Merge main into feature/context-routes and optimize

- Fixed all merge conflicts by choosing simpler database export approach
- Simplified tests to use mock.module instead of global testDb
- Optimized context routes:
  - Use INSERT OR REPLACE for simpler upsert logic
  - Added transaction support for atomic operations
  - Use JSON.stringify for embeddings (more efficient)
  - Added index migration for (external_id, source) queries
  - Improved validation error messages
- Fixed asterisks in test file paths (replaced with :memory:)
- Removed asterisk file from root directory

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <[email protected]>

Author: zackify

README.md

@@ -12,10 +12,19 @@
 
 **Re**call **vect**ors is your personal memory vault 🔒 - a self-hosted tool to persist and recall any information indefinitely. Never lose valuable knowledge again!
 
+## 🎯 What was this made for?
+
+Originally built for **recalling articles and blog posts** you've read in the past without remembering the exact title or content. You know you read something relevant, but can't quite recall the details? revect helps you find it with semantic search! 📚
+
+But we've realized it's incredibly useful for **shared and persistent memory for teams and individuals across many AI tools**. Whether you're working solo or collaborating with others, revect becomes your collective knowledge base that any AI can tap into. 🤝
+
 With MCP support, you can use revect as a private way to own your data and recall it seamlessly in any AI system. Your data, your control. 🛡️
 
+> 🔄 **Embedding Model Freedom**: Bring any embedding model, any size - from tiny efficient models to massive powerhouses. Swap between OpenAI, local Ollama models, or any OpenAI-compatible provider instantly. revect automatically re-embeds your entire database when you switch, ensuring zero data loss and maximum flexibility! 🪄
+
+- 🏗️ **Single Service Simplicity** - Just one container vs complex multi-service alternatives - simpler architecture, easier setup
+- 💾 **Portable & Browsable** - Store data in SQLite that you can browse anytime with any SQLite viewer
 - 🔍 **Semantic Superpowers** - Find and retrieve articles from your past with powerful semantic search
-- 💾 **Portable & Simple** - Store your data in a simple, portable SQLite file format
 - 💬 **AI Memory Bridge** - Instantly recall past conversations across different AI providers
 - 🔒 **Privacy First** - Enjoy complete privacy with fully local, offline operation
 - 🔌 **Extensible Ecosystem** - Connect with expanding web interfaces and third-party integrations
@@ -36,6 +45,7 @@ With MCP support, you can use revect as a private way to own your data and recal
   - [🐳 Docker + Local AI Setup](#-running-fully-local-with-docker--ollama--lm-studio)
   - [🧠 Running Multiple Containers](#-running-multiple-containers)
   - [🔌 MCP Configuration](#-mcp-setup)
+  - [🔌 Claude Code Setup](#-claude-code-setup)
   - [💬 Using MCP](#-mcp-usage)
   - [🔄 Switching Embedding Models](#-switching-embedding-models)
 - [☁️ Cloud Option](#️-cloud-option)
@@ -48,6 +58,7 @@ It's our hosted platform with additional features and seamless synchronization.
 
 Get excited about what's coming next! 🎉
 
+- [ ] 🔍 **Advanced Search Types** - Date range filtering, exact match search, and other search refinements
 - [ ] 🌐 **Browser Extension** - Auto-save or right-click to save URLs and articles
 - [ ] 📝 **Obsidian Integration** - Pull in all content and search inside Obsidian
 - [ ] 🖥️ **Web Dashboard** - Search more deeply and interact better with your data
@@ -138,6 +149,8 @@ This approach lets you ask your AI to recall from specific knowledge domains. Fo
 
 Connect revect to your favorite AI tools with the Model Context Protocol! 🤝
 
+> **Note:** revect provides both REST API and MCP service in one unified server, unlike many current MCP projects that require separate services or complex architectures.
+
 1. **Direct Connection** (for supported tools):
 
 ```
@@ -158,6 +171,23 @@ url: http://localhost:8000/mcp
 }
 ```
 
+### 🔌 Claude Code Setup
+
+Once you have revect configured in Claude Desktop above, you can also add it to Claude Code for seamless access to your memory across all Claude interfaces.
+
+Add to Claude Code with this command:
+
+```bash
+claude mcp add-from-claude-desktop
+```
+
+Now you can recall information and persist information for quick access outside of Claude Code's hard coded memory files.
+
+**Benefits for multi-environment usage:**
+- 🐳 **Perfect for Docker environments** - Access your memory vault from Claude Code running in containers without duplicating configuration files
+- 🔄 **Unified MCP client management** - Avoid manually configuring MCP servers across multiple environments and development setups
+- 🌐 **Consistent memory access** - Whether you're in Claude Desktop, Claude Code, or other MCP-enabled tools, your memory vault stays connected
+
 ### 💬 MCP Usage
 
