opc-agent 1.1.1 → 1.1.3

package/src/core/knowledge.ts

@@ -1,210 +1,210 @@
-/**
- * Knowledge Base / RAG - Local vector storage with semantic search
- */
-import * as fs from 'fs';
-import * as path from 'path';
-import * as crypto from 'crypto';
-
-// Simple in-memory vector store (PGlite-compatible interface for future migration)
-interface VectorEntry {
-  id: string;
-  content: string;
-  embedding: number[];
-  metadata: Record<string, unknown>;
-}
-
-interface KnowledgeStore {
-  entries: VectorEntry[];
-  version: number;
-  updatedAt: string;
-}
-
-const CHUNK_SIZE = 500; // chars per chunk
-const CHUNK_OVERLAP = 50;
-const STORE_FILE = '.opc-knowledge.json';
-
-function splitText(text: string, chunkSize = CHUNK_SIZE, overlap = CHUNK_OVERLAP): string[] {
-  const chunks: string[] = [];
-  // Split by paragraphs first, then by size
-  const paragraphs = text.split(/\n\s*\n/).filter(p => p.trim());
-  let current = '';
-
-  for (const para of paragraphs) {
-    if (current.length + para.length > chunkSize && current.length > 0) {
-      chunks.push(current.trim());
-      // Keep overlap from end of current
-      current = current.slice(-overlap) + '\n\n' + para;
-    } else {
-      current += (current ? '\n\n' : '') + para;
-    }
-  }
-  if (current.trim()) chunks.push(current.trim());
-
-  // If any chunk is still too large, split by sentences
-  const result: string[] = [];
-  for (const chunk of chunks) {
-    if (chunk.length <= chunkSize * 1.5) {
-      result.push(chunk);
-    } else {
-      const sentences = chunk.split(/(?<=[.!?])\s+/);
-      let buf = '';
-      for (const s of sentences) {
-        if (buf.length + s.length > chunkSize && buf) {
-          result.push(buf.trim());
-          buf = buf.slice(-overlap) + ' ' + s;
-        } else {
-          buf += (buf ? ' ' : '') + s;
-        }
-      }
-      if (buf.trim()) result.push(buf.trim());
-    }
-  }
-  return result;
-}
-
-// Simple TF-IDF-like embedding (no external dependencies)
-// For production, replace with real embedding API
-function simpleEmbed(text: string): number[] {
-  const words = text.toLowerCase().replace(/[^a-z0-9\s]/g, '').split(/\s+/).filter(Boolean);
-  const dim = 128;
-  const vec = new Array(dim).fill(0);
-
-  for (const word of words) {
-    const hash = crypto.createHash('md5').update(word).digest();
-    for (let i = 0; i < dim; i++) {
-      vec[i] += (hash[i % hash.length] - 128) / 128;
-    }
-  }
-
-  // Normalize
-  const mag = Math.sqrt(vec.reduce((s, v) => s + v * v, 0)) || 1;
-  return vec.map(v => v / mag);
-}
-
-function cosineSimilarity(a: number[], b: number[]): number {
-  let dot = 0, magA = 0, magB = 0;
-  for (let i = 0; i < a.length; i++) {
-    dot += a[i] * b[i];
-    magA += a[i] * a[i];
-    magB += b[i] * b[i];
-  }
-  return dot / (Math.sqrt(magA) * Math.sqrt(magB) || 1);
-}
-
-export class KnowledgeBase {
-  private store: KnowledgeStore;
-  private storePath: string;
-
-  constructor(baseDir: string = '.') {
-    this.storePath = path.join(baseDir, STORE_FILE);
-    this.store = this.load();
-  }
-
-  private load(): KnowledgeStore {
-    try {
-      if (fs.existsSync(this.storePath)) {
-        return JSON.parse(fs.readFileSync(this.storePath, 'utf-8'));
-      }
-    } catch { /* ignore */ }
-    return { entries: [], version: 1, updatedAt: new Date().toISOString() };
-  }
-
-  private save(): void {
-    this.store.updatedAt = new Date().toISOString();
-    fs.writeFileSync(this.storePath, JSON.stringify(this.store), 'utf-8');
-  }
-
-  async addFile(filePath: string): Promise<{ chunks: number }> {
-    const absPath = path.resolve(filePath);
-    if (!fs.existsSync(absPath)) {
-      throw new Error(`File not found: ${absPath}`);
-    }
-
-    const content = fs.readFileSync(absPath, 'utf-8');
-    const filename = path.basename(absPath);
-
-    // Remove existing entries for this file
-    this.store.entries = this.store.entries.filter(
-      e => e.metadata.source !== filename
-    );
-
-    const chunks = splitText(content);
-    for (let i = 0; i < chunks.length; i++) {
-      const chunk = chunks[i];
-      this.store.entries.push({
-        id: `${filename}_${i}_${Date.now()}`,
-        content: chunk,
-        embedding: simpleEmbed(chunk),
-        metadata: {
-          source: filename,
-          chunkIndex: i,
-          totalChunks: chunks.length,
-          addedAt: new Date().toISOString(),
-        },
-      });
-    }
-
-    this.save();
