lwazi 1.6.7 → 1.7.0

package/README.md

@@ -1,187 +1,62 @@
-# Lwazi AI Assistant
+# Lwazi
 
-[![npm version](https://badge.fury.io/js/lwazi-ai.svg)](https://badge.fury.io/js/lwazi-ai)
-[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
+AI assistant for Laravel applications.
 
-**AI-powered assistant for Laravel applications.** Add an intelligent chat widget to your Laravel app with just one command.
-
-## Features
-
-- 🤖 **AI-Powered Chat**: Integrates with Ollama for local AI conversations
-- 🎨 **Customizable UI**: Modern chat widget with theme customization
-- 🔍 **Smart Navigation**: AI understands your app structure and helps users navigate
-- 📱 **Responsive Design**: Works on desktop and mobile
-- ⚡ **Easy Installation**: One-command setup for Laravel projects
-- 🔒 **Privacy-Focused**: Runs locally with Ollama, no external API calls
-
-## Quick Start
-
-### Prerequisites
-
-- **Laravel 10+** application
-- **PHP 8.1+**
-- **Node.js 14+** (for installation)
-- **Ollama** running locally
-
-### Installation
+## Install
 
 ```bash
-# Install via npm
-npm install lwazi-ai
-
-# This will automatically run the post-install script
-# Then complete the installation by running:
-npx lwazi
+npm install lwazi
+npx lwazi install
 ```
 
-The installer will:
-- Download the latest Lwazi package
-- Configure your Laravel application
-- Set up environment variables
-- Publish necessary assets
-- Generate your site's knowledge base
-
-### Manual Installation
+## Requirements
 
-If you prefer manual installation:
+- Laravel 10+
+- PHP 8.1+
+- Ollama running locally
 
-```bash
-# Clone the repository
-git clone https://github.com/nigelnkomo/lwazi.git
-cd your-laravel-project
+## Configure
 
-# Copy lwazi folder
-cp -r /path/to/lwazi/lwazi ./
+Add to `.env`:
 
-# Run installation script
-./lwazi/install
 ```
-
-## Configuration
-
-After installation, configure these environment variables in your `.env` file:
-
-```env
-# Lwazi AI Configuration
 LWAZI_OLLAMA_URL=http://localhost:11434
 LWAZI_MODEL=llama3.2:1b
-LWAZI_API_URL=/lwazi
-LWAZI_AUTO_INJECT=true
-LWAZI_POSITION=bottom-right
-LWAZI_THEME=blue
-LWAZI_ACCENT="#007aff"
-LWAZI_ACCENT_DARK="#0051d4"
 ```
 
 ## Usage
 
-Once installed, Lwazi will automatically appear as a chat widget in your Laravel application. Users can:
+Lwazi adds a chat widget to your app automatically. Users can ask about your site and get AI-powered responses.
 
-- Ask questions about your website
-- Get navigation help
-- Interact with your content using AI
-
-### Laravel Commands
+## Commands
 
 ```bash
-# Re-analyze your application structure
-php artisan lwazi:ingest
-
-# Run setup wizard
-php artisan lwazi:setup
-
-# Clear Lwazi cache
-php artisan lwazi:clear
+php artisan lwazi:ingest    # Re-analyze app structure
+php artisan lwazi:clear     # Clear cache
 ```
 
-## How It Works
-
-Lwazi analyzes your Laravel application to understand:
-
-- **Routes & Navigation**: Maps your site's structure
-- **Content & Pages**: Indexes your views and content
-- **Database Schema**: Understands your data relationships
-- **Configuration**: Learns about your app's setup
-
-This knowledge allows the AI to provide contextual, helpful responses specific to your application.
-
 ## Customization
 
-### Themes
-
-Lwazi supports multiple themes and custom colors. Configure via environment variables or publish the config:
+Publish config to customize:
 
 ```bash
 php artisan vendor:publish --provider="Lwazi\Core\Providers\LwaziServiceProvider"
 ```
 
-### Advanced Configuration
-
-Edit `config/lwazi.php` to customize:
+Edit `config/lwazi.php` to change:
 - AI model settings
 - Widget appearance
 - Response behavior
-- Security settings
-
-## Requirements
-
-- **Laravel**: 10.0, 11.0, or 12.0
-- **PHP**: 8.1 or higher
-- **Ollama**: Latest version
-- **Node.js**: 14+ (for installation only)
-
-## Supported AI Models
-
-Lwazi works with any Ollama-compatible model:
-
-- **Llama 3.2**: `llama3.2:1b`, `llama3.2:3b`
-- **Llama 3**: `llama3:8b`, `llama3:70b`
-- **Mistral**: `mistral:7b`
-- **Phi-3**: `phi3:3.8b`
-- **Gemma**: `gemma:2b`, `gemma:7b`
 
