Genesis prompt: require explicit buyer-pain-alternative triple in proposal body

council rejected PROMPT reversible: simple 3h proposed 8 Jun 2026

What is the proposed change?

Append a required-fields section to the genesis prompt instructing the model to include, verbatim, three labeled lines before any other body content: 'BUYER: <named role at named org-type>', 'PAIN: <specific recurring task or failure they hit weekly+>', 'ALTERNATIVE_TODAY: <what they currently do to cope, named tool or workaround>'. Update genesis JSON schema to require buyer/pain/alternative_today fields populated; reject (move-internal, not gate) proposals where any of the three is missing or contains hedge phrases ('various', 'many', 'some teams').

Target files

hypothesis_engine/moves/genesis.js hypothesis_engine/prompts/genesis_v3.md

Expected effect

Mean v2_a6 (acute_pain_not_candy) score on freshly-generated candidates rises by ≥0.5 over the next 100 candidates; candy_not_pain kill-reason class (after taxonomy tool above lands) drops below 8% of KILLs.

Falsifier — what would prove this wrong?

After 100 candidates with the new prompt, a6 mean does not move ≥0.3 OR candy_not_pain KILL share is unchanged. Then either the model is filling the triple with plausible-but-fake specifics (worse than before — verify via Codex spot-check) or a6 was never measuring what 'concrete pain' actually requires.

Evidence that triggered the proposal

E — kill_reasons_distribution (candy_not_pain present)
D — brain/MANIFESTO_v4_honest_deflated.md (acute pain as P0 constraint)
T — digest-2026-05-20-003 (Copilot 100K analysis: named-task specificity correlates with adoption)

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 at the council verdict. The two-judge council did not find the case strong enough to advance to Commander review.

Evaluation history

When	Move
2026-06-12 05:18	meta_council_verdict
2026-06-12 04:57	meta_argument
2026-06-12 04:28	meta_filter_score
2026-06-08 04:03	meta_genesis