Skip to main content

Live AI-assisted prototype — agent outputs are generated through OpenRouter and require human review.

Agentic Lesson Studio

Human-governed AI instructional design

A human instructional designer and a group of specialised AI agents collaborate to analyse, design, develop, test, publish, deliver and evaluate a lesson. Every agent output requires explicit human approval — no exceptions.

Agentic instructional design is a prototype to demonstrate the coordinated execution of a complete instructional-design lifecycle under human-authored specifications, explicit quality controls and accountable approval gates.

Open the studio
Learner view (publish a lesson first)Open instructor queue
What is this?

Eleven specialised agents

Analysis, learning design, subject research (with real web search), assessment, content, learner simulation, quality review, accessibility, runtime tutor, assessment judgement, evaluation.

Approval-gated workflow

Every AI output enters awaiting-approval. Approve, edit-and-approve, request revision, or reject — each decision recorded as a separate audit event.

Deterministic controller

Workflow progression is pure TypeScript logic, not another LLM. Revisions cascade downstream approvals to stale.

Distinguishable provenance

AI-generated, human-edited, human-authored, application-generated and synthetic content are visually distinct throughout.

What is Agentic Lesson Studio?

Agentic Lesson Studio is a live AI-assisted prototype that demonstrates how a human instructional designer and specialised AI agents collaborate to analyse, design, develop, test, publish, deliver and evaluate a lesson. The human defines the educational intent, remains accountable for all decisions, and approves every agent output. The agents perform bounded instructional-design tasks, operate from explicit specifications, and return structured, reviewable artefacts.

This is not a general autonomous agent platform. It is a controlled demonstration of a complete instructional-design lifecycle with deterministic workflow control. Every AI output requires explicit human approval, and human edits.

Human-authoredSeed lesson

Verify an AI-generated claim before using it

A 45-minute lesson for adult learners beginning to use generative AI. Loaded automatically for each account so you can experience the full lifecycle end to end.