MATLAB Development & AI Coding Projects

Supports Claude Code, GitHub Copilot, Cursor, OpenAI Codex, Sourcegraph Amp, and Gemini CLI.

• MATLAB MCP Server - Run MATLAB using AI applications with the official MCP server from MathWorks.
• MCP Framework for MATLAB Production Server - Publish MATLAB functions to MATLAB Production Server as MCP tools.
• MATLAB MCP HTTP Client - MCP client in pure MATLAB code for calling external tools in LLM workflows.

• Large Language Models (LLMs) with MATLAB - Connect MATLAB to LLM APIs, including OpenAI Chat Completions, Azure OpenAI Services, and Ollama.
• MATLAB Local Agent Quickstart - Set up a local AI coding agent for MATLAB on your own hardware, with nothing leaving your machine.

• Agent Skills Playground - A sandbox for prototyping and demonstrating Agent Skills for MATLAB and Simulink work. Includes standalone skills and bundled demos with tutorials.
• MATLAB AI Coding Rules - Coding rules and guidelines for AI coding assistants.
• AI Coding Agent Prompts for MATLAB - Prompts for MATLAB code generation and analysis.
• MATLAB Slash Commands - Custom commands for AI-powered development workflows.

• Terminal in MATLAB - Run a terminal inside MATLAB to use command-line tools such as AI coding agents, git, and docker without leaving the MATLAB desktop.
• MATLAB AI Agent SDK - Build and run AI agents in MATLAB. Connect to OpenAI or Ollama, use MATLAB functions and toolbox workflows as tools, and maintain conversational state across turns.
• Virtual Lab for MATLAB - Run structured multi-agent research meetings in MATLAB, where specialist agents with their own roles and models take turns in one shared conversation. Includes an interactive app, workflow presets, and saved transcripts, and makes model calls through MATLAB AI Agent SDK.
• xToSim - Agent skill that turns research papers, system descriptions, equations, and block diagrams into validated, hierarchical Simulink models. Delivers the model with an HTML report and standalone test scripts.
• VLA-Guided Mobile Manipulation - A Simulink model of a mobile manipulator that turns one plain-English instruction into a drive, a grasp, and a place in an Unreal Engine apartment. Gemini Robotics-ER grounds the instruction in four camera views, and four Stateflow supervisors sequence the task.
• Soft Gripper Fruit Picking - A UR5e arm with a soft pneumatic gripper picks fruit, modeled physically in Simscape and validated against published hardware data. An AI agent built the project by following a written Model-Based Design guide, which ships with the example.

• MATLAB Extension for Visual Studio Code - Syntax highlighting, code analysis, navigation, and debugging for MATLAB in VS Code.
• MATLAB Coding Guidelines - Recommended coding standards with Code Analyzer integration for consistency across teams.
• MATLAB Toolbox Best Practices - File organization, testing, and release practices for sharing MATLAB toolboxes.

Recent posts on AI coding with MATLAB and Simulink.

• Nov 12, 2026 - Agentic AI for Climate Data in MATLAB
• Nov 6, 2026 - AI エージェントにエンジニアの実践知を伝える (JP)
• Nov 5, 2026 - Practice Agentic Coding with the MATLAB One-Button Game Mini Hack
• Nov 3, 2026 - Engineering Embedded AI with MATLAB Agentic Toolkit
• Nov 2, 2026 - MATLAB EXPO 2026: Agentic Engineering in Practice
• Nov 2, 2026 - AI x Simulink 協調設計：40 時間でコンバータの回路・制御を設計してみた ～これからのシステム設計におけるAIとの付き合い方とは？～ (JP)
• Oct 30, 2026 - Behavioral Models with Agentic-AI-powered MBSE
• Oct 27, 2026 - Teaching AI Agents How Engineers Work