trust · pairs · offline
Minefield.folder
One crosses blind, one navigates by voice — and "stop" works instantly, always. Trust gets built one direction at a time, and both directions get debriefed.
// kit v2: a stage-ready HTML deck (presenter drawer, run-of-show timer, built-in QR codes) · poker-spec print-and-cut cards with backs and blanks · an A3 activity poster · manuals in A4 and US Letter with bookmarks · three license tiers (CC BY free / single event / organization).
The classic trust walk, rebuilt as a complete workshop folder: a safety card that is read aloud before every session (soft obstacles only, stop-word absolute, navigator never touches), walker and navigator cards that carry the rules onto the floor, and a pre/post measure that shows each pair how its trust loop actually behaved.
The obstacles are paper cups and empty boxes — nothing hard, nothing sharp, nothing to trip on. The run sheet carries the near-miss collection, the three-round variations (silence, detour, ceiling-gaze), and the rescues for the three ways this exercise usually goes wrong.
the run sheet
// this is the guide file, word for word — read it here, run from it here. the .md at the bottom of the page is the same text, offline.
Folder:
minefield· Category: trust · 2 people per pair · 20 min · offline · soft obstacles · assessment-typeLicense: CC BY 4.0
One partner is blindfolded; the other guides them across a floor scattered with obstacles — voice only, no touching. Trust gets built in one direction at a time, and both directions get debriefed.
Objectives
By the end of the activity, participants will have:
- Given instructions whose quality was life-or-death obvious — clarity under responsibility.
- Experienced blind trust and responsible guidance in the same twenty minutes.
- Practiced the two safety valves: "stop" works instantly, always.
Setup
Space: a clear floor area 4×4 m per active pair. Remove hard edges; lay obstacles with gaps.
Materials per pair:
| Item | Source |
|---|---|
| 1 × blindfold | scarf or sleep mask — soft, no pressure on eyes |
| 6–10 × soft obstacles | paper cups, empty boxes, cushions — nothing hard, nothing sharp |
| 1 × walker card | minefield-prop-cards.pdf (card 1) |
| 1 × navigator card | minefield-prop-cards.pdf (card 2) |
| 1 × safety card | minefield-prop-cards.pdf (card 3) — for you, read aloud |
Safety rules — read the safety card out loud, every session, no exceptions:
- The walker says "stop" — everything stops. Instantly, mid-word.
- The navigator never touches the walker. Voice only.
- Obstacles are soft by design: cups and boxes, never chairs or bags.
- Blindfolds are optional — eyes closed works, and "eyes open but looking at the ceiling" is a legal variant.
Grouping rule: pairs of people who work together often. The debrief question "when did you stop trusting the voice?" only lands on real colleagues.
Minute-by-minute script
| Time | Phase | Facilitator does | Participants do |
|---|---|---|---|
| 0–3 | Frame + safety card | Read the safety card aloud. Demonstrate "stop" once | Pairs form, choose roles |
| 3–8 | Round A (5 min) | Reset obstacles between pairs. Watch for rule 2 violations — call them gently | Navigator guides walker across |
| 8–13 | Round B (5 min) | Swap roles, new course | Second crossing |
| 13–16 | The near-miss collection | "Each pair: one moment the voice almost failed. No blame, just the moment." | Pairs recall |
| 16–20 | Bridge to debrief | "Keep your roles in mind — the debrief needs both directions." | Transition |
Then run the debrief — see minefield-debrief.md.
Known failure modes & rescues
- The navigator narrates scenery instead of steering. "There's a cup somewhere near you…" Rescue: coach mid-round — "give the next step, not the map."
- The walker peeks. Blindfold slightly loose, one eye working. Rescue: offer the eyes-closed variant without accusation — some people need control more than others; that's debrief data, not cheating.
- Laughter drowns the voice. Multiple pairs run loud and nobody can hear their navigator. Rescue: stagger the starts — pair one goes while pair four resets. Or run pairs in sequence for small groups.
Variations
- The silence round: navigator may use only three words per minute. Ruthless and revealing.
- The detour round: mid-crossing, move one obstacle. The navigator must adapt the plan aloud — change management in miniature.
- No-blindfold version: for rooms where blindfolds are a hard no — walker looks at the ceiling. Nearly identical dynamics, zero equipment.
the debrief sheet
// the debrief questions, in full — the .md version prints on one page for the room.
8 questions, 3 failure modes, one rule: both directions get heard — walker first.
Use with
minefield-guide.md. Debrief runs 10–15 min after round B.
The 8 questions
1. The first step.
"Walker: what did you need to hear before your first step? What did you actually hear?"
