←Leaderboard/Science/simulink-agentic-toolkit
Last commit on September 30, 2026·Created on April 15, 2026

matlab/simulink-agentic-toolkit

“Automates the configuration and authoring of Simulink models for embedded code generation”
Combined rank
#298
across all skills
In Science
#7
category rank
Stars
1.2k
+2.9% in last 7d
Forks
108
+0.9% in last 7d
Watchers
33
+3.1% in last 7d
Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
TL;DR

This toolkit provides a set of capabilities for embedded software engineers to manage the transition from Simulink models to deployable code. It streamlines the technical requirements for generating C and C++ code, ensuring that model elements are correctly represented in the final output.

The skills focus on precise control over code interface configurations and the authoring of Classic AUTOSAR Software Components, making the process of preparing models for embedded targets more structured and verifiable.

WHO IT'S FOR
Embedded software engineers
automate Simulink model configuration for code generation
Automotive systems engineers
author and manage Classic AUTOSAR SWCs
Control systems architects
optimize Simulink models for memory and precision
Calibration engineers
automate the customization of A2L files
Repository contents

36 skill files

Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.

Use this skill collection

Follow the documented setup, then try a first task.

I’m using

Automated setup through MATLAB Add-On Manager

Installs: Complete Simulink Agentic Toolkit installation and skill registration; the automated installer supports both the MATLAB and Simulink Agentic Toolkits. The listed skill paths are a bounded selection relevant to the starter tasks, not the bundle's complete inventory. · Claude Code, Copilot, Codex, Amp, Gemini CLI

Before you start
  • MATLAB R2023a or later with Simulink.
  • A supported AI coding agent.
  • Some skills require additional products; check each skill's manifest.yaml requires-products field.
  • Existing users of the previous agent-driven workflow must first uninstall that custom installation and global configuration following the repository's migration guide.
  1. Download agenticToolkitInstaller.mltbx from the latest Simulink Agentic Toolkit release using the documented release link.

  2. Open the downloaded agenticToolkitInstaller.mltbx file to install the installer add-on through MATLAB Add-On Manager.

  3. After the add-on is installed, enter this command in MATLAB.

    setupAgenticToolkit("install")
README.md ↗ · Checked Oct 2, 2026.claude-plugin/marketplace.json ↗ · Checked Oct 2, 2026

Give it something to do.

Suggested first task

Find a shipped Simulink example

Uses finding-simulink-examples

  • A search topic to replace <topic>, such as motor control, PID, or battery.
  • The owning MathWorks product must be installed only if you later choose to open a returned example.
Use the finding-simulink-examples skill to search the shipped MathWorks examples database for five examples about <topic>. Do not open anything. Return each match's name, open command verbatim, and a concise description.
Suggested first task

Run a non-destructive configured simulation

Uses simulating-simulink-models

  • An accessible Simulink model name to replace <model-name>.
  • A stop-time value to replace <stop-time>.
  • MATLAB R2023a or later with Simulink.
Use the simulating-simulink-models skill to run <model-name> non-destructively with StopTime set to <stop-time> through Simulink.SimulationInput. Report which output properties are available, including any logged signal names, without modifying the .slx file.
Suggested first task

Draft a standalone bug report

Uses filing-bug-reports

  • A defect description to replace <defect>.
  • Conversation context or reproduction details showing what happened and what was expected.
  • Any exact error messages or relevant tool output.
Use the filing-bug-reports skill to create a standalone Markdown bug report for <defect>. Include the action and intent, actual behavior and exact errors, expected behavior, reproduction details, and available environment information. Do not prescribe a root cause or solution.