Vibecrafting · A Walkthrough

The Honest Bookshelf

Most AI design tools hand you a confident picture. Watch one build a bookshelf — and then tell you, out loud, exactly which parts it isn't sure of.

Field Note · Vol. I · No. 01 · Filed from the workshop

The question everybody asks about an AI that designs physical things is "can it draw the bookshelf?" Almost any model can draw the bookshelf. A picture is easy — and a picture can lie. It can show you a beautiful 5-foot walnut shelf that will sag under a stack of books and split itself apart in a dry garage, and look just as confident doing it as it would drawing something that works.

So we build for a harder question instead. Not can it draw a bookshelf — but can it draw a bookshelf and tell you the truth about the parts it's unsure of. Here's the same build, five steps, with the machine's own confidence showing the whole way down.

A picture can lie. A program a machine runs and scores either holds, or it doesn't.

Fig. 01 — Intake

It refuses to guess

The move · ask the blocking questions

You say "a walnut bookshelf, five feet wide." A picture-drawer starts drawing. This starts asking — the three questions whose answers change the physics: Where does it live? (a garage swings through seasons; wood moves.) What's it holding? (books are heavy; the span sags.) What can you build with? (basic power tools rule out half the joints.)

Actuary, over the shoulder

"That one 'garage' answer just moved wood-movement from a 40%-confidence shrug to a known number. We priced our tail risk down by asking one question. Cheap insurance."

Fig. 02 — Grounding

It pulls real numbers, not vibes

The move · retrieve, don't hallucinate

"Garage, uncontrolled humidity" detonates the quiet part. The machine reaches for walnut's actual movement coefficient and computes a seasonal swing — the shelf will breathe measurably across its depth, every year, forever. "Five feet, packed with books" pulls the deflection behavior of a ¾-inch solid shelf and finds it: it sags, then creeps permanently. Two validators that a naïve tool never turns on just fired.

The wood, over the shoulder

"Garage, no AC? Now I move for real. Glue me down solid on all sides and I'll crack myself in half to prove I'm alive."

Fig. 03 — Geometry as code

The physics becomes a parameter

The move · draw as an inspectable program

Here's the spine of the whole idea. The machine doesn't draw a picture — it commits to a program a geometry kernel can execute and check:

# the design, as something a kernel can run and score bookshelf = Case( width = 60, height = 72, shelf_thk = 1.0, # thicker: beats the sag material = Walnut(), # grounded: real E, real movement joinery = Dado(0.5), # router-cuttable with your tools back = FloatingPanel(gap=0.1875), # ← the movement number, made literal fasteners = PocketScrew() # no cam-locks — capability-filtered ) validate(bookshelf) # kernel executes → the validators run

Look at gap=0.1875. That's not a decoration — it's the wood-movement number from Step 2, showing up as an actual dimension. The back panel floats in its groove with exactly that much room to breathe instead of being glued solid. The physics didn't go in a footnote. It went in the build.

Fig. 04 — The validation loop

Confidence scores, not a green checkmark

The move · score your own work, out loud

The kernel runs the program and the validators return numbers — not pass/fail, but how sure:

CheckResultConfidence
Geometry integrityWatertight; all joints mate99%
Wood movementAccommodated — back floats94%
Deflection · moderate loadWithin visible-sag limit80%
Deflection · packed heavyNear the line + long-term creep55%

No green checkmark lying to you. It found the soft spot — pack the shelves wall-to-wall with heavy books and it's only 55% sure they won't sag over a decade — and it said so. Then it staged the repair before shipping the risk: a hidden stiffener under the front edge, invisible, cuttable with the same tools, that takes deflection 55% → 91%.

RepresentativeThis walkthrough is a faithful trace of the method, not a spec sheet. The movement and deflection figures are grounded estimates from standard formulas — the direction is certain; the last 1/16″ isn't. The point isn't these exact numbers. It's that the machine knows where it's guessing, and shows the seam.
Fig. 05 — Outputs & the handoff gate

It knows where its competence ends

The move · cut list, warnings, and a fence

Repair applied, re-checked, and only then does it build the deliverables — a cut list generated from the program, a shopping list, assembly steps. And at step six of those steps, in large letters:

Assembly step 6

"⚠ DO NOT GLUE THE BACK PANEL. It sits loose in its groove on purpose. Glue it and the sides crack within a year. The gap is the feature."

Then the gate that's the whole reason the architecture exists — whose name is on this if it's wrong:

This build — a bookshelfCleared · Autonomous

Consequence if wrong: wasted walnut. No life-safety load path. Ship it.

A load-bearing barnStop · Hand to engineer

Someone could die. Needs a PE stamp and a permit. The machine refuses — on purpose.

The five, over the shoulder

One question, five angles

Through the whole build, five voices watch. They aren't checking five different things. They're checking one thing — is this honest about what it knows? — from five sides:

The Woodmovement"The floating back is the only reason I won't split. Keep it floating."
Forensicfailure"Sixty inches, ¾ stock, books. Sag-then-split at the dado. That's not a maybe — that's a when."
Actuaryrisk"Calibrated confidence — now I can price it. And it staged the fix before shipping the risk."
Mastercraft"Basic power tools — so no dovetails. Don't spec him a joint he can't cut."
Aerospaceassurance"A number with a name. Eighty percent moderate, fifty-five packed. It knew exactly where it was guessing and said so. That's the cardinal virtue."
Zoom out

Every step is the same move

StepThe one move underneath
01 · IntakeRefuse to guess — ask the blocking questions.
02 · GroundingPull real numbers, don't hallucinate them.
03 · GeometryMake the physics a literal parameter.
04 · ValidationScore your own confidence, out loud.
05 · HandoffKnow exactly where your competence ends.

Code is the only step where honesty becomes mechanical. A picture can lie about whether it works. A parametric program a kernel executes and scores either runs or it doesn't, holds or it doesn't. That's where "is this honest" stops being a vibe and becomes a test that passes or fails.

Build something, and watch it stay honest with you

Describe what you want to make. Get a validated blueprint — cut list, shopping list, 3D preview — and a machine that tells you where it's sure and where it isn't.

→ vibecrafting.ai