You asked
How should we prepare everything before the first turn?
Answered counted as answeredAsked with: Love Letterbg-129622-en-1e
1. Shuffle the 16 cards to form a face-down draw deck [1]. 2. Remove the top card of the deck from the game without looking at it [1]. 3. If playing a two-player game, take three more cards from the top of the deck and place them to the side, face up; these will not be used during this round [1]. 4. Each player draws one card from the deck to serve as their secret hand [1].
🔧 debug — where this ruling came from
- outcome
- ANSWERED — Answered
- corpus profile
- live
- answerer pin (deploy-frozen, attribution only)
- gemma4:26b think:false 5571076f3d70
- timings
- search 204 ms · ask 2740 ms · total 35394 ms
- single-flight (PR-5)
- queued 32439 ms · generation 2955 ms
- retrieval refs (raw)
- [{"page": 9, "score": 0.032787, "snippet": "Shuffle the 16 cards to form a face-down draw deck. Remove the top card of the deck from the game without looking at it. If you are playing a two-player game, take three more cards from the top of the deck and place them to the side, face up. They will not be used during this round. Each player draw", "headingPath": "setup", "sharpsignalDocId": 396}, {"page": 11, "score": 0.032258, "snippet": "See page 15 for the effects of each individual card. Likewise, if you have any questions about special cases regarding the card, you'll find the answer there. All discarded cards remain in front of the player who discarded them. Overlap the cards so that it's clear in which order they were discarded", "headingPath": "taking a turn", "sharpsignalDocId": 396}, {"page": 12, "score": 0.031746, "snippet": "Once you finish applying the card's effect, the turn passes to the player on your left. If a player is knocked out of the round, that player discards the card in his or her hand face up (do not apply the card's effect) and takes no more turns until next round. A player could cheat when chosen with t", "headingPath": "taking a turn · out of the round · honesty", "sharpsignalDocId": 396}, {"page": 10, "score": 0.03125, "snippet": "Love Letter is played in a series of rounds. Each round represents one day. At the end of each round, one player's letter reaches Princess Annette, and she reads it. When she reads enough letters from one suitor, she becomes enamored and grants that suitor permission to court her. That player wins t", "headingPath": "how to play · taking a turn", "sharpsignalDocId": 396}, {"page": 19, "score": 0.030769, "snippet": "When you discard Prince Arnaud, choose one player still in the round (including yourself). That player discards his or her hand (do not apply its effect) and draws a new card. If the deck is empty, that player draws the card that was removed at the start of the round. If all other players are protec", "headingPath": "5: prince arnaud · 4: handmaid susannah", "sharpsignalDocId": 396}, {"page": 14, "score": 0.030303, "snippet": "The winner receives a token of affection. Shuffle all 16 cards together, and play a new round following all of the setup rules. The winner of the previous round goes first, because the princess speaks kindly of him or her at breakfast. A player wins the game after winning a number of tokens based on", "headingPath": "end of a round · winning", "sharpsignalDocId": 396}, {"page": 13, "score": 0.029851, "snippet": "A round ends if the deck is empty at the end of a turn. The royal residence closes for the evening, the person closest to the princess delivers the love letter, and Princess Annette retires to her chambers to read it. All players still in the round reveal their hands. The player with the highest ran", "headingPath": "end of a round", "sharpsignalDocId": 396}, {"page": 20, "score": 0.029412, "snippet": "When you discard the Handmaid, you are immune to the effects of other players' cards until the start of your next turn. If all players other than the player whose turn it is are protected by Susannah, the player must choose him- or herself if possible. The scion of an esteemed house that has long be", "headingPath": "4: handmaid susannah · 3: baron talus", "sharpsignalDocId": 396}]
- citations (raw)
- [{"page": 9, "quote": null, "headingPath": "setup", "sharpsignalDocId": 396}]
- boost provenance — errata + base rulebook (E2/E3)
-
no registered errata card matched this query; base rulebook doc 396 boosted; rulesage.errata.boost-enabled=true, rulesage.errata.supersedence-enabled=true
base rulebook doc 396 boosted
- answered-before recall
-
this ruling was answered, so past rulings were not offered
similarity threshold in force: 0.55 (rulesage.answered-before.similarity-threshold; the exact-question tier is not gated by it)
- debug row retained until
- Nov 7, 2026 05:01
retrieval — what the search arm returned
score = the fused retrieval score (reciprocal-rank fusion over the search arms) — NOT a cosine similarity. This ruling ran the THREE-ARM fuse (full-text + vector + the doc-priority boost arm, which carried the base rulebook and/or the errata corrections this query matched), whose arithmetic ceiling is 0.04918 — all three arms ranking the same chunk #1 (ceiling = arms/(k+1) at k=60). The measured healthy band 0.016–0.033 was measured on TWO-arm asks and does not bound this one: its high end is the two-arm ceiling, and a third arm has room above it. A chunk only one arm ranked highly lands near 0.016 (that one arm's #1 vote), so 0.03 here is an excellent hit, not a 3% match.
