agentdb 1.0.5 → 1.0.8

package/CHANGELOG.md

@@ -5,6 +5,32 @@ All notable changes to AgentDB will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [1.0.7] - 2025-10-18
+
+### Fixed
+- **Browser Bundle:** Removed better-sqlite3 dependency from browser builds completely
+- Made NativeBackend import fully dynamic to prevent bundling Node.js dependencies
+- Browser bundle size reduced to 89KB (down from 196KB in v1.0.5)
+- Fixed "Failed to resolve module specifier 'better-sqlite3'" error in browser WASM
+
+### Changed
+- NativeBackend now uses lazy loading with require() instead of static imports
+- Removed require.resolve() check from backend detection for browser compatibility
+- Added native-backend paths to esbuild external list for browser builds
+
+## [1.0.6] - 2025-10-18
+
+### Added
+- **Claude Code Setup Guide** - Complete documentation for MCP integration
+- Quick setup command: `claude mcp add agentdb npx agentdb@1.0.5 mcp`
+- Updated README with all 20 MCP tools listed (10 core + 10 learning)
+- Direct link to comprehensive Claude Code setup guide
+
+### Changed
+- README now shows quick setup command first (better UX)
+- Improved MCP tools documentation with descriptions
+- Added link to detailed setup guide in docs/
+
 ## [1.0.5] - 2025-10-18
 
 ### Added

package/README.md

@@ -73,28 +73,55 @@ It's the missing layer that lets agents **remember what worked, learn what didn'
 npm install agentdb
 ```
 
-### For Claude Desktop / MCP Integration
+### For Claude Code / MCP Integration
 
-Add AgentDB as an MCP server in your Claude Desktop config:
+**Quick Setup (Recommended):**
+
+```bash
+claude mcp add agentdb npx agentdb@1.0.7 mcp
+```
+
+This automatically configures Claude Code with all 20 AgentDB tools (10 core + 10 learning tools).
+
+**Manual Setup:**
+
+Add AgentDB to your Claude Desktop config (`~/.config/claude/claude_desktop_config.json`):
 
 ```json
 {
   "mcpServers": {
     "agentdb": {
       "command": "npx",
-      "args": ["agentdb", "mcp"]
+      "args": ["agentdb@1.0.5", "mcp"]
     }
   }
 }
 ```
 
-**Available MCP Tools:**
+**Available MCP Tools (20 total):**
+
+*Core Tools (10):*
 - `agentdb_init` - Initialize vector database
 - `agentdb_insert` / `agentdb_insert_batch` - Store vectors
-- `agentdb_search` - Semantic search
-- `agentdb_pattern_store` / `agentdb_pattern_search` - ReasoningBank
+- `agentdb_search` - Semantic similarity search
+- `agentdb_pattern_store` / `agentdb_pattern_search` - ReasoningBank patterns
 - `agentdb_stats` - Database metrics
-- ...and 5 more tools
+- `agentdb_delete` - Delete vectors
+- `agentdb_pattern_stats` - Pattern statistics
+- `agentdb_clear_cache` - Clear query cache
+
+*Learning Tools (10):*
+- `learning_start_session` / `learning_end_session` - Session management
+- `learning_predict` - AI-recommended actions with confidence
+- `learning_feedback` - Provide user feedback
+- `learning_train` - Train policies on experience
+- `learning_metrics` - Performance metrics
+- `learning_transfer` - Transfer learning between tasks
+- `learning_explain` - Explain AI predictions
+- `experience_record` - Record tool executions
+- `reward_signal` - Calculate multi-dimensional rewards
+
+[📚 Full Claude Code Setup Guide](docs/CLAUDE_CODE_SETUP.md)
 
 ### CLI Usage
 

package/bin/agentdb.js

@@ -370,28 +370,34 @@ function initDatabase(path) {
   console.log('');
 
   try {
-    const { SqliteVectorDB, Presets } = require('../dist/index.js');
+    const { SQLiteVectorDB, Presets } = require('../dist/index.js');
 
     // Create a small test database
     const config = Presets.smallDataset(1536, path);
-    const db = SqliteVectorDB.new(config);
+    const db = new SQLiteVectorDB(config);
 
     console.log('✅ Database initialized successfully!');
     console.log('');
     console.log('Configuration:');
     console.log(`  Path: ${path}`);
-    console.log(`  Dimension: 1536`);
-    console.log(`  Mode: persistent`);
+    console.log(`  Mode: ${config.memoryMode ? 'in-memory' : 'persistent'}`);
+    console.log(`  Cache Size: ${config.cacheSize}MB`);
+    console.log(`  WAL Mode: ${config.walMode ? 'enabled' : 'disabled'}`);
     console.log('');
     console.log('Next steps:');
-    console.log('  1. Import the library: const { SqliteVectorDB } = require("agentdb")');
-    console.log('  2. Insert vectors: await db.insert({ data: [...], metadata: {...} })');
-    console.log('  3. Search: await db.search(queryVector, 5)');
+    console.log('  1. Import the library: const { SQLiteVectorDB } = require("agentdb")');
+    console.log('  2. Insert vectors: db.insert({ embedding: [...], metadata: {...} })');
+    console.log('  3. Search: db.search(queryVector, 5, "cosine", 0.0)');
+    console.log('  4. View the documentation: https://github.com/ruvnet/agentic-flow/tree/main/packages/agentdb');
     console.log('');
 
     db.close();
   } catch (error) {
     console.error('❌ Failed to initialize database:', error.message);
+    if (error.stack) {
+      console.error('\nStack trace:');
+      console.error(error.stack);
+    }
     process.exit(1);
   }
 }