name: training-engagement-check description: > Audit session for passive stretches, energy gaps, and expectation-checking moments. Flags problems with exact scripted questions and timers. Use for "engagement check", "energy audit", "is this interactive enough". version: 2.0.0
Engagement Check
Audit a session plan or run-of-show for engagement gaps, energy problems, and missing expectation-check moments. For every flagged issue, output an actionable fix with exact words Tim can say and a timer.
Input
Pass a session plan, run-of-show, transcript, or outline. Include timestamps or time markers.
Output Format
ENGAGEMENT AUDIT: [Session Title]
WHERE | WHAT'S WRONG | DO THIS | SAY THIS (exact words) | TIMER
──────────|─────────────────────────────────|────────────────|─────────────────────────────────────────────────────────────────|──────
Min 12-22 | 10 min passive, no interaction | Pair-share | "Turn to the person next to you. What's one thing in your | 90 sec
| | | org that this framework explains? 90 seconds." |
Critical Design Rules
For Tim:
- Tim is self-described "really bad at engagement questions." Every question must be EXACT WORDS he can say. Not a concept. Not "ask about their experience." The literal sentence.
- Questions are calibrated to his High I style — designed to give the room something to give HIM so he can riff.
- Timers are mandatory because Tim will either cut it short (impatience) or let it run long (he got interested).
For Balance:
- Flag sections where he's so interaction-heavy there's no space to land a concept.
- Target: 3-4 interactive moments per 30 minutes, varied in type.
Expectation Architecture Integration
Expectation-Checking Engagement Moves
Beyond standard engagement, flag opportunities for Tim to CHECK whether the audience's expectations are being met during the session. These are expectation monitoring operations disguised as interaction.
Expectation Check Move — "Is This Landing?" Use at: ~25% into session (after first major concept)
SAY THIS: "Quick pulse check — on a scale of 1 to 5, hold up your fingers:
how well does what we've covered so far match what you expected when you
walked in? Don't overthink it. Just a gut number."
TIMER: 30 sec
PURPOSE: Surfaces expectation alignment early enough to course-correct.
IF LOW (lots of 1-2s): Tim needs to address the gap. "Okay, I see some
low numbers. Tell me — what were you expecting that we haven't hit yet?"
IF HIGH (lots of 4-5s): Confirm and continue. "Good. We're tracking."
Expectation Redirect Move — "Let Me Reset" Use at: When Tim detects drift (audience disengaged, wrong questions)
SAY THIS: "I want to pause for a second. I promised you [declared expectation
from opening]. Let me check — are we on track for what you need? Raise your
hand if you want me to go deeper on [current topic]. Keep it down if you'd
rather I move to [next topic]."
TIMER: 30 sec
PURPOSE: Real-time expectation redirect. Prevents Tim from pushing content
the room doesn't need.
Peak-End Awareness Flag the session's peak moment and closing moment. If neither is designed for interaction, flag it:
- The PEAK moment (most intense) should include audience participation — it's what they'll remember.
- The CLOSE should include a declared callback — "Remember when I said [opening expectation]? Here's what you now have."
Engagement Audit Additions
Add to the summary section:
EXPECTATION CHECKS: [X planned] — target is 1-2 per session
PEAK MOMENT: [Is it interactive?] YES/NO — [If no, flag as risk]
CLOSE MOMENT: [Does it callback to declared expectation?] YES/NO
"Say This" Question Design Principles
- Short setup, specific ask. Not "what do you think about X?" but "what's ONE thing that..."
- Physical engagement when possible. Raise hand, stand up, turn to neighbor.
- Surface experience, not knowledge. "Tell me a time when..." not "Who can define..."
- Give Tim material to riff. Responses should create moments he can improvise from.
- Avoid put-on-the-spot. Never call someone by name cold.
- Vary the type. Pair-share, full-room poll, shout-out, think-then-share, show of hands.
Engagement Toolkit
Load references/engagement-toolkit.md for the full library of 10 engagement move types with "Say This" templates and timing.
Summary Section
ENGAGEMENT SCORE: [X/10]
PASSIVE STRETCHES: [N flagged]
INTERACTION DENSITY: [X interactions per 30 min — target is 3-4]
ENERGY ARC: [Describe the shape — front-loaded? flat? builds? dips?]
EXPECTATION CHECKS: [X planned — target 1-2]
PEAK MOMENT INTERACTIVE: [YES/NO]
CLOSE CALLBACKS DECLARED: [YES/NO]