NVIDIA Opens Physical AI Toolset to Coding Agents Across Robotics and Manufacturing

What happened:  NVIDIA released an open-source collection of physical AI skills and tools. The release spans Omniverse, Cosmos, Alpamayo, and Metropolis. It targets robotics, autonomous vehicles, vision AI, and industrial digital twins.

Why it matters: The skills convert physical AI training, evaluation, and deployment steps into repeatable, agent-executable instructions. Agents can call NVIDIA libraries, models, and frameworks directly. The tools are available on GitHub and skills.sh for use with any coding agent.

Industry context:  Industrial software firms including Siemens, Cadence, Dassault Systèmes, and Synopsys are applying the libraries to engineering data, simulation, and digital twins, whereas several manufacturers have reported measured results – some have cut model training and deployment time and some have improved defect detection.

Our take: The release furthers agentic automation into hardware-centric workflows, with early adoption concentrated in electronics manufacturing.