skills.rest
matlab
matlabOFFICIAL
Explore all skills by this author →
WebsiteGitHub

matlab-package-toolbox

Packages raw MATLAB code into a published .mltbx toolbox through an eight-phase checkpointed pipeline.

995|122|Updated Apr 3, 2026
Search Tags:#dependency analysis#packaging#release#build pipeline#toolbox#matlab#mltbx
One-click install
npx skills add https://github.com/matlab/matlab-agentic-toolkit --skill matlab-package-toolbox
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: matlab-package-toolbox
Source: https://github.com/matlab/matlab-agentic-toolkit/tree/main/skills-catalog/matlab-software-development/matlab-package-toolbox
Command: npx skills add https://github.com/matlab/matlab-agentic-toolkit --skill matlab-package-toolbox

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Turning a folder of MATLAB code into a distributable .mltbx toolbox involves many error-prone steps: defining the public API, tracing dependencies, creating a project, writing documentation, building, and publishing. This Skill orchestrates that entire pipeline with human checkpoints so nothing ships without review.

Core Features & Use Cases

  • Eight-Phase Pipeline: Drives scope definition, dependency analysis, project creation, documentation, build planning, readiness assessment, build, and publish in strict order.
  • Human Checkpoints: Pauses for user approval at each phase boundary, producing reviewable artifacts like toolboxSpecification.m, tbxManifest.m, and buildfile.m.
  • Readiness Assessment: Runs up to 16 ordered quality checks (H1 help, tests, coverage, code issues, README, license) with Quick, Standard, and Deep levels.
  • Use Case: You have a folder of MATLAB functions for linear algebra utilities. Invoke the Skill to classify the public API, resolve external dependencies, generate documentation and examples, build the .mltbx, and publish version 1.0 to a release folder.

Quick Start

Ask your AI agent to package the folder containing your MATLAB code as a toolbox and state its purpose, for example by saying to package the mylib folder as a toolbox for signal processing.

Frequently Asked Questions about matlab-package-toolbox

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I package MATLAB code into a .mltbx toolbox?▼

Provide the folder path and the toolbox purpose, then approve the pipeline plan. The Skill walks through eight phases from API definition to publishing, producing a versioned .mltbx in the release folder after you approve each checkpoint.

How to analyze MATLAB toolbox dependencies before packaging?▼

Phase 2 uses matlab.codetools.requiredFilesAndProducts and ToolboxOptions to trace transitive file and product dependencies, detect unresolved symbols, and flag ignore-file conflicts. Results are saved as buildUtilities/tbxManifest.m for your review.

What MATLAB version is required to build a .mltbx toolbox?▼

This Skill requires MATLAB R2026a or later, plus an AI agent connected to the MATLAB MCP server. It uses matlab.addons.toolbox.packageToolbox and the matlab.buildtool framework for building and packaging.

Can I skip documentation when packaging a MATLAB toolbox?▼

Yes, state that you want to skip documentation when the pipeline plan is presented and Phase 4 will be marked as skipped. Phase 3 is also skipped automatically if a MATLAB project with a .prj file already exists.

When should I not use the MATLAB toolbox packaging pipeline?▼

Do not use it to compile standalone apps or web apps, which require MATLAB Compiler or Compiler SDK. It is also unnecessary for distributing a single script or function without toolbox structure.

Why does the packaging pipeline stop and ask for confirmation?▼

Checkpoints are the core contract of the pipeline: each phase produces an artifact such as the interface spec or dependency manifest that you must approve before work continues. This prevents unwanted files from being written or packaged without review.

Related Skills

Explore More Software Engineering Skills→
matlabmatlab
883

matlab-list-products

List installed MATLAB products and support packages for a given installation folder or release name.

Official
Intermediate
agoodwayagoodway
1

publish

Validate project readiness and confirm version before running the just publish workflow.

Official
Intermediate
mtschoenmtschoen

maintaining-full-coverage

Enforce 100% test coverage and commit TEST-REPORT.md with git hash.

Community
Advanced
matlabmatlab
883

matlab-agentic-toolkit-setup

Install and configure the MATLAB Agentic Toolkit with MCP server registration.

Official
Advanced