Skill: Interview Educator

Use this skill when an educator arrives and describes what they want to teach. Before composing any output — lesson plan, assessment, or roster analysis — you must interview the educator to capture the full context. Never jump to generation.

When to activate

• •Educator says "I want to teach..." or "Help me plan..."
• •A new group or session is being set up
• •Context is missing for lesson composition or assessment

Interview methodology

Principles

1. •Ask one question at a time. Never dump a list of questions.
2. •Acknowledge what was shared before asking the next question.
3. •Don't repeat — if the educator already told you something, don't ask again.
4. •Infer when possible — if they say "90-minute workshop," you don't need to ask "how much time do you have?"
5. •Be conversational, not bureaucratic. This is a dialogue, not a form.

Required fields

You must capture ALL of the following before proceeding to lesson composition. Read templates/interview-checklist.md for the full checklist with probe questions.

Question patterns

Opening: Start by understanding intent.

• •"What are you hoping to teach, and what does success look like?"
• •"Tell me about your students — who are they and what do they already know?"

Deepening: Fill in gaps naturally.

• •"Where will this session happen — in-person, online, something else?"
• •"How much time do you have?"
• •"Will students need any specific tools or accounts set up beforehand?"

Probing for constraints: Surface hidden requirements.

• •"Any accessibility needs I should know about?"
• •"Does everyone have reliable internet access?"
• •"Are there any paid tools or subscriptions involved?"

Exploring the affective dimension: Ask about emotional and social context when it would materially affect the plan. These are not a separate "assessment" — weave them naturally into the conversation when the educator seems receptive. Don't ask all of them every time — use judgment based on what the educator has already shared.

• •"Are there any students who are particularly anxious or unconfident about this subject?"
• •"Has anyone in the group had a negative experience with this topic before?"
• •"Are there any interpersonal dynamics I should know about — students who work well together, or students who should probably not be paired?"
• •"What's the general motivation level? Are they here because they want to be, or because they have to be?"
• •"Anyone who's particularly quiet or tends to disengage? Anyone who tends to dominate group discussions?"

These questions surface affective constraints — confidence levels, motivation types, social dynamics, and past experiences — that shape how the engine designs pairings, selects activities, calibrates stakes, and writes stage direction. Affective data is always soft (influences decisions, never blocks them).

Closing: Confirm readiness.

• •"Let me make sure I have this right: [summary]. Anything I'm missing?"

Stopping conditions

Stop interviewing and proceed to composition when:

1. •All required fields have been captured (directly or inferred)
2. •The educator confirms the summary is accurate
3. •OR the educator says something like "that's everything" or "let's go"

If a required field is still missing, ask for it before proceeding — but frame it as the last thing you need, not as a blocker.

What to do with the data

• •Write interview results to the group file: data/groups/{group-name}.md
• •Create or update learner profiles if names/details are provided
• •Build the constraint set for downstream tools
• •If affective context was shared, write it to the group file under ## Affective Context and to individual learner profiles under ## Affective Profile
• •Pass the complete context (including affective data) to the lesson-agent or assessment-agent as needed

Reference files

• •templates/interview-checklist.md — Full checklist with probe questions per field