SDXL Prompt Styler Node Documentation

Overview

The SDXL Prompt Styler is a custom node developed for use with ComfyUI. It is designed to style text prompts based on predefined templates stored across multiple JSON files. This node provides flexibility and creativity by efficiently managing both positive and negative prompts in text-based workflows.

Features

• Template-Based Styling: Allows users to select from predefined styles to enhance the textual content of prompts.
• Flexible Negative Prompt Management: Users can specify negative prompts in conjunction with or in place of those defined in the templates, enabling diverse expression.
• Toggle Styling Options: Users can disable styling for positive, negative, or both sentiments, offering bypass mode for specific scenarios.
• Advanced Styling Capabilities: The accompanying SDXL Prompt Styler Advanced node offers additional control over the formatting and combination of prompts.

Primary Use Cases

• Generating Complex Text Prompts: Ideal for creative workflows that require rich, detailed, and versatile text prompts.
• Customizing AI Model Inputs: Enhances the inputs fed into AI models by stylizing the structure and content to achieve specific outcomes.
• Managing Dual Sentiments: Facilitates a structured approach to handle both positive and negative sentiments within text prompts.

Inputs

• text_positive: This input receives the base text for the positive prompt that will be styled based on the selected template.
• text_negative: This input is intended for the negative base prompt. If no negative prompt is provided, the default from the template is used.
• style: Users select a style from the available list of templates, which dictates the styling process.
• log_prompt: A boolean input that controls whether the node's actions (inputs and outputs) are logged to the console.
• style_positive: Toggle input allowing the user to enable or disable styling for positive prompts.
• style_negative: Toggle input allowing the user to enable or disable styling for negative prompts.

Outputs

• text_positive: The output is the styled positive text prompt, which incorporates the base text and the desired stylistic elements from the selected template.
• text_negative: Similar to the positive prompt, this output represents the styled negative text prompt.

Usage in ComfyUI Workflows

In ComfyUI, the SDXL Prompt Styler node can be integrated into workflows that require dynamically styled text prompts. By selecting appropriate templates, users can generate prompts that cater to specific scenarios, aiding in scenarios like natural language generation, sentiment analysis experiments, and more. Incorporating this node helps transform basic input text into detailed, expressive, and contextually relevant prompts by leveraging predefined styles.

Special Features and Considerations

• Style Selection from JSON: Styles are dynamically loaded from JSON files located in a specified directory, and the node ensures the uniqueness of style names.
• Error Handling: Robust error handling is implemented to manage file reading and data validation efficiently.
• Updates and Versatility: The node's structure allows for easy updates and integration with additional templates, allowing ongoing enhancement of text styling capabilities.
• Advanced Node: For more intricate workflows, the SDXL Prompt Styler Advanced node provides additional options like manipulating specific parts of the prompt and deciding negative prompt split behavior. This makes it suitable for users seeking more granular control over their styled prompts.