-    return { chunks: chunks.length };
-  }
-
-  async addText(text: string, source: string = 'manual'): Promise<{ chunks: number }> {
-    const chunks = splitText(text);
-    for (let i = 0; i < chunks.length; i++) {
-      this.store.entries.push({
-        id: `${source}_${i}_${Date.now()}`,
-        content: chunks[i],
-        embedding: simpleEmbed(chunks[i]),
-        metadata: { source, chunkIndex: i, totalChunks: chunks.length, addedAt: new Date().toISOString() },
-      });
-    }
-    this.save();
-    return { chunks: chunks.length };
-  }
-
-  async search(query: string, topK: number = 5): Promise<Array<{ content: string; score: number; source: string }>> {
-    if (this.store.entries.length === 0) return [];
-
-    const queryEmb = simpleEmbed(query);
-    const scored = this.store.entries.map(entry => ({
-      content: entry.content,
-      score: cosineSimilarity(queryEmb, entry.embedding),
-      source: String(entry.metadata.source ?? 'unknown'),
-    }));
-
-    scored.sort((a, b) => b.score - a.score);
-    return scored.slice(0, topK);
-  }
-
-  /** Build context string for injection into LLM calls */
-  async getContext(query: string, topK: number = 3, minScore: number = 0.1): Promise<string> {
-    const results = await this.search(query, topK);
-    const relevant = results.filter(r => r.score >= minScore);
-    if (relevant.length === 0) return '';
-
-    return `\n\n--- Relevant Knowledge ---\n${relevant.map((r, i) =>
-      `[${i + 1}] (source: ${r.source}, relevance: ${(r.score * 100).toFixed(0)}%)\n${r.content}`
-    ).join('\n\n')}\n--- End Knowledge ---\n`;
-  }
-
-  getStats(): { totalEntries: number; sources: string[]; updatedAt: string } {
-    const sources = [...new Set(this.store.entries.map(e => String(e.metadata.source)))];
-    return {
-      totalEntries: this.store.entries.length,
-      sources,
-      updatedAt: this.store.updatedAt,
-    };
-  }
-
-  clear(): void {
-    this.store.entries = [];
-    this.save();
-  }
-
-  removeSource(source: string): number {
-    const before = this.store.entries.length;
-    this.store.entries = this.store.entries.filter(e => e.metadata.source !== source);
-    this.save();
-    return before - this.store.entries.length;
-  }
-}
+/**
+ * Knowledge Base / RAG - Local vector storage with semantic search
+ */
+import * as fs from 'fs';
+import * as path from 'path';
+import * as crypto from 'crypto';
+
+// Simple in-memory vector store (PGlite-compatible interface for future migration)
+interface VectorEntry {
+  id: string;
+  content: string;
+  embedding: number[];
+  metadata: Record<string, unknown>;
+}
+
+interface KnowledgeStore {
+  entries: VectorEntry[];
+  version: number;
+  updatedAt: string;
+}
+
+const CHUNK_SIZE = 500; // chars per chunk
+const CHUNK_OVERLAP = 50;
+const STORE_FILE = '.opc-knowledge.json';
+
+function splitText(text: string, chunkSize = CHUNK_SIZE, overlap = CHUNK_OVERLAP): string[] {
+  const chunks: string[] = [];
+  // Split by paragraphs first, then by size
+  const paragraphs = text.split(/\n\s*\n/).filter(p => p.trim());
+  let current = '';
+
+  for (const para of paragraphs) {
+    if (current.length + para.length > chunkSize && current.length > 0) {
+      chunks.push(current.trim());
+      // Keep overlap from end of current
+      current = current.slice(-overlap) + '\n\n' + para;
+    } else {
+      current += (current ? '\n\n' : '') + para;
+    }
+  }
+  if (current.trim()) chunks.push(current.trim());
+
+  // If any chunk is still too large, split by sentences
+  const result: string[] = [];
+  for (const chunk of chunks) {
+    if (chunk.length <= chunkSize * 1.5) {
+      result.push(chunk);
+    } else {
+      const sentences = chunk.split(/(?<=[.!?])\s+/);
+      let buf = '';
+      for (const s of sentences) {
+        if (buf.length + s.length > chunkSize && buf) {
+          result.push(buf.trim());
+          buf = buf.slice(-overlap) + ' ' + s;
+        } else {
+          buf += (buf ? ' ' : '') + s;
+        }
+      }
+      if (buf.trim()) result.push(buf.trim());
+    }
+  }
+  return result;
+}
+
+// Simple TF-IDF-like embedding (no external dependencies)
+// For production, replace with real embedding API
+function simpleEmbed(text: string): number[] {
+  const words = text.toLowerCase().replace(/[^a-z0-9\s]/g, '').split(/\s+/).filter(Boolean);
+  const dim = 128;
+  const vec = new Array(dim).fill(0);
+
+  for (const word of words) {
+    const hash = crypto.createHash('md5').update(word).digest();