 Transform your AI into a knowledge powerhouse! Here's how to use revect's MCP features:

src/database/migrations.ts

@@ -98,6 +98,18 @@ export const migrations: Migration[] = [
       return db.exec(`DROP TABLE metadata;`);
     },
   },
+  {
+    name: "create_external_id_source_index",
+    up: () => {
+      return db.exec(`
+        CREATE INDEX IF NOT EXISTS idx_documents_external_id_source 
+        ON documents(external_id, source);
+      `);
+    },
+    down: () => {
+      return db.exec(`DROP INDEX IF EXISTS idx_documents_external_id_source;`);
+    },
+  },
 
   // Add more migrations here
 ];

src/routes/context/getContext.ts

@@ -14,7 +14,7 @@ export const getContext = async (body: GetContextProps): Promise<GetContextResul
 
   if (!success) {
     return {
-      error: "Validation failed",
+      error: error.issues?.[0]?.message || "Validation failed",
     };
   }
 

src/routes/context/setContext.ts

@@ -16,7 +16,7 @@ export const setContext = async (body: SetContextProps): Promise<SetContextResul
 
   if (!success) {
     return {
-      error: "Validation failed",
+      error: error.issues?.[0]?.message || "Validation failed",
     };
   }
 
@@ -30,44 +30,33 @@ export const setContext = async (body: SetContextProps): Promise<SetContextResul
   }
 
   try {
-    // Check if context with this key already exists
-    const existingContext = db
-      .query("SELECT id FROM documents WHERE external_id = ? AND source = 'context'")
-      .get(data.key);
-
-    if (existingContext) {
-      // Update existing context
-      db.query(
-        `
-          UPDATE documents 
-          SET 
-            text = ?,
-            embeddings = ?,
-            metadata = ?
-          WHERE external_id = ? AND source = 'context'
-        `
-      ).run(
-        data.message,
-        `[${embeddings.join(",")}]`,
-        JSON.stringify({ type: "context" }),
-        data.key
-      );
-    } else {
-      // Insert new context
+    // Use a transaction for atomic operation
+    db.run("BEGIN TRANSACTION");
+    
+    try {
+      // Use JSON.stringify for embeddings array (more efficient than join)
+      const embeddingsStr = JSON.stringify(embeddings);
+      const metadata = JSON.stringify({ type: "context" });
+      
+      // Use INSERT OR REPLACE for simpler logic
       db.query(
         `
-          INSERT INTO documents (external_id, text, embeddings, source, metadata)
+          INSERT OR REPLACE INTO documents (external_id, text, embeddings, source, metadata)
           VALUES (?, ?, ?, 'context', ?)
         `
       ).run(
         data.key,
         data.message,
-        `[${embeddings.join(",")}]`,
-        JSON.stringify({ type: "context" })
+        embeddingsStr,
+        metadata
       );
+      
+      db.run("COMMIT");
+      return { message: "Context successfully set" };
+    } catch (e) {
+      db.run("ROLLBACK");
+      throw e;
     }
-
-    return { message: "Context successfully set" };
   } catch (e) {
     console.error("Error setting context:", e);
     return { error: "Failed to set context" };

tests/contextRoutes.test.ts

@@ -12,7 +12,7 @@ import * as sqliteVec from "sqlite-vec";
 import { createDocumentsTableSQL } from "../src/database/migrations";
 
 // Set test environment variables
-process.env.DATABASE_PATH = "******"; // In-memory database for tests
+process.env.DATABASE_PATH = ":memory:"; // In-memory database for tests
 process.env.AI_API_KEY = "test-key";
 process.env.AI_EMBEDDING_MODEL = "test-model";
 
@@ -34,7 +34,7 @@ describe("Context Routes", () => {
 
   beforeAll(async () => {
     // Create fresh database
-    db = new Database("******");
+    db = new Database(":memory:");
 
     // Configure database
     db.exec("PRAGMA journal_mode = WAL;");
@@ -127,7 +127,7 @@ describe("Context Routes", () => {
 
     // Verify validation error
     expect(response.status).toBe(400);
-    expect(responseData).toHaveProperty("error", "Validation failed");
+    expect(responseData).toHaveProperty("error", "Key field is required");
   });
 
   test("should handle non-existent context key", async () => {

tests/documentRoute.test.ts

@@ -13,7 +13,7 @@ import * as sqliteVec from "sqlite-vec";
 import { createDocumentsTableSQL } from "../src/database/migrations";
 
 // Set test environment variables
-process.env.DATABASE_PATH = "******"; // In-memory database for tests
+process.env.DATABASE_PATH = ":memory:"; // In-memory database for tests
 process.env.AI_API_KEY = "test-key";
 process.env.AI_EMBEDDING_MODEL = "test-model";
 
@@ -38,7 +38,7 @@ describe("Document Route", () => {
   // Create a fresh test setup before tests
   beforeAll(async () => {
     // Create fresh database
-    db = new Database("******");
+    db = new Database(":memory:");
 