per-arm ranks (the rank each search arm gave a chunk): NOT DISCLOSED by the gateway. Fusion happens gateway-side and only the FUSED score crosses the wire, so rulesage cannot show them without a gateway change. The rank column below is the FUSED rank (position in the returned list), which is real.
token count per chunk: NOT DISCLOSED — chunk token counts are not reported on the ask path. The character counts below are of the bounded snippet/quote rulesage actually received — a real measurement of what we hold, never an estimate of the chunk.
| fused rank | document | heading | page | fused RRF score | snippet chars |
|---|---|---|---|---|---|
| #1 | RULEBOOK Love Letter — Rulebook (corpus doc 396) | setup | p.9 | 0.032787 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one)) | 300 |
| #2 | RULEBOOK Love Letter — Rulebook (corpus doc 396) | taking a turn | p.11 | 0.032258 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one)) | 300 |
| #3 | RULEBOOK Love Letter — Rulebook (corpus doc 396) | taking a turn · out of the round · honesty | p.12 | 0.031746 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one)) | 300 |
| #4 | RULEBOOK Love Letter — Rulebook (corpus doc 396) | how to play · taking a turn | p.10 | 0.03125 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one)) | 300 |
| #5 | RULEBOOK Love Letter — Rulebook (corpus doc 396) | 5: prince arnaud · 4: handmaid susannah | p.19 | 0.030769 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one)) | 300 |
| #6 | RULEBOOK Love Letter — Rulebook (corpus doc 396) | end of a round · winning | p.14 | 0.030303 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one)) | 300 |
| #7 | RULEBOOK Love Letter — Rulebook (corpus doc 396) | end of a round | p.13 | 0.029851 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one)) | 300 |
| #8 | RULEBOOK Love Letter — Rulebook (corpus doc 396) | 4: handmaid susannah · 3: baron talus | p.20 | 0.029412 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one)) | 300 |
chosen chunks — what the answer actually cited
token count per chunk: NOT DISCLOSED — chunk token counts are not reported on the ask path. The character counts below are of the bounded snippet/quote rulesage actually received — a real measurement of what we hold, never an estimate of the chunk.
| # | document | heading | page | quote chars |
|---|---|---|---|---|
| [1] | RULEBOOK Love Letter — Rulebook (corpus doc 396) | setup | p.9 | no quote on the wire |
generation — timing and the path taken
- timing split
- retrieval 204 ms · first delta 2176 ms · total 35394 ms (retrieval and first-delta both measured from the START of gateway work, so first-delta INCLUDES retrieval — do not subtract)
- full breakdown
- queued 32439 ms · retrieval 204 ms · ask 2740 ms · gateway work 2955 ms · total 35394 ms
- tokens in (the prompt the answerer read)
- 1237
- tokens out (what the answerer generated)
- 102
- tok/sec
- 136.5 tok/sec — 102 output tokens in 0.747 s of generation
- live seat (read from the running child)
- answering — gemma4:26b think:false · num_ctx 32768 (live-verified 32s ago)
- stream path
- streamed — the gateway pushed at least one accumulated draft
- watchdog
- first-token watchdog: 30s per arming window (rulesage.table.first-token-watchdog-seconds), re-armed once by the passages frame — so the worst case before a reader falls back to polling is 60s. Whether it actually fired is a BROWSER-side decision that never reaches the server, so it is not recorded here. The stream path above is the server's half of that story: whether a delta was emitted at all.
- num_ctx · think posture
- both are the CHILD's settings, not rulesage's, and no ask reply carries them — an answer never states which context window or think posture produced it. The live-seat line above reads them from the running child instead, where the gateway can answer for itself; it describes the child NOW, not this ask.
disposition
- outcome
- ANSWERED — Answered
- abstain rationale (n/a — this ruling answered)
- answered — no abstain rationale applies
setup-shape — why the checklist band did or didn't render
- decision
- NOT_SETUP_SHAPED — no setup-shape pattern matched this question
errata registry — what the gate above had to match against
this game registers NO errata cards (edition 325), so "no registered errata card matched" above could never have matched one
Errata scope: the gate above matched this ruling's own edition only. When an ask also reads an expansion, that expansion's own registered errata cards stay outside the match — a stated phase-1 limitation, not a failed lookup.