+    for (let i = 0; i < dim; i++) {
+      vec[i] += (hash[i % hash.length] - 128) / 128;
+    }
+  }
+
+  // Normalize
+  const mag = Math.sqrt(vec.reduce((s, v) => s + v * v, 0)) || 1;
+  return vec.map(v => v / mag);
+}
+
+function cosineSimilarity(a: number[], b: number[]): number {
+  let dot = 0, magA = 0, magB = 0;
+  for (let i = 0; i < a.length; i++) {
+    dot += a[i] * b[i];
+    magA += a[i] * a[i];
+    magB += b[i] * b[i];
+  }
+  return dot / (Math.sqrt(magA) * Math.sqrt(magB) || 1);
+}
+
+export class KnowledgeBase {
+  private store: KnowledgeStore;
+  private storePath: string;
+
+  constructor(baseDir: string = '.') {
+    this.storePath = path.join(baseDir, STORE_FILE);
+    this.store = this.load();
+  }
+
+  private load(): KnowledgeStore {
+    try {
+      if (fs.existsSync(this.storePath)) {
+        return JSON.parse(fs.readFileSync(this.storePath, 'utf-8'));
+      }
+    } catch { /* ignore */ }
+    return { entries: [], version: 1, updatedAt: new Date().toISOString() };
+  }
+
+  private save(): void {
+    this.store.updatedAt = new Date().toISOString();
+    fs.writeFileSync(this.storePath, JSON.stringify(this.store), 'utf-8');
+  }
+
+  async addFile(filePath: string): Promise<{ chunks: number }> {
+    const absPath = path.resolve(filePath);
+    if (!fs.existsSync(absPath)) {
+      throw new Error(`File not found: ${absPath}`);
+    }
+
+    const content = fs.readFileSync(absPath, 'utf-8');
+    const filename = path.basename(absPath);
+
+    // Remove existing entries for this file
+    this.store.entries = this.store.entries.filter(
+      e => e.metadata.source !== filename
+    );
+
+    const chunks = splitText(content);
+    for (let i = 0; i < chunks.length; i++) {
+      const chunk = chunks[i];
+      this.store.entries.push({
+        id: `${filename}_${i}_${Date.now()}`,
+        content: chunk,
+        embedding: simpleEmbed(chunk),
+        metadata: {
+          source: filename,
+          chunkIndex: i,
+          totalChunks: chunks.length,
+          addedAt: new Date().toISOString(),
+        },
+      });
+    }
+
+    this.save();
+    return { chunks: chunks.length };
+  }
+
+  async addText(text: string, source: string = 'manual'): Promise<{ chunks: number }> {
+    const chunks = splitText(text);
+    for (let i = 0; i < chunks.length; i++) {
+      this.store.entries.push({
+        id: `${source}_${i}_${Date.now()}`,
+        content: chunks[i],
+        embedding: simpleEmbed(chunks[i]),
+        metadata: { source, chunkIndex: i, totalChunks: chunks.length, addedAt: new Date().toISOString() },
+      });
+    }
+    this.save();
+    return { chunks: chunks.length };
+  }
+
+  async search(query: string, topK: number = 5): Promise<Array<{ content: string; score: number; source: string }>> {
+    if (this.store.entries.length === 0) return [];
+
+    const queryEmb = simpleEmbed(query);
+    const scored = this.store.entries.map(entry => ({
+      content: entry.content,
+      score: cosineSimilarity(queryEmb, entry.embedding),
+      source: String(entry.metadata.source ?? 'unknown'),
+    }));
+
+    scored.sort((a, b) => b.score - a.score);
+    return scored.slice(0, topK);
+  }
+
+  /** Build context string for injection into LLM calls */
+  async getContext(query: string, topK: number = 3, minScore: number = 0.1): Promise<string> {
+    const results = await this.search(query, topK);
+    const relevant = results.filter(r => r.score >= minScore);
+    if (relevant.length === 0) return '';
+
+    return `\n\n--- Relevant Knowledge ---\n${relevant.map((r, i) =>
+      `[${i + 1}] (source: ${r.source}, relevance: ${(r.score * 100).toFixed(0)}%)\n${r.content}`
+    ).join('\n\n')}\n--- End Knowledge ---\n`;
+  }
+
+  getStats(): { totalEntries: number; sources: string[]; updatedAt: string } {
+    const sources = [...new Set(this.store.entries.map(e => String(e.metadata.source)))];
+    return {
+      totalEntries: this.store.entries.length,
+      sources,
+      updatedAt: this.store.updatedAt,
+    };
+  }
+
+  clear(): void {
+    this.store.entries = [];
+    this.save();
+  }
+
+  removeSource(source: string): number {
+    const before = this.store.entries.length;
+    this.store.entries = this.store.entries.filter(e => e.metadata.source !== source);
+    this.save();
+    return before - this.store.entries.length;
+  }
+}

package/src/core/logger.ts

@@ -1,57 +1,57 @@
-/**
- * Structured logger with log levels.