     // Configure database
     db.exec("PRAGMA journal_mode = WAL;");

tests/helpers/mockDb.ts

@@ -8,7 +8,7 @@ import { createDocumentsTableSQL, createDocumentChunksTableSQL } from "../../src
  */
 export function createTestDb() {
   // Create a new in-memory database
-  const testDb = new Database("******");
+  const testDb = new Database(":memory:");
 
   // Enable WAL mode
   testDb.exec("PRAGMA journal_mode = WAL;");

tests/index.test.ts

@@ -17,7 +17,7 @@ import {
 import { indexRoute } from "../src/routes/index/index";
 
 // Set test environment variables
-process.env.DATABASE_PATH = "******"; // In-memory database for tests
+process.env.DATABASE_PATH = ":memory:"; // In-memory database for tests
 process.env.AI_API_KEY = "test-key";
 process.env.AI_EMBEDDING_MODEL = "test-model";
 
@@ -39,13 +39,24 @@ describe("Index and Search routes", () => {
 
   beforeAll(async () => {
     // Create fresh database
-    db = new Database("******");
+    db = new Database(":memory:");
 
     // Configure database
     db.exec("PRAGMA journal_mode = WAL;");
     sqliteVec.load(db);
     db.exec(createDocumentsTableSQL("1536"));
     db.exec(createDocumentChunksTableSQL("1536"));
+    
+    // Create metadata table
+    db.exec(`
+      CREATE TABLE IF NOT EXISTS metadata (
+        id INTEGER PRIMARY KEY AUTOINCREMENT,
+        key TEXT UNIQUE NOT NULL,
+        value TEXT,
+        created_at DATETIME DEFAULT CURRENT_TIMESTAMP,
+        updated_at DATETIME DEFAULT CURRENT_TIMESTAMP
+      );
+    `);
 
     // Spy on database module to return our test db
     mock.module("../src/database/database", () => ({

tests/indexRoute.test.ts

@@ -13,7 +13,7 @@ import * as sqliteVec from "sqlite-vec";
 import { createDocumentsTableSQL, createDocumentChunksTableSQL } from "../src/database/migrations";
 
 // Set test environment variables
-process.env.DATABASE_PATH = "******"; // In-memory database for tests
+process.env.DATABASE_PATH = ":memory:"; // In-memory database for tests
 process.env.AI_API_KEY = "test-key";
 process.env.AI_EMBEDDING_MODEL = "test-model";
 
@@ -35,14 +35,25 @@ describe("Index Route", () => {
 
   beforeAll(async () => {
     // Create fresh database
-    db = new Database("******");
+    db = new Database(":memory:");
 
     // Configure database
     db.exec("PRAGMA journal_mode = WAL;");
     sqliteVec.load(db);
     db.exec(createDocumentsTableSQL("1536"));
     db.exec(createDocumentChunksTableSQL("1536"));
 
+    // Create metadata table
+    db.exec(`
+      CREATE TABLE IF NOT EXISTS metadata (
+        id INTEGER PRIMARY KEY AUTOINCREMENT,
+        key TEXT UNIQUE NOT NULL,
+        value TEXT,
+        created_at DATETIME DEFAULT CURRENT_TIMESTAMP,
+        updated_at DATETIME DEFAULT CURRENT_TIMESTAMP
+      );
+    `);
+    
     // Spy on database module to return our test db
     mock.module("../src/database/database", () => ({
       db: db

tests/searchRoute.test.ts

@@ -13,7 +13,7 @@ import * as sqliteVec from "sqlite-vec";
 import { createDocumentsTableSQL, createDocumentChunksTableSQL } from "../src/database/migrations";
 
 // Set test environment variables
-process.env.DATABASE_PATH = "******"; // In-memory database for tests
+process.env.DATABASE_PATH = ":memory:"; // In-memory database for tests
 process.env.AI_API_KEY = "test-key";
 process.env.AI_EMBEDDING_MODEL = "test-model";
 
@@ -35,14 +35,25 @@ describe("Search Route", () => {
 
   beforeAll(async () => {
     // Create fresh database
-    db = new Database("******");
+    db = new Database(":memory:");
 
     // Configure database
     db.exec("PRAGMA journal_mode = WAL;");
     sqliteVec.load(db);
     db.exec(createDocumentsTableSQL("1536"));
     db.exec(createDocumentChunksTableSQL("1536"));
 
+    // Create metadata table
+    db.exec(`
+      CREATE TABLE IF NOT EXISTS metadata (
+        id INTEGER PRIMARY KEY AUTOINCREMENT,
+        key TEXT UNIQUE NOT NULL,
+        value TEXT,
+        created_at DATETIME DEFAULT CURRENT_TIMESTAMP,
+        updated_at DATETIME DEFAULT CURRENT_TIMESTAMP
+      );
+    `);
+    
     // Spy on database module to return our test db
     mock.module("../src/database/database", () => ({
       db: db