prompt_gen

AI Prompt Generator

Project Overview

This project is an AI Prompt Generator that creates prompts for Midjourney v6 image generation and Udio music generation using the Anthropic API. It includes a command-line interface for generating prompts and a Flask web application for viewing the outputs.

Project Structure

• app.py: Flask web application for viewing generated prompts
• main.py: Command-line interface for generating prompts
• prompts/: Directory containing prompt templates
  • midjourney.txt: Template for Midjourney v6 prompts
  • udio.txt: Template for Udio music prompts
• outputs/: Directory where generated prompts are stored
• templates/: HTML templates for the web application

Setup and Usage

1. Install required dependencies:

   pip install flask anthropic termcolor python-dotenv
   

2. Set up your Anthropic API key in a .env file:

   ANTHROPIC_API_KEY=your_api_key_here
   

3. Run the prompt generator:

   python main.py
   

4. Run the Flask web application:

   python app.py
   

   Access the web interface at http://127.0.0.1:5000

Features

Prompt Generation (main.py)

• Supports Midjourney v6 and Udio prompt generation
• Uses Anthropic's Claude 3.5 Sonnet model
• Saves generated prompts to files with timestamps

Web Interface (app.py)

• Displays generated prompts in a responsive layout
• Supports pagination for large numbers of prompts
• Parses and formats XML output from the prompt generator

Prompt Templates

Midjourney v6 (midjourney.txt)

• Generates 5 diverse image prompts
• Incorporates Midjourney v6 best practices and parameters
• Includes explanations for each generated prompt

Key parameters:

• Aspect ratio (--ar)
• Chaos (--chaos)
• Quality (--quality)
• Seed (--seed)
• Stylize (--stylize)
• Weird (--weird)
• Tile (--tile)
• No (negative prompting)

Udio Music (udio.txt)

• Generates 5 diverse music prompts
• Focuses on genre, mood, instruments, tempo, and lyrical themes
• Utilizes Udio-specific features and parameters

Key features:

• Custom lyrics
• Instrumental mode
• Manual mode
• Extension mode
• Remixing
• Inpainting

Development

• The project uses Flask for the web framework
• Jinja2 is used for HTML templating
• CSS is used for responsive design
• JavaScript handles dynamic content loading and pagination

Contributing

Contributions are welcome! Please submit pull requests or open issues for any bugs or feature requests.