AI API Application Skeleton

This is a comprehensive skeleton for AI API applications with support for multiple AI model types (REST and WebSocket), input/output testing interface, and modular architecture.

Features

• Multiple AI Model Support: Supports different AI models (General, GPT, Grok, Custom, etc.) with a pluggable architecture
• REST API Endpoints: Dedicated endpoints for each model type plus a generic endpoint
• WebSocket Support: Real-time communication for interactive AI applications
• Web Interface: Enhanced UI with input/output boxes for testing AI model responses
• Modular Architecture: Clean separation of concerns with model manager and configuration
• Configuration Management: Environment variables and configuration files

Architecture

• app.py: Main Flask application with SocketIO support
• models.py: AI model manager with abstract base classes
• config.py: Configuration management
• index.html: Enhanced web interface with tabs for REST, WebSocket, and configuration
• .env: Environment variables (API keys, etc.)
• requirements.txt: Python dependencies

API Endpoints

• GET / - Main UI interface
• POST /api/test - Generic AI model endpoint
• POST /api/general/test - General model endpoint
• POST /api/grok/test - Grok model endpoint (x-ai/grok-4-fast:free via OpenRouter)
• POST /api/gpt/test - GPT model endpoint
• POST /api/custom/test - Custom model endpoint
• GET /api/models - List available models
• GET /health - Health check endpoint

WebSocket Events

• connect - Client connected event
• disconnect - Client disconnected event
• test_request - Request to process AI model
• ping/pong - Connection validation

Setup

1. Install dependencies: pip install -r requirements.txt
2. Configure environment variables in .env
3. Run the application: python3 app.py
4. Access the UI at http://localhost:3000

UI Features

• Tabbed interface for REST API, WebSocket, and Configuration
• Model type selector for different AI models
• Parameter configuration (temperature, max tokens, etc.)
• Real-time chat interface for WebSocket communication
• Connection status indicators
• Input/output boxes for testing AI responses

Extending the Skeleton

To add a new AI model:

1. Create a new class extending AIModel in models.py
2. Add the model type to the AIModelManager
3. Optionally add a dedicated endpoint in app.py

Environment Variables

• PORT - Server port (default: 3000)
• GENERAL_API_KEY, GPT_API_KEY, GROK_API_KEY, CUSTOM_API_KEY - API keys for different models
• GENERAL_API_URL, GPT_API_URL, CUSTOM_API_URL - API endpoints for different models
• DEFAULT_MODEL - Default model to use
• MODEL_* parameters - Default parameter values# ai_dev_project-