BY file_path
+      )
+      ORDER BY updated_at DESC
+    `);
+
+    const rows = stmt.all() as any[];
+    return rows.map(row => ({
+      id: row.id,
+      file_path: row.file_path,
+      file_level: row.file_level,
+      module_path: row.module_path,
+      updated_at: row.updated_at,
+      update_source: row.update_source,
+      git_commit_hash: row.git_commit_hash,
+      files_changed_before_update: row.files_changed_before_update,
+      metadata: row.metadata
+    }));
+  }
+
+  /**
+   * Delete update records for a file
+   */
+  deleteClaudeUpdateRecords(filePath: string): number {
+    const stmt = this.db.prepare(`
+      DELETE FROM claude_update_history
+      WHERE file_path = ?
+    `);
+    const result = stmt.run(filePath);
+    return result.changes;
+  }
+
+  /**
+   * Close database connection
+   */
+  close(): void {
+    this.db.close();
+  }
+}
+
+// Singleton instance cache
+const storeCache = new Map<string, CoreMemoryStore>();
+
+/**
+ * Get or create a store instance for a project
+ */
+export function getCoreMemoryStore(projectPath: string): CoreMemoryStore {
+  const normalizedPath = projectPath.toLowerCase().replace(/\\/g, '/');
+
+  if (!storeCache.has(normalizedPath)) {
+    storeCache.set(normalizedPath, new CoreMemoryStore(projectPath));
+  }
+  return storeCache.get(normalizedPath)!;
+}
+
+// ============================================================================
+// Cross-workspace management functions
+// ============================================================================
+
+import { readdirSync, writeFileSync, readFileSync } from 'fs';
+import { homedir } from 'os';
+
+export interface ProjectInfo {
+  id: string;
+  path: string;
+  memoriesCount: number;
+  clustersCount: number;
+  lastUpdated?: string;
+}
+
+export interface ExportedMemory {
+  version: string;
+  exportedAt: string;
+  sourceProject: string;
+  memories: CoreMemory[];
+}
+
+/**
+ * Get CCW home directory
+ */
+function getCCWHome(): string {
+  return process.env.CCW_DATA_DIR || join(homedir(), '.ccw');
+}
+
+/**
+ * List all projects with their memory counts
+ */
+export function listAllProjects(): ProjectInfo[] {
+  const projectsDir = join(getCCWHome(), 'projects');
+
+  if (!existsSync(projectsDir)) {
+    return [];
+  }
+
+  const projects: ProjectInfo[] = [];
+  const entries = readdirSync(projectsDir, { withFileTypes: true });
+
+  for (const entry of entries) {
+    if (!entry.isDirectory()) continue;
+
+    const projectId = entry.name;
+    const coreMemoryDb = join(projectsDir, projectId, 'core-memory', 'core_memory.db');
+
+    let memoriesCount = 0;
+    let clustersCount = 0;
+    let lastUpdated: string | undefined;
+
+    if (existsSync(coreMemoryDb)) {
+      try {
+        const db = new Database(coreMemoryDb, { readonly: true });
+
+        // Count memories
+        const memResult = db.prepare('SELECT COUNT(*) as count FROM memories').get() as { count: number };
+        memoriesCount = memResult?.count || 0;
+
+        // Count clusters
+        try {
+          const clusterResult = db.prepare('SELECT COUNT(*) as count FROM session_clusters').get() as { count: number };
+          clustersCount = clusterResult?.count || 0;
+        } catch {
+          // Table might not exist
+        }
+
+        // Get last update time
+        const lastMemory = db.prepare('SELECT MAX(updated_at) as last FROM memories').get() as { last: string };
+        lastUpdated = lastMemory?.last;
+
+        db.close();
+      } catch {
+        // Database might be locked or corrupted
+      }
+    }
+
+    // Convert project ID back to approximate path
+    const approximatePath = projectId
+      .replace(/^([a-z])--/, '$1:/')  // d-- -> d:/
+      .replace(/--/g, '/')
+      .replace(/-/g, ' ');
+
+    projects.push({
+      id: projectId,
+      path: approximatePath,
+      memoriesCount,
+      clustersCount,
+      lastUpdated
+    });
+  }
+
+  // Sort by last updated (most recent first)
+  return projects.sort((a, b) => {
+    if (!a.lastUpdated) return 1;
+    if (!b.lastUpdated) return -1;
+    return b.lastUpdated.localeCompare(a.lastUpdated);
+  });
+}
+
+/**
+ * Get memories from another project by ID
+ */
+export function getMemoriesFromProject(projectId: string): CoreMemory[] {
+  const projectsDir = join(getCCWHome(), 'projects');
+  const coreMemoryDb = join(projectsDir, projectId, 'core-memory', 'core_memory.db');
