We don't record what you typed.
We record what you meant.
MMC is deckwright's intent-capture layer. While you work on the Bench - our endless spatial canvas, on desktop or iPad - it records the whole living act of creation: what you draw, how your hand moves, what you say and exactly when you say it, and how the thing you're building actually behaves. Those streams fuse into a single, durable record of intent the crew can read, replay, and build against.
Most tools take your click as a coordinate and throw the rest away. MMC keeps the rest. The rest is where the meaning is.
Why we built it this way
Artists don't think in specifications. They think responsively, in real time - gesturing, pointing, reacting to what's in front of them. "More like this. No, here. Bigger. Softer. That one." The intent lives in the motion, the glance, the timing, and the word said at the moment the hand arrives - not in a paragraph written after the fact.
Prompt-based tools force that fluid, embodied way of thinking through the keyhole of a text box. You stop, translate yourself into prose, and hand the machine a flattened description of something you were showing it a second ago. Everything expressive - the hesitation, the decisiveness, the pointing, the pace - is lost in the translation.
MMC refuses that trade. It meets you in your native mode: you gesture and speak and arrange the way you always have, and the system treats that - the actual performance of creating - as the real specification. It's the difference between describing a melody and humming it.
How it works
Four moves: capture, align, decode, keep.
- Capture - every channel at once, always on: canvas, motion, voice, the live surface, the system underneath.
- Align - the streams are stitched onto one shared timeline, so a spoken "make this rounder" locks to the exact shape your cursor was on the instant you spoke.
- Decode - the crew reads the fused signal and infers the referent, the operation, and how confident that reading is.
- Keep - the part no one else does: it all persists. A replayable artifact of intent, not a command that fires once and vanishes.
No screenshots. Nothing inferred that can be recorded.
Deckwright does not work from screenshots. A screenshot is a single flattened surface with no time axis: no motion, no order of operations, no hesitation, no voice - everything a model reads from it is inference, and inference is lossy. Multi-Modal Capture never flattens. Each channel of intent is recorded at the source as its own discrete, first-class, timestamped stream, and the streams are fused on one shared clock. Signal-native, not pixel-derived. The result is not a picture of the work to be interpreted after the fact - it is the work's own instrumentation: exact, replayable, and made of what actually happened.
And no vision model in the loop. MMC does not rely on pre-baked vision modalities to understand the work: no screenshots are taken, and no screen-recorded video is parsed by a model's eyes. The capture arrives as structured, machine-native state - exact coordinates, exact timings, exact canvas, DOM, and console events - so nothing is estimated from pixels that the system knew precisely at the source. Three consequences follow: precision, because a cursor position is a number rather than a guess; economy, because structured deltas cost a fraction of the tokens of video frames; and independence, because the record is legible to any model, with or without vision.
What it captures
Far more than the picture. MMC listens on every channel a moment of creation actually happens on:
- Canvas & structureEvery shape, placement, and spatial relationship - and the whole becoming of the scene, not just its final frame: created, moved, resized, restyled, checkpointed as it evolves.
- Generated imagery, in the loopNot just vector shapes. Images are generated right on the canvas, and the whole loop - generate, react, mark it up, regenerate - is captured as intent, so "more like this, warmer, lose the background" lands on the exact frame you meant.
- Motion & dynamicsThe signature capability. Not just where your cursor went but how: velocity, the easing of a stroke, the difference between a decisive point-and-land and a tentative hunting circle. Dwell. Hesitation. The rhythm and cadence of a whole session. Motion carries intent that words never state - MMC reads it.
- Pencil, on iPadA released surface, not a port: the Bench runs on iPad with full Apple Pencil support, and the pencil joins the capture first-class - stroke, pressure, tilt, and timing fused on the same clock as voice and canvas. A sketch stops being a picture and becomes a narrated, timed act of intent.
- Voice & deixisSpoken narration bound to the exact place and instant it belongs to. "Put that there" resolves, because the system knows where "there" was pointing when you said it. The pause before you speak and the move right after are timed and kept.
- IntentWhere the streams fuse: gesture plus voice plus timing, read together into what you meant - the target, the action, the confidence - instead of a literal transcript of what you did.
- Live surfaces, wired to the consoleLoad a real, running interface onto the canvas and work it like the live thing it is. As you click an actual element, MMC captures the developer console, the errors, and the network calls it fires - at that instant, bound to the element you touched. One channel carries you from designing a surface, to building it, to the part that matters most: catching its bugs in the act. You click the thing that's misbehaving, and the stack trace, the failing request, the console error are already captured and tied to what you meant. The reproduction is the recording.
- System traceThe session tied to what the system actually did underneath, so nothing is read in a vacuum.
- Privacy by designIt captures how you work, never your private contents: values typed into fields are never recorded, sensitive data is redacted at the source. Expressive signal, not surveillance.
- Durable & replayableAll of it is one artifact you can return to, re-watch, and re-decode. Your intent stops being a fleeting keystroke and becomes an asset the whole crew can build from.
Why it matters
The paired gesture-and-voice idea is 45 years old. Everyone who ever built it built it as a command - fire it, execute it, forget it. We're the ones who keep it. That one decision - treating the living act of creation as a durable record of intent instead of a disposable input - is what turns a canvas into a context engine, and a sketch into a spec.