MathWorks announced new capabilities that let AI agents run and refine MATLAB workflows through two new open-source tools – MATLAB MCP Server and MATLAB Agentic Toolkit – using Model Context Protocol. With these tools, agents can write MATLAB code, run it in a live session, examine outputs or errors, and iterate toward a correct result, while engineers remain responsible for validating results and applying their expertise.
These new capabilities are intended for MATLAB users, applied AI engineers building agent-driven workflows, and platform teams managing AI-assisted engineering environments. By running code directly in MATLAB, agents base their reasoning on deterministic computation, numerical analysis, and executable models. This shifts AI from probabilistic reasoning to execution-based results, reinforcing engineering rigor. Engineers can verify results by reviewing outputs, comparing them to expected behaviour, and refining workflows through iteration – core practices in engineering development.
“As organisations adopt agentic AI in Model-Based Design and engineering, the focus is shifting from code generation to reliable execution within established multi-disciplinary toolchains,” said Diego Tamburini, AI Practice Director at CIMdata. “Engineers remain responsible for defining problems, validating outcomes, and maintaining oversight, while AI agents increasingly handle iterative and repetitive tasks – augmenting human efficiency and effectiveness. This reinforces the importance of human-in-the-loop workflows, where real execution and validation underpin trust in AI-driven engineering processes.”
MATLAB MCP Server and MATLAB Agentic Toolkit are open-source packages, allowing developers and organisations to inspect, extend, and incorporate agent-based workflows with MATLAB in their own environments using agentic tools such as Claude Code, GitHub Copilot, OpenAI Codex, Gemini CLI, and more. This supports interoperability across diverse AI agent frameworks and positions MATLAB as foundational infrastructure for emerging agent-based engineering ecosystems.
“AI agents are most effective in engineering when they can directly interact with the tools used for design, simulation, and analysis,” said Seth DeLand, Generative AI Product Manager at MathWorks. “By enabling agents to execute and iterate MATLAB workflows, we’re connecting AI-driven iteration to the same computational environment engineers use to develop and validate their work. This allows teams to move from LLM generated code to executable, testable results within a consistent engineering framework.”
