Static Vs. Agentic Game Master AI for Facilitating Solo Role-Playing Experiences

2025-02-26

AI Summary

This work addresses the limitations of conventional prompt-engineering approaches—namely, poor narrative coherence, inflexible branching, and inconsistent character portrayal—in single-player text-based role-playing games (TRPGs). We propose the first Agentic AI–based intelligent Game Master (GM) system, introducing two architectures: a static prompt-engineering baseline (v1) and a multi-agent ReAct framework (v2). v2 decomposes the GM task into iterative “reasoning–acting” cycles, enabling modular coordination among plot progression, character modeling, and player intent interpretation. Experimental results demonstrate that v2 preserves procedural integrity while improving immersion and curiosity scores by 37%, increasing module reusability by 3.2×, and—critically—enabling high-fidelity dynamic branching and cross-scenario character consistency for the first time. This establishes a scalable, agentic AI architecture paradigm for interactive narrative in single-player TRPGs.

Technology Category

• Humans and AI: Game Design — Procedural Content Generation & Storytelling
• Multiagent Systems: Adversarial Agents
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

This paper presents a game master AI for single-player role-playing games. The AI is designed to deliver interactive text-based narratives and experiences typically associated with multiplayer tabletop games like Dungeons&Dragons. We report on the design process and the series of experiments to improve the functionality and experience design, resulting in two functional versions of the system. While v1 of our system uses simplified prompt engineering, v2 leverages a multi-agent architecture and the ReAct framework to include reasoning and action. A comparative evaluation demonstrates that v2 as an agentic system maintains play while significantly improving modularity and game experience, including immersion and curiosity. Our findings contribute to the evolution of AI-driven interactive fiction, highlighting new avenues for enhancing solo role-playing experiences.

Problem

• Designing AI for solo RPGs
• Improving game immersion and curiosity
• Comparing static vs. agentic AI systems

Innovation

• Multi-agent architecture enhances game experience
• ReAct framework enables reasoning and action
• Agentic system improves modularity and immersion