 ## Troubleshooting
 
-### Common Issues
-
-**"Ollama not available"**
-- Ensure Ollama is running: `ollama serve`
-- Check the URL in your `.env` file
-
-**"Model not found"**
-- Pull the model: `ollama pull llama3.2:1b`
-- Update `LWAZI_MODEL` in your `.env`
+**Ollama not available**
+- Run `ollama serve`
+- Check `LWAZI_OLLAMA_URL` in `.env`
 
-**Widget not appearing**
-- Clear Laravel caches: `php artisan config:clear && php artisan cache:clear`
-- Check browser console for JavaScript errors
-
-### Getting Help
-
-- 📖 **Documentation**: [GitHub Wiki](https://github.com/nigelnkomo/lwazi/wiki)
-- 🐛 **Issues**: [GitHub Issues](https://github.com/nigelnkomo/lwazi/issues)
-- 💬 **Discussions**: [GitHub Discussions](https://github.com/nigelnkomo/lwazi/discussions)
-
-## Contributing
-
-We welcome contributions! Please see our [Contributing Guide](CONTRIBUTING.md) for details.
-
-1. Fork the repository
-2. Create a feature branch
-3. Make your changes
-4. Add tests if applicable
-5. Submit a pull request
+**Model not found**
+- Run `ollama pull llama3.2:1b`
 
 ## License
 
-This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
-
-## Credits
-
-- **Ollama**: For providing the local AI inference engine
-- **Laravel**: The PHP framework that makes this possible
-- **Open Source Community**: For the amazing tools and libraries
-
----
-
-**Made with ❤️ for the Laravel community**
+MIT

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "lwazi",
-  "version": "1.6.7",
+  "version": "1.7.0",
   "description": "Lwazi is an AI assistant for Laravel. Install with one command to add an AI assistant to your Laravel app.",
   "main": "bin/lwazi.js",
   "bin": {

package/src/Console/AnalyzeProjectCommand.php

@@ -69,6 +69,16 @@ class AnalyzeProjectCommand extends Command
 
         $root = $this->extractRoot($rootUrl);
         
+        if (!isset($manifest['flat'])) {
+            $manifest['flat'] = [];
+        }
+        if (!isset($manifest['nodes'])) {
+            $manifest['nodes'] = [];
+        }
+        if (!isset($manifest['adjacency'])) {
+            $manifest['adjacency'] = [];
+        }
+        
         foreach ($routes as $route) {
             $path = '/' . ltrim($route['uri'], '/');
             

package/src/Console/SetupCommand.php

@@ -149,6 +149,16 @@ class SetupCommand extends Command
 
         $root = $this->extractRoot($rootUrl);
         
+        if (!isset($manifest['flat'])) {
+            $manifest['flat'] = [];
+        }
+        if (!isset($manifest['nodes'])) {
+            $manifest['nodes'] = [];
+        }
+        if (!isset($manifest['adjacency'])) {
+            $manifest['adjacency'] = [];
+        }
+        
         foreach ($routes as $route) {
             $path = '/' . ltrim($route['uri'], '/');
             

package/src/Http/Controllers/ChatController.php

@@ -55,7 +55,7 @@ class ChatController extends Controller
                 $ragService = app(RagService::class);
                 $routes = $ragService->getRouteIndex();
                 
-                $prompt = "What is the single most important word (noun) this user is asking about? Just return the word, nothing else. Question: {$message}";
+                $prompt = "One word only. What does the user want to find? this user is asking about? Just return the word, nothing else. Question: {$message}";
                 
                 $terms = [];
                 $llmError = null;

package/src/Services/GraphVisualizer.php

@@ -10,7 +10,7 @@ class GraphVisualizer
         $nodes = $manifest['nodes'] ?? [];
         $rootUrl = $manifest['root_url'] ?? '';
 
-        if (empty($adjacency)) {
+        if (empty($adjacency) || empty($nodes)) {
             return "No graph data available.\n";
         }
 

package/src/Services/LwaziAgent.php

@@ -43,7 +43,7 @@ class LwaziAgent
 
     public function isReady(): bool
     {
-        return !empty($this->loadManifest()) && $this->isOllamaAvailable();
+        return $this->isOllamaAvailable();
     }
 
     protected function isOllamaAvailable(): bool

package/src/Services/LwaziService.php

@@ -168,7 +168,7 @@ class LwaziService
             return null;
         }
 
-        $prompt = "What is the single most important word (noun) this user is asking about? Just return the word, nothing else. Question: {$message}";
+        $prompt = "One word only. What does the user want to find? this user is asking about? Just return the word, nothing else. Question: {$message}";
 
         $response = $this->callOllama([
             ['role' => 'system', 'content' => 'Return only one word.'],
@@ -337,7 +337,7 @@ class LwaziService
             return null;
         }
 
-        $prompt = "What is the single most important word (noun) this user is asking about? Just return the word, nothing else. Question: {$message}";
+        $prompt = "One word only. What does the user want to find? this user is asking about? Just return the word, nothing else. Question: {$message}";
 
         $response = $this->callOllama([
             ['role' => 'system', 'content' => 'Return only one word.'],