Promote closed-thesis prune from ranking preference to pre-genesis GATE

filter rejected GATE reversible: simple 5h proposed 8 Jun 2026

What is the proposed change?

Add a pre-genesis gate that hard-rejects proposals whose thesis matches any entry in a rolling 30-day Commander KILL log. Currently (post-S183 decage) closed-thesis matching is a ranking demotion only, so genesis still spends meta_genesis tokens producing variants Commander has already killed. The gate reads commander_overrides.json, builds a thesis-key set (use existing thesis_class tag if present, otherwise normalized title trigram), and short-circuits before the LLM call returns. Add an override flag --allow-closed-thesis for explicit Commander re-explore.

Target files

hypothesis_engine/moves/genesis.js hypothesis_engine/gates/closed_thesis_gate.js

Expected effect

meta_genesis token cost in the move_cost_rollup drops 15-25% over the next 7d window; KILL-with-reason-off-thesis rate on graduated candidates drops below 10% (currently dominant in Corpus E kill-reason distribution).

Falsifier — what would prove this wrong?

After 7 days of gate-on operation, if (a) meta_genesis cost does not drop measurably AND (b) off-thesis KILL rate stays flat, the gate is either misclassifying or the off-thesis kills were not actually from closed theses — proposal is wrong.

Evidence that triggered the proposal

E — kill_reasons_distribution_product_lane_7d (off_thesis dominant)
D — brain/S183_decage_genesis.md (demoted shape/side hard-rejects to preferences)
E — move_cost_rollup_7d.meta_genesis

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
2026-06-12 04:27	meta_filter_score
2026-06-08 04:03	meta_genesis