Composable Prompting Workspaces for Creative Writing: Exploration and Iteration Using Dynamic Widgets

2025-03-27

AI Summary

Current generative AI GUIs lack support for iterative prompt exploration and fail to treat prompts as manipulable, first-class interface objects. To address this, we propose the “prompt-as-widget” paradigm, introducing a composable prompt canvas that encapsulates prompt elements as dynamic, interactive widgets—enabling spatial arrangement, real-time editing, modular composition, and immediate generative feedback. Our approach integrates GUI interaction design, generative AI orchestration, context-aware prompt suggestion, and adaptive widget layout. A comparative user study with 18 participants demonstrates that our system significantly outperforms conversational baselines on creativity support metrics. Users consistently rated it higher in perceived control, output value, and preference. This work advances structured, user-driven prompt engineering through interface-level abstractions, bridging the gap between high-level intent and low-level prompt manipulation in generative workflows.

Technology Category

• Humans and AI: Intelligent User Interfaces
• Natural Language Processing: Prompt Engineering / Prompting
• Cognitive Modeling & Cognitive Systems: Computational Creativity

Application Category

• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

Generative AI models offer many possibilities for text creation and transformation. Current graphical user interfaces (GUIs) for prompting them lack support for iterative exploration, as they do not represent prompts as actionable interface objects. We propose the concept of a composable prompting canvas for text exploration and iteration using dynamic widgets. Users generate widgets through system suggestions, prompting, or manually to capture task-relevant facets that affect the generated text. In a comparative study with a baseline (conversational UI), 18 participants worked on two writing tasks, creating diverse prompting environments with custom widgets and spatial layouts. They reported having more control over the generated text and preferred our system over the baseline. Our design significantly outperformed the baseline on the Creativity Support Index, and participants felt the results were worth the effort. This work highlights the need for GUIs that support user-driven customization and (re-)structuring to increase both the flexibility and efficiency of prompting.

Problem

• Lack of iterative exploration in current AI prompting GUIs
• Need for actionable interface objects for prompt representation
• Requirement for user-driven customization in text generation

Innovation

• Composable prompting canvas with dynamic widgets
• User-generated widgets for text exploration
• Customizable spatial layouts for prompting