World Craft: Agentic Framework to Create Visualizable Worlds via Text

TL;DR

World Craft is a framework that enables users to create visualizable, interactive worlds like AI Town through natural language descriptions, combining structured scaffolding and multi-agent analysis to improve customization and stability.

Contribution

The paper introduces a novel agentic framework with modules for structured scene development and intent analysis, enhancing user-driven environment creation without programming skills.

Findings

Outperforms existing code agents and LLMs in scene construction

Enhances spatial knowledge and layout stability

Provides scalable democratization of environment creation

Abstract

Large Language Models (LLMs) motivate generative agent simulation (e.g., AI Town) to create a ``dynamic world'', holding immense value across entertainment and research. However, for non-experts, especially those without programming skills, it isn't easy to customize a visualizable environment by themselves. In this paper, we introduce World Craft, an agentic world creation framework to create an executable and visualizable AI Town via user textual descriptions. It consists of two main modules, World Scaffold and World Guild. World Scaffold is a structured and concise standardization to develop interactive game scenes, serving as an efficient scaffolding for LLMs to customize an executable AI Town-like environment. World Guild is a multi-agent framework to progressively analyze users' intents from rough descriptions, and synthesizes required structured contents (\eg environment layout and assets) for World Scaffold . Moreover, we construct a high-quality error-correction dataset via reverse engineering to enhance spatial knowledge and improve the stability and controllability of layout generation, while reporting multi-dimensional evaluation metrics for further analysis. Extensive experiments demonstrate that our framework significantly outperforms existing commercial code agents (Cursor and Antigravity) and LLMs (Qwen3 and Gemini-3-Pro). in scene construction and narrative intent conveyance, providing a scalable solution for the democratization of environment creation.