+
+  if (!existsSync(coreMemoryDb)) {
+    throw new Error(`Project not found: ${projectId}`);
+  }
+
+  const db = new Database(coreMemoryDb, { readonly: true });
+
+  const stmt = db.prepare('SELECT * FROM memories ORDER BY updated_at DESC');
+  const rows = stmt.all() as any[];
+
+  db.close();
+
+  return rows.map(row => ({
+    id: row.id,
+    content: row.content,
+    summary: row.summary || '',
+    raw_output: row.raw_output,
+    created_at: row.created_at,
+    updated_at: row.updated_at,
+    archived: Boolean(row.archived),
+    metadata: row.metadata
+  }));
+}
+
+/**
+ * Find a memory by ID across all projects
+ * Searches through all project databases to locate a specific memory
+ */
+export function findMemoryAcrossProjects(memoryId: string): { memory: CoreMemory; projectId: string } | null {
+  const projectsDir = join(getCCWHome(), 'projects');
+
+  if (!existsSync(projectsDir)) {
+    return null;
+  }
+
+  const entries = readdirSync(projectsDir, { withFileTypes: true });
+
+  for (const entry of entries) {
+    if (!entry.isDirectory()) continue;
+
+    const projectId = entry.name;
+    const coreMemoryDb = join(projectsDir, projectId, 'core-memory', 'core_memory.db');
+
+    if (!existsSync(coreMemoryDb)) continue;
+
+    try {
+      const db = new Database(coreMemoryDb, { readonly: true });
+      const row = db.prepare('SELECT * FROM memories WHERE id = ?').get(memoryId) as any;
+      db.close();
+
+      if (row) {
+        return {
+          memory: {
+            id: row.id,
+            content: row.content,
+            summary: row.summary || '',
+            raw_output: row.raw_output,
+            created_at: row.created_at,
+            updated_at: row.updated_at,
+            archived: Boolean(row.archived),
+            metadata: row.metadata
+          },
+          projectId
+        };
+      }
+    } catch {
+      // Database might be locked or corrupted, skip
+    }
+  }
+
+  return null;
+}
+
+/**
+ * Export memories to a JSON file
+ */
+export function exportMemories(
+  projectPath: string,
+  outputPath: string,
+  options?: { ids?: string[]; includeArchived?: boolean }
+): number {
+  const store = getCoreMemoryStore(projectPath);
+  let memories = store.getMemories({ archived: options?.includeArchived || false, limit: 10000 });
+
+  // Filter by IDs if specified
+  if (options?.ids && options.ids.length > 0) {
+    const idSet = new Set(options.ids);
+    memories = memories.filter(m => idSet.has(m.id));
+  }
+
+  const exportData: ExportedMemory = {
+    version: '1.0',
+    exportedAt: new Date().toISOString(),
+    sourceProject: projectPath,
+    memories
+  };
+
+  writeFileSync(outputPath, JSON.stringify(exportData, null, 2), 'utf-8');
+  return memories.length;
+}
+
+/**
+ * Import memories from a JSON file or another project
+ */
+export function importMemories(
+  targetProjectPath: string,
+  source: string,  // File path or project ID
+  options?: { overwrite?: boolean; prefix?: string }
+): { imported: number; skipped: number } {
+  const store = getCoreMemoryStore(targetProjectPath);
+  let memories: CoreMemory[];
+
+  // Check if source is a file or project ID
+  if (existsSync(source) && source.endsWith('.json')) {
+    // Import from file
+    const content = readFileSync(source, 'utf-8');
+    const data = JSON.parse(content) as ExportedMemory;
+    memories = data.memories;
+  } else {
+    // Import from project ID
+    memories = getMemoriesFromProject(source);
+  }
+
+  let imported = 0;
+  let skipped = 0;
+
+  for (const memory of memories) {
+    // Generate new ID with optional prefix
+    let newId = memory.id;
+    if (options?.prefix) {
+      newId = `${options.prefix}-${memory.id}`;
+    }
+
+    // Check if already exists
+    const existing = store.getMemory(newId);
+    if (existing && !options?.overwrite) {
+      skipped++;
+      continue;
+    }
+
+    // Import memory
+    store.upsertMemory({
+      id: newId,
+      content: memory.content,
+      summary: memory.summary,
+      raw_output: memory.raw_output,
+      metadata: memory.metadata
+    });
+
+    imported++;
+  }
+
+  return { imported, skipped };
+}
+
+/**
+ * Close all store instances
+ */
+export function closeAllStores(): void {
+  const stores = Array.from(storeCache.values());
+  for (const store of stores) {
+    store.close();
+  }
+  storeCache.clear();
+}
+
+export default CoreMemoryStore;