Quickstart

The promise: install the toolkit, scaffold a course, describe what you want to your AI assistant, and package the result for your LMS.

Beta

OELTKit is pre-v1.0 and the packages aren’t on npm yet. Install from the main repository for now; the install command below is what you’ll use once @oeltkit/cli publishes with v0.1.

1. Install

npm install -g @oeltkit/cli

This gives you the oelt command. See the CLI reference for the full command surface.

2. Scaffold

oelt new my-course --title "Data Privacy Essentials"

This creates a course.json manifest and a first page at pages/p1.html. The manifest is the source of truth — it declares your modules, pages, completion rules, and score weighting. See the authoring guide for its shape.

3. Chat

Connect OELTKit to your assistant (MCP setup) and describe the course:

Build me a 15-minute course on GDPR basics with a scored quiz…

Your assistant writes the content layer as plain HTML and drops in interaction components (<oelt-mcq>, <oelt-quiz>, <oelt-branching>, …) where they belong. Tracking and accessibility come from the toolkit, not hand-wiring.

Validate as you go — findings are machine-readable, so the assistant fixes its own mistakes:

oelt validate my-course

4. Preview

oelt preview my-course

Opens the course in a local fake-LMS harness with a live tracking inspector, so you can see completion and score reporting before you ship.

5. Package

oelt package my-course --target scorm12

Import the resulting zip into your LMS. Other targets: scorm2004, cmi5, web. If you’re unsure which your LMS supports, start with SCORM 1.2 — see Standards for how to choose.