Cinema + labs + WebGL. The labeled map, Hub canvas, field sketch, and every live engine. Use when this is the app that owns the GPU on the machine.Every feature, no graphics. Same engines, same API, same honesty grammar. No WebGL, no custom cursor, no Three.js. For when other apps already hold the GPU, or after learning loops when you only need the work.Terminal + one API out. Every live verb as a command and as HTTP. TUCE agents load skill.md and call /api/v1. No cinema overhead. Occupancy stays nominated | live-bound | unavailable.
Lab · 25 · L0 · refuse-first
Policy pin
FAST cannot commit. UNATTENDED requires model × adapter × lobe. Missing pin fails closed.
NCM 1.19.0 · Hub 2.2.0 · invented_metrics = false
illustrative
Refuse-first
This host cannot commit. The pin still decides the refuse.
can_commit stays false here — token completion and Hub weight bind are unavailable. FAST cannot commit. UNATTENDED without the triple fails closed. There is no silent “routed” success on the marketing caller.
L0
FAST
Cheap path. A FAST seed cannot silently become RESEARCH, and it cannot commit.
L0
STANDARD
Default hub-spoke route with occupancy attached.
L0
RESEARCH
Wider complementary portfolio. Still refuse-first.
L0
UNATTENDED
Pins model × adapter × lobe at the start. No mid-task hot-swap.
STANDARD · can_commit=false
spoke_not_live
Token completion and Hub weight bind are unavailable on the marketing caller. SAT, fit, occupancy, and the agent roster are live. Point TUCE at a live Hub when a model spoke is live-bound.
surface=marketing-caller. Live MoA on a Hub is experimental; demo MoA stays labeled. This host will not invent a completion.
← Watched DPLLSame caps as the Constraint lamp. Two-watched literals plus activity. Capped ≠ UNSAT. SAT ⟂ HAL.All labsEvery this-tab tool. Real DPLL, occupancy, adapters, SAT. Not the Hub runtime.JSON constraints →Paste JSON. Write path == value rules. Unknown fields stay free SAT vars. SAT ⟂ HAL.