trust · whole group · offline + online
The Reliability Ledger.folder
Trust leaks from expectations nobody wrote down. Say what you'll reliably do, what you quietly need, and the one thing nobody named. Then commit one repair.
// commitment & need cards · 25 min · print & run · tool pack · no projector, no account
A trust session, built as a complete workshop folder: a timed run sheet, printable cards, and a write → read & post → surface-the-unspoken → repair mechanic that turns quiet resentment into stated commitments and one concrete fix — with an owner and a check-in date, so it's a ledger, not a complaint box.
The props are poker-spec print cards (63.5 × 88.9 mm, bleed and cut guides included) — two commitment cards, a need card, an unspoken-expectation card, a repair-action card, a blank card and branded backs. Nothing needs a projector, and nothing needs an account to download.
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: `the-reliability-ledger` · Category: trust · 4–10 people · 25 min · offline + online · print props · tool pack License: CC BY 4.0
A 25-minute, 4–10 person trust tool. A team makes its implicit contract explicit: what each person will reliably do, what each quietly needs from the others, and the one broken-but-unspoken expectation everyone has been tiptoeing around — then commits to one repair with an owner and a check-in date. Trust here is treated as a record of kept promises, not a feeling you manufacture with games.
How this relates to the other trust folders: The Reliability Ledger writes down what The Safety Check only measures and what The Trust Timeline only maps. Step In If… shows shared experience without leaving a record; the ledger deliberately does — commitments, needs, one surfaced expectation, one owned repair. Run it when the trust leak smells like "I thought you had that."
Objectives
By the end of the activity, participants will have:
- Written and read aloud two operational commitments ("I'll reliably…") and one honest need ("I quietly need…") — the implicit contract made explicit.
- Surfaced, as a group, one expectation everyone quietly held but nobody ever said — and named what actually happened instead.
- Committed to one concrete repair for that expectation, with a named owner and a check-in date, plus one personal commitment each that keeps it real.
When to use this activity
- A team with enough history that expectations have quietly formed — and quietly started disappointing.
- Recurring low-grade friction nobody raises because "it's not that serious": missed handovers, silent resentments, assumed coverage.
- Two people who keep hitting "I thought you were handling that" — the peer-pair version fits over one table.
Setup
Clock: twenty-five minutes end to end. Give it thirty on a first run with a team that has real history — naming the unspoken one takes longer the honest it is.
Space: one wall or shared board with two columns — Commitments · Needs — plus a spot for The unspoken one. Online, the columns live on a shared board.
Materials per person:
| Item | Source |
|---|---|
| 2 × commitment faces — "I'll reliably…" | `the-reliability-ledger-prop-cards.pdf` |
| 1 × need face — "I quietly need…" | `the-reliability-ledger-prop-cards.pdf` |
| 1 × "The unspoken one" face + 1 × "Repair action" face + blank (the team's, not each person's) | `the-reliability-ledger-prop-cards.pdf` |
| 1 × pen; wall tape or board pins | your own |
| Optional: the poster with the two rules in view | `the-reliability-ledger-activity-poster.pdf` (A3) on the wall |
Grouping: 4–10 people who work together and share obligations; run the whole group against one wall. A peer pair of 2 works as the table version.
Roles: everyone writes and reads their own cards; you keep phrasing honest (requests, not grievances), guard the no-debating rule during the read, and force the close — one repair, one owner, one date. No projector, no account, no online form required.
How to run it
| Time | Phase | Facilitator does | Participants do |
|---|---|---|---|
| 0–7 | Write | Hand out the faces. "Commitments are operational, not confessions: things you will actually do. Needs are requests: 'I need X by Y.'" | Write two commitments ("I'll reliably…") and one need ("I quietly need…") |
| 7–13 | Read & post | Read each card aloud and post it under Commitments or Needs. No debate, no defending — just hearing. | Listen to what colleagues have been carrying and waiting for; post without explaining |
| 13–19 | Surface the unspoken | Ask: what did everyone quietly expect but nobody ever said? Write it on The unspoken one and say what actually happened instead | Name it as a shared pattern ("we've all started assuming…"), not one person's failure |
| 19–25 | Repair | Force the close: one concrete repair, an owner who takes it (not one assigned), a check-in date. "What will 'this got fixed' look like on that day?" | Agree the repair action; each person adds one commitment they'll personally keep to make it real |
Then run the debrief — see `the-reliability-ledger-debrief.md`.
The card
The faces mirror the ledger's double entry: two commitment faces because reliability is a record of kept promises, not one grand vow; one need face because what you quietly wait for is the leak's other half. The unspoken one and Repair action exist so the conversation can't end at awareness — a ledger with no repair action is a complaint box, and the check-in date is what makes it a book.
Failure modes
- Needs come out as complaints. Rescue: rewrite each one as "I need X by Y" before it goes on the wall. A grievance posted is a grievance re-lived; a request posted is a promise someone can keep.
- The unspoken one names a person. Rescue: reframe it as a pattern — "we've all started assuming…" — and move on. The moment it becomes one person's failure, the room closes.
- No one wants to write a commitment. Rescue: start with boring, operational ones: "respond within a day", "flag blockers early", "say when I'm swamped". Safety is built by kept small promises, not by confessions.
Variations
- New team, week one: write only commitments and needs; skip the unspoken one until there's history to surface.
- After a specific rupture: go straight to the unspoken one, then work backwards to the commitments that would have prevented it.
- Peer pair: two people run it across a table; the two columns make "I thought you had that" visible fast.
