Laravel Log Analysis AI

An AI-powered log analysis system that automatically monitors, vectorizes, and analyzes Laravel application logs to detect incidents and assess severity levels.

Features

• Real-time Log Monitoring: MCP server integration for continuous log file watching
• AI-Powered Analysis: Uses LLMs (Claude, GPT, etc.) via Prism to analyze log entries
• Semantic Search: Vector embeddings enable similarity-based log grouping
• Incident Detection: Automatic severity classification (low, medium, high, critical)
• Critical Alert System: Dismissible banners for critical incidents with viewed status tracking
• Test Error Generator: One-click generation of 5-10 random test logs with automatic AI analysis
• Interactive Dashboard: Livewire-powered UI with real-time updates and polling
• Asynchronous Processing: Queue-based architecture for scalable analysis

Tech Stack

• Laravel 12: Latest Laravel framework
• Prism v0.96: Multi-provider LLM integration
• Overpass v0.7: Python ML bridge for embeddings
• Livewire 3 + Volt: Interactive, real-time dashboard
• Laravel MCP v0.3: Model Context Protocol server
• SQLite: Vector embeddings stored as JSON for future semantic search
• Tailwind CSS v4: Modern styling with dark mode
• Sentence Transformers: all-MiniLM-L6-v2 for 384-dim embeddings

Prerequisites

• PHP 8.2 or higher
• Composer
• Node.js & NPM
• Python 3.9 or higher
• SQLite 3.x

Installation

1. Clone the repository

   git clone <repository-url>
   cd LogAnalysisAi
   

2. Install PHP dependencies

   composer install
   

3. Install Node dependencies

   npm install
   

4. Set up environment file

   cp .env.example .env
   php artisan key:generate
   

5. Configure database

   The default SQLite database is already configured in .env:

   DB_CONNECTION=sqlite
   DB_DATABASE=database/database.sqlite
   

   Create the database file:

   touch database/database.sqlite
   

6. Run migrations

   php artisan migrate
   

7. Install Python dependencies for Overpass

   cd overpass-ai
   pip3 install -r requirements.txt
   cd ..
   

   Note: The first run will download the sentence-transformers model (~80MB).

Configuration

AI Provider Setup

Configure your preferred LLM provider in .env:

For Anthropic Claude:

LOG_ANALYSIS_PROVIDER=anthropic
LOG_ANALYSIS_MODEL=claude-3-haiku-20240307
ANTHROPIC_API_KEY=your_api_key_here

For OpenAI:

LOG_ANALYSIS_PROVIDER=openai
LOG_ANALYSIS_MODEL=gpt-4-turbo
OPENAI_API_KEY=your_api_key_here

Analysis Parameters

Adjust AI analysis settings:

LOG_ANALYSIS_MAX_TOKENS=200
LOG_ANALYSIS_TEMPERATURE=0.3

Python Bridge Configuration

Overpass is pre-configured to use the overpass-ai directory. Ensure Python 3.9+ is active:

python3 --version  # Should show Python 3.9 or higher

Running the Application

Quick Start (Recommended)

Start all services at once with a single command:

composer run dev

This starts: web server, queue worker, log viewer (Pail), and Vite dev server.

Individual Services

Alternatively, run services separately:

1. Start the Development Server

php artisan serve

2. Start the Queue Worker In a separate terminal:

php artisan queue:work

3. Build Frontend Assets For development:

npm run dev

For production:

npm run build

4. MCP Server (Optional) The MCP server runs automatically with the web server. It's accessible at:

http://loganalysisai.test/mcp/log-watcher

To test it with MCP Inspector:

php artisan mcp:inspector

Usage

Web Dashboard

Visit http://loganalysisai.test (if using Laravel Herd) or `http://lo

... View full README on GitHub