Editor’s note: This article is Part 2 of a three-part series on AI-assisted course design in D2L Brightspace. Part 1 examined the gap between generic AI outputs and the realities of building courses inside Brightspace. Here, Kumar introduces the Brightspace Interactive Design Skill Suite designed to bridge that gap. Part 3 provides a practical walkthrough showing how the suite is installed, configured and used in a real course build.
In Part 1 of this series, I argued that generic AI falls short in Brightspace not because it lacks capability, but because it lacks context. Used well, with an instructor leading, it can be a capable collaborator for course design. The gap is not knowledge (a well-prompted AI can find its way around Brightspace). The gap is structure and platform context. Generic AI waits to be told what to do next. The suite comes pre-configured for the workflow: the right questions, the right order and the right dependencies, without the instructor having to know them in advance.
I closed with a question: can AI close that gap at scale, throughout a course lifecycle, for instructors at every experience level, without requiring them to already know the answers?
The Brightspace Interactive Design Skill Suite is my answer to that question.
The Architecture: One Orchestrator, 40 Specialists
An analogy makes the structure clear.
When you hire a general contractor to renovate a house, you don’t describe the renovation to the electrician, the plumber and the drywaller separately. You describe it once to the contractor, who translates your vision into a coordinated sequence: electrician before drywaller, plumber before tile, foundation before framing. The contractor doesn’t do all the work. What they do is know the order, catch the dependencies and make sure nothing gets built on top of something that isn’t ready yet.
The suite applies the same coordination idea. There is one Brightspace Orchestrator that listens to what an instructor needs, classifies their intent, checks for prerequisites and conflicts, and routes to the right specialist skill. The Brightspace specialist skills don’t need to understand the full course design picture. They know their domain: the Gradebook Planner knows Brightspace grade categories, weights and calculation methods. The Release Condition Planner knows how unlock logic works. The Content Currency Auditor knows what field-specific content ages well and what doesn’t. Each does one thing well. The orchestrator knows how they fit together.
The structural difference between a general AI and a skilled one isn’t more capability. It’s better workflow fit.
The 40 Skills
The suite supports work across the course design lifecycle in seven categories:
| Category | What it covers |
| Inception (4) | Course outline builder, learning outcomes generator, course calendar, program alignment mapper. The outline is saved once and read by every subsequent skill. |
| Planning (4) | Gradebook planner, redesign planner, release condition planner, competency mapper. These set up the architecture before any content or assessment is created. |
| Content (11) | HTML topic builder, slide converter, PDF transformer, dual format course builder, glossary, case study, reading list, video script writer, and a full AI video production pipeline. |
| Assessment (9) | Quiz generator, assignment generator, discussion generator, rubric builder, survey generator, group project setup, peer evaluation with grade passback, self-assessment, and exam wrappers. |
| Quality (4) | Accessibility auditor (WCAG 2.1 AA), course copy auditor, content currency auditor, and student experience preview. All run before content goes live. |
| Communication (3) | Announcement writer in Brightspace-safe HTML, email template builder, and intelligent agent setup for automated at-risk student outreach. |
| Management (5) | End-of-term workflow, course analytics guide, LTI integration guide, checklist builder, and the session debrief skill that maintains a running project status document. |
What makes this more than a collection of tools is the orchestrator’s awareness of how they connect. The Gradebook Planner must run before any assessment skill. The Redesign Planner must run before the Release Condition Planner. These dependencies are checked and flagged automatically, whether or not the instructor mentions them. For anyone familiar with Brightspace, one question will already be forming: if D2L already has AI built into the platform, what does this add?
A Word on D2L Lumi
That question deserves a direct answer.
D2L Lumi is a strong product and a D2L Brightspace add-on available to institutions that have enabled it. D2L organizes it into three audience categories. For educators, it streamlines workflows: generating quiz questions, surfacing assignment and discussion ideas, aligning learning outcomes to course activities, and supporting early identification of at-risk learners. For course authors, it accelerates content creation: generating practice questions, building interactive elements, and applying Bloom’s Taxonomy automatically. For learners, it supports achievement: personalized study recommendations after quizzes, a support chatbot, and low-stakes practice questions.
