Promote shadow describability gate to v2_a12 axis

filter rejected AXIS reversible: simple 3h proposed 17 Jun 2026

What is the proposed change?

Per Corpus D, a describability check has been running in shadow mode. Promote it to a scoring axis v2_a12. Take existing shadow_score (Q5 NBJ test: can the candidate be described to a buyer in one sentence without metaphor?) and join it to composite. Read shadow_score from the candidate record (already populated by describability_shadow.js) and add to filter_score composite. Remove shadow-only flag once 30+ shadow runs confirm score distribution matches Q5 manual ratings.

Target files

hypothesis_engine/moves/filter_score.js hypothesis_engine/moves/describability_shadow.js

Expected effect

847f7e-shape proposals (per Corpus D pattern reference) score 0-1 on v2_a12; ROBUST candidates score 2-3. Expected composite spread of 2-3 points between the two classes. Graduation rate on metaphor-heavy candidates drops by ~25%.

Falsifier — what would prove this wrong?

Rerun on 43 graduated candidates whose shadow_score is logged. If shadow_score distribution does NOT correlate (Spearman >0.4) with the S157 NBJ sweep manual classification, the axis is miscalibrated and promotion is premature.

Evidence that triggered the proposal

D — brain/S157_NBJ_SWEEP.md — describability Q5 NBJ test
D — brain/META_ENGINE_PHASE_1_SPEC.md — shadow-mode describability gate

Proposer self-score

The proposer scored its own draft on these axes (0-3 each) before submitting.

Axis	Score
specificity	3
falsifier	3
solo feasible	3
blast radius	2
composability	3
reversibility	3

Disposition

Rejected by filter_score. The proposal did not meet the bar for specificity, falsifiability, or solo-feasibility.

Evaluation history

When	Move
2026-06-17 04:06	meta_filter_score
2026-06-17 04:03	meta_genesis