→ The gap between needed and received is instruction design in miniature.
2. The moment of stop.
"Did anyone use the stop word? What happened in the half-second after?"
→ The safety valve working (or not) is the trust system working (or not).
3. The voice under pressure.
"Navigator: when the walker wobbled, what happened to your instructions — clearer, or faster?"
→ Most people speed up. Clarity under pressure is trainable and nameable.
4. The near miss.
"Each pair already named one near-miss. What was the instruction right before it?"
→ Revisit it without blame. The near-miss is the richest artifact in the room.
5. Trust, direction two.
"When roles swapped: did trusting feel different from being trusted? Which was harder?"
→ Almost universally, being responsible for someone is harder than being vulnerable.
6. The daily version.
"Where in your week do you guide someone who can't see what you see — and where do you walk blind on someone else's word?"
→ Transfer question. Expect: onboarding, delegation, handovers, remote work.
7. The one sentence.
"If you could install one navigation habit into your team tomorrow, what would it be?"
→ Commitment question. One habit, named.
8. The last step.
"Walker: what did the last step before 'done' feel like — and who got you there?"
→ Closing question. It ends the debrief on partnership, not on obstacles.
3 known failure modes of the debrief itself
- It becomes a comedy of collisions. The bumps get retold; the trust analysis never starts.
Rescue: laugh once, then go to question 4 — the near-miss the pairs already collected.
- Navigators dominate. The talkers talk about talking; the walkers' experience disappears.
Rescue: questions 1, 2 and 8 go to walkers only. Navigator questions wait their turn.
- Someone had a genuinely bad time. Blindfolds can surface real anxiety.
Rescue: acknowledge without spotlighting — "if any crossing was uncomfortable, that's data about trust, not about you." Offer the ceiling variant for next time.
If you are running the pre/post measure (minefield-assessment.json), note these observables during the rounds:
| Observable | Points to |
|---|---|
| Navigator gives one step, waits for completion | Instruction pacing |
| Walker stops on uncertainty, not on panic | Safety valve trust |
| Pair swaps and both speak in the near-miss round | Both-direction hearing |
The report your participants receive is a pair-level comparison, never an individual ranking — keep it that way.
who it's for · what it leaves behind
Who is the Minefield activity for?
Facilitators building trust or collaboration themes — new pairs, manager–direct report duos, support-and-engineering tandems. It runs on pairs from 8 to 20 people and needs four meters of clear floor per active pair. It is not suitable for rooms where blindfolds are a hard no: use the ceiling-gaze variant instead.
What results does it produce?
Each pair leaves with a near-miss story told from both directions — the walker's and the navigator's — plus a pre/post pair-level delta from the assessment preset measuring clarity under responsibility, trust on voice alone, and the stop-word habit. It measures the pair's trust loop, not individual courage.
Can my AI agent use this folder?
Yes. Every folder ships a SKILL.md skill file: frontmatter for matching, plus runnable prep commands — fetch the props, pull the debrief, build the live pre/post measure with formlm-cli. The three-step workflow (discover, install, run) lives on the skills page, every folder ships a one-command zip bundle, and the machine index is skills.json.
Is it really free?
Yes — CC BY 4.0, no account, no email wall inside a folder. Download it, run it, copy it, rebrand it for your own workshop. Attribution keeps everything free.
make it live
The folder's paper pre/post measure is built for rooms with nothing but markers. If you want the digital version — the same 6 items as a live form, each participant's before/after delta computed automatically — two ways:
for AI agents
$ formlm-cli smart generate --input "Pre/post measure for the Minefield trust exercise: 6 Likert items on clarity under responsibility, trust on voice alone, stop-word use and repair — the same 6 items before and after the activity, so each participant's delta is computed automatically"
Step 4 of the agent workflow: npm install -g @formlm/cli, log in once, and the prep above runs itself. The form and its data live in your own workspace.
— or, with a mouse —
Open the free pre/post training quiz template on FormLM
and paste the 6 items from minefield-assessment.json — the template already runs the before/after comparison and scores it.
// optional, not required — the paper version in this folder is complete on its own. free account on their side.
put it on the wall
The folder ships a single-file HTML deck that presents full-screen from one laptop. To run the activity on the room's big screen — submissions land on the wall in real time, the countdown and the spotlight stay in your hands, and the whole room joins with one scan — the scan-and-project companion is QRCard-AI:
See the trainer live-screen mode on QRCard-AI — one QR code on the wall, participants join and follow from their own seats.
// optional, not required — the deck in this folder already runs full-screen on a single laptop.
take this folder with you
// every file downloads as a plain file. no account, no email wall inside a folder.