feat: Add DefaultEmbeddingFunction for local vector embeddings

- Added @chroma-core/default-embed dependency for local embeddings
- Updated ChromaSync to use DefaultEmbeddingFunction with collections
- Added isServerReachable() async method for reliable server detection
- Fixed start() to detect and reuse existing Chroma servers
- Updated build script to externalize native ONNX binaries
- Added runtime dependency to plugin/package.json

The embedding function uses all-MiniLM-L6-v2 model locally via ONNX,
eliminating need for external embedding API calls.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

src/services/sync/ChromaServerManager.ts

@@ -48,6 +48,7 @@ export class ChromaServerManager {
   /**
    * Start the Chroma HTTP server
    * Spawns `npx chroma run` as a background process
+   * If a server is already running (from previous worker), reuses it
    */
   async start(): Promise<void> {
     if (this.ready || this.starting) {
@@ -57,6 +58,24 @@ export class ChromaServerManager {
       });
       return;
     }
+
+    // Check if a server is already running (from previous worker or manual start)
+    try {
+      const response = await fetch(
+        `http://${this.config.host}:${this.config.port}/api/v2/heartbeat`
+      );
+      if (response.ok) {
+        logger.info('CHROMA_SERVER', 'Existing server detected, reusing', {
+          host: this.config.host,
+          port: this.config.port
+        });
+        this.ready = true;
+        return;
+      }
+    } catch {
+      // No server running, proceed to start one
+    }
+
     this.starting = true;
 
     // Cross-platform: use npx.cmd on Windows
@@ -159,9 +178,29 @@ export class ChromaServerManager {
 
   /**
    * Check if the server is running and ready
+   * Returns true if we manage the process OR if a server is responding
    */
   isRunning(): boolean {
-    return this.ready && this.serverProcess !== null;
+    return this.ready;
+  }
+
+  /**
+   * Async check if server is running by pinging heartbeat
+   * Use this when you need to verify server is actually reachable
+   */
+  async isServerReachable(): Promise<boolean> {
+    try {
+      const response = await fetch(
+        `http://${this.config.host}:${this.config.port}/api/v2/heartbeat`
+      );
+      if (response.ok) {
+        this.ready = true;
+        return true;
+      }
+    } catch {
+      // Server not reachable
+    }
+    return false;
   }
 
   /**

src/services/sync/ChromaSync.ts

@@ -13,6 +13,7 @@
  */
 
 import { ChromaClient, Collection } from 'chromadb';
+import { DefaultEmbeddingFunction } from '@chroma-core/default-embed';
 import { ParsedObservation, ParsedSummary } from '../../sdk/parser.js';
 import { SessionStore } from '../sqlite/SessionStore.js';
 import { logger } from '../../utils/logger.js';
@@ -108,11 +109,12 @@ export class ChromaSync {
       const port = parseInt(settings.CLAUDE_MEM_CHROMA_PORT || '8000', 10);
       const ssl = settings.CLAUDE_MEM_CHROMA_SSL === 'true';
 
-      // In local mode, verify server is running
+      // In local mode, verify server is reachable
       if (mode === 'local') {
         const serverManager = ChromaServerManager.getInstance();
-        if (!serverManager.isRunning()) {
-          throw new Error('Chroma server not running. Ensure worker started correctly.');
+        const reachable = await serverManager.isServerReachable();
+        if (!reachable) {
+          throw new Error('Chroma server not reachable. Ensure worker started correctly.');
         }
       }
 
@@ -159,8 +161,11 @@ export class ChromaSync {
 
     try {
       // getOrCreateCollection handles both cases
+      // Use DefaultEmbeddingFunction for local embeddings (all-MiniLM-L6-v2)
+      const embeddingFunction = new DefaultEmbeddingFunction();
       this.collection = await this.chromaClient.getOrCreateCollection({
-        name: this.collectionName
+        name: this.collectionName,
+        embeddingFunction
       });
 
       logger.debug('CHROMA_SYNC', 'Collection ready', { collection: this.collectionName });