Clarify Before You Draw: Proactive Agents for Robust Text-to-CAD Generation

TL;DR

ProCAD introduces a proactive agentic framework for text-to-CAD generation that improves robustness by resolving specification ambiguities through targeted clarifications before code synthesis.

Contribution

It presents a novel proactive clarification approach with a dedicated agent that enhances the accuracy and reliability of text-to-CAD systems, outperforming existing models.

Findings

Significantly reduces Chamfer distance by 79.9%

Lowers invalidity ratio from 4.8% to 0.9%

Improves robustness to ambiguous prompts

Abstract

Large language models have recently enabled text-to-CAD systems that synthesize parametric CAD programs (e.g., CadQuery) from natural language prompts. In practice, however, geometric descriptions can be under-specified or internally inconsistent: critical dimensions may be missing and constraints may conflict. Existing fine-tuned models tend to reactively follow user instructions and hallucinate dimensions when the text is ambiguous. To address this, we propose a proactive agentic framework for text-to-CadQuery generation, named ProCAD, that resolves specification issues before code synthesis. Our framework pairs a proactive clarifying agent, which audits the prompt and asks targeted clarification questions only when necessary to produce a self-consistent specification, with a CAD coding agent that translates the specification into an executable CadQuery program. We fine-tune the coding agent on a curated high-quality text-to-CadQuery dataset and train the clarifying agent via agentic SFT on clarification trajectories. Experiments show that proactive clarification significantly improves robustness to ambiguous prompts while keeping interaction overhead low. ProCAD outperforms frontier closed-source models, including Claude Sonnet 4.5, reducing the mean Chamfer distance by 79.9 percent and lowering the invalidity ratio from 4.8 percent to 0.9 percent. Our code and datasets will be made publicly available.