- */
-export type LogLevel = 'debug' | 'info' | 'warn' | 'error';
-
-const LOG_LEVEL_ORDER: Record<LogLevel, number> = {
-  debug: 0,
-  info: 1,
-  warn: 2,
-  error: 3,
-};
-
-export class Logger {
-  private context: string;
-  private level: LogLevel;
-
-  constructor(context: string, level: LogLevel = 'info') {
-    this.context = context;
-    this.level = level;
-  }
-
-  private shouldLog(level: LogLevel): boolean {
-    return LOG_LEVEL_ORDER[level] >= LOG_LEVEL_ORDER[this.level];
-  }
-
-  private format(level: LogLevel, message: string, data?: Record<string, unknown>): string {
-    const ts = new Date().toISOString();
-    const prefix = `[${ts}] [${level.toUpperCase()}] [${this.context}]`;
-    return data ? `${prefix} ${message} ${JSON.stringify(data)}` : `${prefix} ${message}`;
-  }
-
-  debug(message: string, data?: Record<string, unknown>): void {
-    if (this.shouldLog('debug')) console.debug(this.format('debug', message, data));
-  }
-
-  info(message: string, data?: Record<string, unknown>): void {
-    if (this.shouldLog('info')) console.info(this.format('info', message, data));
-  }
-
-  warn(message: string, data?: Record<string, unknown>): void {
-    if (this.shouldLog('warn')) console.warn(this.format('warn', message, data));
-  }
-
-  error(message: string, data?: Record<string, unknown>): void {
-    if (this.shouldLog('error')) console.error(this.format('error', message, data));
-  }
-
-  setLevel(level: LogLevel): void {
-    this.level = level;
-  }
-
-  child(context: string): Logger {
-    return new Logger(`${this.context}:${context}`, this.level);
-  }
-}
-
-export const defaultLogger = new Logger('opc');
+/**
+ * Structured logger with log levels.
+ */
+export type LogLevel = 'debug' | 'info' | 'warn' | 'error';
+
+const LOG_LEVEL_ORDER: Record<LogLevel, number> = {
+  debug: 0,
+  info: 1,
+  warn: 2,
+  error: 3,
+};
+
+export class Logger {
+  private context: string;
+  private level: LogLevel;
+
+  constructor(context: string, level: LogLevel = 'info') {
+    this.context = context;
+    this.level = level;
+  }
+
+  private shouldLog(level: LogLevel): boolean {
+    return LOG_LEVEL_ORDER[level] >= LOG_LEVEL_ORDER[this.level];
+  }
+
+  private format(level: LogLevel, message: string, data?: Record<string, unknown>): string {
+    const ts = new Date().toISOString();
+    const prefix = `[${ts}] [${level.toUpperCase()}] [${this.context}]`;
+    return data ? `${prefix} ${message} ${JSON.stringify(data)}` : `${prefix} ${message}`;
+  }
+
+  debug(message: string, data?: Record<string, unknown>): void {
+    if (this.shouldLog('debug')) console.debug(this.format('debug', message, data));
+  }
+
+  info(message: string, data?: Record<string, unknown>): void {
+    if (this.shouldLog('info')) console.info(this.format('info', message, data));
+  }
+
+  warn(message: string, data?: Record<string, unknown>): void {
+    if (this.shouldLog('warn')) console.warn(this.format('warn', message, data));
+  }
+
+  error(message: string, data?: Record<string, unknown>): void {
+    if (this.shouldLog('error')) console.error(this.format('error', message, data));
+  }
+
+  setLevel(level: LogLevel): void {
+    this.level = level;
+  }
+
+  child(context: string): Logger {
+    return new Logger(`${this.context}:${context}`, this.level);
+  }
+}
+
+export const defaultLogger = new Logger('opc');