- Standing ritual: re-run quarterly against the previous wall — kept promises earn a tick, broken ones earn a repair.
Facilitation notes
- The unspoken one is named as a shared pattern, never a culprit — that single rule is what keeps the room open.
- A ledger with no repair action is a complaint box: no date, no owner, no deal.
- Don't average or merge needs away; two people needing the same thing is free trust, and you'll find those overlaps in the read-aloud if you let it breathe.
Put it on the wall
Print `the-reliability-ledger-activity-poster.pdf` (A3) to keep the two rules — "needs are requests, not grievances" and "no repair action = a complaint box" — in view through the whole run.
Props in this folder
- `the-reliability-ledger-prop-cards.pdf` — poker-spec print sheet (63.5 × 88.9 mm, bleed and cut guides): two commitment faces, one need face, the unspoken one, repair action, blank card, branded backs
- `the-reliability-ledger-guide.md` — this run sheet
- `the-reliability-ledger-debrief.md` — the debrief questions and the failure modes of the debrief
- `the-reliability-ledger-run-sheet.pdf` — the print edition: this run sheet + the debrief, bound as one A4 document with bookmarks
- `the-reliability-ledger-activity-poster.pdf` — the whole tool on one A3 wall sheet: the two columns, the flow, the watch-outs
- `the-reliability-ledger-start-here.md` — one page, thirty seconds, then print and go
- `the-reliability-ledger-license.md` — what you may do with all of this
- `SKILL.md` — machine-readable manifest for AI agents
- `*-Letter.pdf` — US Letter editions of the print documents
Agent version
Every file in this folder has a stable public URL — an AI agent can fetch the cards, the debrief and this run sheet with three curl commands (see `SKILL.md`). The paper version needs nothing but a wall, a deck of faces and pens.
the debrief sheet
// the debrief questions, in full — the .md version prints on one page for the room.
A ledger earns trust only when it gets read, tested and repaid — the debrief is where commitments stop being wallpaper. Use with `the-reliability-ledger-guide.md`. The repair phase at minute 19 is the compressed version; this sheet is the full one.
How to run this debrief
- Keep the wall up — Commitments, Needs and the repaired unspoken one all visible. This debrief reads a document the team just wrote, not a memory.
- Separate the three verbs: read (what's on the wall), test (will the repair hold), commit (what each person keeps). Don't blend them into feelings talk.
- Close with the check-in date read aloud. A repair nobody will be asked about again is a note, not a promise.
Read the ledger
- Scan the Commitments column. Which ones surprise you — things people were already doing that you didn't know about?
- Scan the Needs column. Which need was hardest to say aloud, and what made it risky?
- How many needs are actually requests for something someone else already committed to? That overlap is free trust.
The six debrief questions
1. The hidden overlap.
"Which need on this wall is already someone's commitment — and were we solving that one by accident before?"
→ Targets: free trust; the overlap the implicit contract was hiding, now visible and nameable.
2. The risky need.
"Which need was hardest to say out loud today? What would have to change here for it to be cheap to ask for next time?"
→ Targets: the cost curve of asking; the unsayable need is the ledger's early-warning line.
3. Projection check.
"Is the unspoken one genuinely shared, or did two people quietly project it onto the room?"
→ Targets: honesty about the group's story; a projected expectation gets rewritten as a pair conversation, not a team fact.
4. The repair's load-bearing part.
"Will the repair action actually hold, or does it need a decision right nobody has yet? And did the owner take it — or was it assigned?"
→ Targets: whether the repair can fail for structural reasons; an assigned owner is a complaint with a name on it.
5. Fixed looks like what.
"What is the check-in date, and what will 'this got fixed' look like on that date — specifically enough that anyone could spot it?"
→ Targets: converting intent into an observable; unfalsifiable repairs quietly dissolve.
6. The personal keep.
"What is the one commitment each person will personally keep to make the repair real — and what did you assume someone would do that they never knew you were waiting for?"
→ Targets: the ledger's double entry closing on every person, not just the culprit of the story.
Failure modes of the debrief
| Failure mode | Why it happens | The rescue |
|---|---|---|
| Needs resurface as grievances | The wall has started feeling safe enough to unload on | Re-phrase live: "you never X" becomes "I need X by Y" before the sentence finishes. A posted request is a promise someone can keep. |
| The unspoken one gets re-personalised | In discussion, the pattern quietly slides back onto one person | Reframe as "we've all started assuming…" out loud, twice if needed. The moment it becomes one person's failure, the room closes. |
| The repair fades after the meeting | Nobody is scheduled to ask whether it held | Set the check-in date before anyone leaves and put it on the team calendar; if it fails, ask what was missing — owner, decision right, or willingness to name it again. |
Who is The Reliability Ledger for?
Teams with some shared history where expectations have quietly formed — and quietly broken. It needs 4–10 people and 25 minutes; it runs on a shared board online.
What does it produce?
A wall of stated commitments and needs, one surfaced unspoken expectation, and one repair action with a named owner and a check-in date. No pre/post measure; the artifact is the ledger plus each person's kept promise.
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 cards, pull the debrief. 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 yours
This ledger needs no online form and no account — two columns, cards, a marker. The card prompts are the one file to change: rename I'll reliably / I quietly need / The unspoken one / Repair action to how your team already talks, and reprint at poker spec; the check-in-date rule stays.
Files in this folder
All files are free under CC BY 4.0. Attribution: “The Reliability Ledger · WorkshopSkill · CC BY 4.0.”