PlayGenie - AI-Powered Ansible Automation Platform

An intelligent automation platform that uses AI to generate, validate, and execute Ansible playbooks through a natural language interface.

Overview

PlayGenie combines the power of Large Language Models (LLM) with Ansible automation to make infrastructure management as simple as having a conversation. Describe what you want to automate, and PlayGenie will generate, validate, and help you execute Ansible playbooks.

Features

• AI-Powered Playbook Generation: Describe tasks in natural language, get Ansible playbooks
• Intelligent Validation: 5-step validation layer with automatic error fixing (3 retries)
• Code Editor: Professional Monaco editor for YAML editing
• Real-time Validation: Automatic retry with LLM-based error correction
• Chat Interface: Ask Ansible questions and get expert guidance
• Quick Execution: Save playbooks for immediate execution

Architecture

The system consists of:

• Frontend: React application with Monaco code editor
• Backend: FastAPI server with LLM integration
• LLM: Ollama (local LLM) for AI-powered generation
• Validation: 5-step validation pipeline with retry mechanism

Project Structure

PlayGenie/
├── client/              # React frontend application
├── server/              # FastAPI backend application
├── ARCHITECTURE.md      # Detailed system architecture
├── QUICK_START.md       # Setup and running instructions
└── README.md           # This file

Quick Start

Prerequisites

• Node.js 16+
• Python 3.9+
• Ollama installed and running
• Ansible (optional, for execution)

Setup

1. Install Ollama and pull a model:

   ollama pull llama2

2. Setup Backend:

   cd server
   source ansible/bin/activate
   pip install -r requirements.txt
   # Create .env file with MODEL_NAME=llama2
   uvicorn main:app --reload --host 127.0.0.1 --port 8000

3. Setup Frontend:

   cd client
   npm install
   npm run dev

4. Access: Open http://localhost:3000 in your browser

For detailed instructions, see QUICK_START.md

Documentation

• ARCHITECTURE.md: Complete system architecture and design
• QUICK_START.md: Step-by-step setup guide
• server/README.md: Backend documentation
• client/README.md: Frontend documentation
• server/VALIDATION.md: Validation layer details
• server/VALIDATION_EXPLAINED.md: 5-step validation explained

Technology Stack

Frontend

• React 18
• Vite
• Tailwind CSS
• Monaco Editor
• Lucide React

Backend

• FastAPI
• Python 3.9+
• Ollama (LLM)
• PyYAML
• Ansible Runner

Core Workflow

1. User describes task → Frontend sends to /api/chat
2. LLM generates playbook → Using system prompts
3. Validation layer → 5-step validation with 3-retry mechanism
4. Playbook displayed → In Monaco editor for editing
5. User executes → Saves to ansible/current-playbook.yml

Security

• Local LLM processing (no external API calls)
• Environment variables for sensitive data
• Input validation on all endpoints
• CORS protection

License

Built for Ansible automation enthusiasts

Contributing

This is a specialized Ansible automation assistant. Contributions welcome!

---

PlayGenie - Making infrastructure automation as simple as having a conversation!