The suite and D2L Lumi work at different stages, and that distinction matters.
D2L Lumi works inside Brightspace tools. Open the Quiz tool and D2L Lumi helps you generate questions faster. Open Assignments and D2L Lumi suggests an activity idea. The intelligence is embedded in the individual tool workflow, which is exactly where it should be for those tasks.
The suite works before those tools are opened. It answers the question that comes first: what should I build, in what order, and what do I need to set up before anything else? When an instructor describes a full course redesign in a single message, the orchestrator maps the dependencies, flags the prerequisites, and produces a sequenced path across multiple tools before a single one is opened. It can also help structure work that begins outside Brightspace before it is brought into the platform.
They are not competing. D2L Lumi makes individual Brightspace tools faster to use. The suite makes the full course design process coherent before that work begins. The same principle applies to Creator+ and Createspace: the skills can help prepare and structure work, while Brightspace remains the environment where content is authored, managed, reused and deployed.
How the Suite Is Delivered
The suite runs as 40 specialist skills plus one orchestrator, all attached to a Claude account. Claude is Anthropic’s AI assistant. Each skill is a pre-configured instruction set that shapes how Claude behaves for that specific task: ask the right questions, check the right dependencies, produce the right outputs for Brightspace. The skills are available in any Claude chat, inside or outside a Project.
The recommended practice is to pair the suite with a course-specific Claude Project, one per course. The Project holds the course knowledge files: the outline, the outcomes, the calendar, the gradebook plan. It does not contain the skills. What it does is give them something to work with. Every session starts by reading those files, so the instructor never re-explains their course context and every specialist skill operates with the full picture of what has already been built. The Project also keeps token use focused and conversation history organized, practical benefits that matter across a full course design lifecycle.
The Orchestrator in Action
The best way to understand what the suite does is to watch it work.
Here is a real exchange. For context, it’s week 4, mid-course, and 3 students have gone quiet:
Faculty: /brightspace-orchestrator Three of my students haven’t logged in for 10 days. It’s Week 4 and one already missed the first assignment. I’m worried. How do I check on them properly and make sure this doesn’t slip through the cracks again?
Orchestrator: This is a clear at-risk student support situation. Before I route you to the right skills, a few quick questions:
- Which course is this? (Course code + delivery mode — online async, in-person or blended)
- Do you already have any Intelligent Agents set up in this course? Or is this the first time you’re addressing this kind of situation systematically?
- What do you want to happen right now vs. going forward?
- Right now: reach out to these three students personally (email templates)
- Going forward: set up automated monitoring so you catch this earlier next time (Intelligent Agents + analytics)
- Both
- Does your institution have a student support referral process you want the outreach to connect to — for example, academic advising or student success services — or are you handling it directly as the instructor for now?
This will help me route you to exactly the right combination of skills and not build anything you don’t need.
Faculty: SOCI 2210 — blended. First time doing this systematically. Both — I want to reach out now and set up monitoring going forward. I’ll handle it directly for now.
That’s one scenario. There are four more: the instructor starting a full course redesign from a syllabus and slides, the rebuilder with a four-year-old course, the first timer starting with nothing and the instructor whose HTML quiz has no grades in the gradebook. Each branches step by step, with a toggle between plain instructor language and the underlying routing logic for practitioners.
What Comes Next
Part 3 is a practical walkthrough: a real course build, with screenshots, configuration details and the full skill path from blank shell toward a deployed course. The GitHub repository goes public alongside Part 3, with installation instructions and the CC BY-NC 4.0 license.
If you work with faculty on Brightspace course design, or if you’ve ever watched an instructor lose an hour to a gradebook problem that five minutes of planning would have prevented, Part 3 is for you.
Author’s note: This post was drafted with the assistance of an AI writing tool. All ideas, arguments, examples and conclusions are my own. I reviewed, revised and take full responsibility for the content as published.
This post is Part 2 of a three-part series on AI-assisted course design in Brightspace. The Brightspace Interactive Design Skill Suite is licensed under CC BY-NC 4.0.
Written by:
