PTC Brings Natural Language AI to CAD Automation via Onshape

PTC Brings Natural Language AI to CAD Automation via Onshape

PTC has made its FeatureScript MCP Server available within Onshape, enabling engineers to create custom CAD features through natural language instructions rather than manual coding. The capability, released through PTC's Onshape Labs early-access program, connects large language models directly to Onshape's CAD automation framework.

Engineers can describe the design functionality they need in plain English, and AI systems including Claude, ChatGPT, and Gemini can generate, test, debug, and refine the corresponding FeatureScript code. FeatureScript is Onshape's proprietary programming language for CAD automation, used to capture engineering knowledge and automate repetitive design tasks.

Why it matters

This approach shifts AI's role in CAD from generating one-off models to creating reusable engineering tools. Once developed, these custom features can be shared across teams and projects, transforming established engineering practices into repeatable design automation. The capability also lowers the barrier for engineers who understand design requirements but may not routinely write FeatureScript code themselves.

Built on the Model Context Protocol

The FeatureScript MCP Server uses the Model Context Protocol to establish direct connections between coding-capable LLMs and Onshape's FeatureScript environment. Rather than limiting AI to individual design tasks, this architecture enables the creation of persistent, reusable CAD capabilities that organizations can deploy across multiple workflows.

PTC positions this as a text-to-code-to-CAD pipeline that makes CAD customization more accessible while maintaining the precision and reusability that engineering teams require. However, AI-generated outputs still require engineering review to ensure designs meet functional, manufacturing, safety, and regulatory standards.

Part of Broader AI Expansion

The FeatureScript MCP Server represents one component of PTC's wider AI integration across its product development portfolio. In July 2026, PTC launched Onshape Labs as a testing environment for emerging AI capabilities, initially outlining plans for AI agents, drawing validation, AI-assisted rendering, and robotics simulation alongside the FeatureScript capability.

PTC has also expanded its Onshape AI Advisor with real-time design guidance and agent-based workflows, and introduced Creo AI Assistant in its Creo 13 and Creo+ 13.3 releases earlier this year to provide embedded guidance within traditional CAD environments.

Engineering Context as Foundation

The capability aligns with PTC's Intelligent Product Lifecycle strategy, which emphasizes maintaining consistent product data across engineering and enterprise functions. Onshape's cloud-native architecture combines CAD and product data management, preserving product history, design relationships, and versioning information that can inform AI-assisted workflows.

By connecting LLMs directly with FeatureScript, PTC extends AI capabilities beyond design assistance toward the creation of reusable engineering automation. The value lies not just in generating immediate design outputs, but in capturing engineering practices within tools that can be applied to future projects.

These details were first reported by ARC Advisory Group.