Submission and debugging checklist

Exact source identity. Retain the supplied version, immutable checkpoints, and configuration. Do not substitute an unexamined repository revision.

Complete state. Store masks, scars, modulation, protection, decay rate, phase, optimizer convention, and named streams.

Endpoint conventions. Floor deletion and minimum-one deletion are different rules. Zero base severity is also affected by the source's lower clamp.

Tie conventions. Stable exact-count ranking is not identical to threshold deletion when values tie.

Initial versus episode scars. Initial placement may exclude inactive repair capacity. Episode scars select currently active coordinates.

Gain reset. The source reset sets $a=g=1$ and restores ReLU. It does not remove protection.

Stage labels. Record pre-challenge, post-decision, and post-scar quantities separately.

Controller and assessment data. A readout driving future updates is controller data. Use fresh blocks or name the finite-design target.

Diagnostic target. Clean activation, perturbed activation, raw mean, and clipped mean are different quantities.

Threshold consistency. A threshold of $0.70$ is not the earlier $0.60$ event. A smaller positive bias is not an untriggered condition.

Counterfactual permanence. Permanent versus regrowable branches start with the same immediate deletion; later eligibility is the differing intervention.

Optimization boundary. Risk infima do not imply Adam convergence or checkpoint ordering.

Conditional-history errors. Normalize rare branches only with a recorded probability lower bound and its error allowance.

Operational Markov witness. Use histories accessible through the claimed observation class. External damage flags alone do not prove a carrier-only witness.

Quantum-resource boundary. Classical memory, coherence sensitivity, and computational advantage are distinct claims.

Evidence types. Analytic derivation, compilation, executed regression, numerical reproduction, and clinical validation remain separate.