docs(questions): add Daily Patch Review Loop Policy v1 and Research Notes v2

seed/questions/DAILY_PATCH_REVIEW_LOOP_POLICY.md

@@ -0,0 +1,116 @@
+# Daily Patch Review Loop Policy v1
+
+This policy exists because a pack can pass schema checks and still be boring, weird, or too generated.
+
+The daily pack must be fixed in patch mode unless a mass rewrite is explicitly justified.
+
+## Core Rule
+
+Review the full pack. Mark only questions that fail. Fix only the marked question IDs.
+
+Do not rewrite passing questions just because a rewrite is happening.
+
+Passing questions are frozen unless they later fail a specific rule.
+
+## Required Inputs
+
+Before editing daily questions, the writer must have:
+
+* the current JSON file
+* the current daily guide files
+* the previous review findings, if any
+* a marked fixes CSV or JSON from the review pass
+
+## Required Review Pass
+
+Each question must be checked for:
+
+* schema validity
+* weekday tag correctness
+* prompt clarity
+* option fit
+* fun-first value
+* therapy or wellness voice
+* household admin drift
+* awkward phrasing
+* repeated opener or repeated mechanic
+* too-random silliness
+
+## Mark Before Fixing
+
+Every failed question must be marked with:
+
+* question ID
+* weekday
+* current prompt
+* failure reason
+* fix scope
+* whether metadata may change
+
+Use these fix scopes:
+
+* `prompt_only`
+* `options_only`
+* `prompt_and_options`
+* `metadata_only`
+* `delete_or_replace`
+
+## Fix Only What Failed
+
+When patching:
+
+* preserve the original ID unless the ID itself is wrong
+* preserve `type` unless the type is wrong
+* preserve `access` unless the count split is wrong
+* preserve `depth` unless the import code or schema requires a change
+* preserve weekday tags unless the tag is wrong
+* preserve compatibility tags unless app code no longer needs them
+* change only the prompt and options for normal content fixes
+
+## Mass Rewrite Exception
+
+A mass rewrite is allowed only when more than 60 percent of a weekday or pack fails for the same root reason.
+
+The rewrite report must state:
+
+* failure percentage
+* shared root cause
+* why patching individual IDs would be worse
+* what will be preserved
+* how the new pack will be reviewed
+
+Without that report, mass rewriting is not allowed.
+
+## Loop Stop Conditions
+
+The loop stops only when:
+
+* every marked question has been fixed
+* every fixed question has been reviewed again
+* remaining hard flags are 0
+* duplicate checks pass
+* option-answer checks pass
+* two random samples per weekday pass cleanly
+* the review findings file says `production_ready`
+
+Do not call the pack production ready with known unresolved content flags.
+
+## Fun But Grounded Rule
+
+Fun does not mean random nonsense.
+
+Reject prompts or options that feel like a carnival generator, such as too many fake awards, too many snack drafts, too many mascot jokes, or cute objects that do not actually fit the prompt.
+
+A daily question should feel playful and usable by adults, not like a children's party game accidentally shipped inside a couples app.
+
+## Required Outputs
+
+Every daily patch must output:
+
+* updated JSON
+* validation report
+* marked fixes file
+* patched fixes file
+* remaining flags file
+* review findings file
+* short apply instructions

seed/questions/FUN_RELATIONSHIP_QUESTIONS_RESEARCH_NOTES.md.md

@@ -0,0 +1,84 @@
+# Fun Relationship Questions Research Notes v2
+
+This research note updates the Closer daily question guides after reviewing modern couple apps, conversation-card games, date-night question lists, and relationship research summaries.
+
+## Main Takeaway
+
+Fun relationship questions work when they feel like a small playable moment.
+
+They are not just "warm prompts". They use choice, humor, tiny missions, preferences, memories, flirtation, and low-pressure honesty.
+
+## What Good Examples Have in Common
+
+* They are quick to answer.
+* They create a tiny story or choice.
+* They ask about favorites, memories, date ideas, silly preferences, attraction, or what sounds good now.
+* They give permission to be playful, weird, honest, or flirty.
+* They avoid sounding like therapy unless the product is clearly built for therapy.
+* They do not make the relationship feel like work.
+
+## What Closer Should Copy As A Pattern
+
+Do not copy outside questions.
+
+Copy these patterns:
+
+* "Would you rather" choices
+* fake awards
+* tiny missions
+* snack drafts
+* date-night picks
+* silly debates
+* comfort-show or music choices
+* flirty low-pressure picks
+* first memory or favorite memory prompts
+* small adventure choices
+* "what should we try tonight" prompts
+
+## What Closer Should Avoid
+
+* chores pretending to be romance
+* bedtime logistics
+* household admin
+* therapy words
+* generic wellness phrasing
+* vague answers like "quality time"
+* awkward domestic answers like "the good blanket saved"
+* options that do not all answer the same prompt
+
+## Specific Guide Changes Made
+
+* Added a research-informed fun rules section.
+* Added required daily game mechanics.
+* Added stronger option checks.
+* Added a research pass before writing daily packs.
+* Added a second sample gate requiring 8 of 10 sampled questions per weekday to feel fun, sweet, flirty, silly, date-like, or game-like.
+* Added metadata fields for `content_policy` and `research_note`.
+
+## Sources Reviewed
+
+* Paired: https://www.paired.com/
+* Gottman Love Maps: https://www.gottman.com/blog/the-sound-relationship-house-build-love-maps/
+* Teen Vogue couples questions: https://www.teenvogue.com/story/145-questions-for-couples-that-are-always-worth-asking
+* Guardian on conversation cards: https://www.theguardian.com/wellness/2024/feb/23/conversation-starters-card-games
+* GQ on Esther Perel's conversation game: https://www.gq.com/story/esther-perel-card-game
+* Verywell Mind spicy questions: https://www.verywellmind.com/spicy-conversation-starters-8737188
+* Partnership through Play research: https://arxiv.org/abs/2505.09509
+
+## Gap Found After v6
+
+The v6 guide improved fun, but still left two gaps.
+
+First, it did not force patch discipline strongly enough. A writer could mark a few weak questions and then rewrite far more than needed. Future guide versions must require `fix_marked_ids_only` unless a mass rewrite exception is documented.
+
+Second, it did not clearly separate fun from random. A pack can avoid therapy voice but drift into snack overload, fake award overload, or childish nonsense. Future daily packs must stay playful and adult, not random for its own sake.
+
+## Added In v7
+
+* Patch mode is required by default.
+* Passing questions must stay frozen.
+* Mass rewrites require a documented over-60-percent shared failure.
+* Every marked question needs a fix scope.
+* Fun but grounded is now a gate.
+* The option answer test is now required.
+* Remaining hard flags must be 0 before production-ready.