33.38 turns multi-phase nuclear outflows into a slow-leak tightening court: after frozen common-beam, common-bandpass, optical-depth, and time-alignment rules, ionized, neutral, and molecular tracers must reproduce the same-sign low-velocity plateau or constant term, align with pore-breathing windows at zero lag or a short predictable lag, and remain azimuthally co-located with near-core pores or the jet axis across arrays and years; under V08/V09-compatible tighten, this remains only a near-core–outflow coordination ledger, not an extreme-universe or structure-formation verdict.
Use this section as a compact machine-readable EFT reference.
33.38 reframes the low-velocity base of nuclear outflows as a co-window court. The question is not whether one can narrate a gentle outflow ramp, but whether a slow-leak plateau or constant term closes with pore-breathing windows across phases and arrays; under compat adjudication, the chapter is tighten.
The section measures a slow-leak spectrum index, low-velocity plateaus and slopes, sign and rank stability across ionized, neutral, and molecular tracer families, zero-lag or short-lag closure with polarimetry and pore-breathing windows, azimuthal co-location with near-core pores or the jet axis, and environment-linked changes across host and lensing layers. Any signal that flips with frequency, optical depth, or line family is reassigned to ordinary transfer or contamination ledgers.
Optical/near-infrared integral-field spectroscopy, millimeter interferometry, centimeter polarimetry, and X-ray timing are forced into one common-beam, common-bandpass, frozen-optical-depth workflow. Two blind pipelines independently output the slow-leak index, all series are tied to a single time anchor, and feed-forward cards use only filament orientation, weak-lensing layers, gas density, star formation, and masks while cross-checking pore and polarization conventions from earlier chapters.
The chapter makes positive controls explicit, then attacks them with negative controls and nulls: line labels, array labels, time-window labels, and spatial-orientation labels are permuted; host-filament and weak-lensing layers are rotated or shifted; and radiative-transfer, dust, beam, point-spread-function, and timing ledgers are allowed to absorb the pattern first. A slow leak that survives equally under these nulls is method bias, not evidence.
The pass line requires at least two cleaning pipelines, two array datasets, and three tracer families to reproduce a non-dispersive slow-leak plateau or constant term with zero-lag or short predictable lag against pore breathing, plus azimuthal co-location and holdout success. If dust, radiative transfer, starburst winds, beam or point-spread-function choices, or timing aliasing dominate, or if permutation baselines stay equally strong, the chapter is falsified.
The section closes with only one near-core–outflow coordination ledger under frozen beam, bandpass, timing, and holdout rules, not an extreme-universe, jet-totality, or structure-genesis verdict. If the slow leak survives, it sharpens a local pore-breathing interface and routes forward to 33.39; otherwise it returns to line-transfer and alignment ledgers.