Ruling — Stonespine Architects

You asked

How do I play?

Answered counted as answered

Asked with: Stonespine Architectsbg-387514-en-1e

Stonespine Architects is played over four rounds, called years.[1] Each year is divided into three phases: Construction, Improvement, and Cleanup.[1]

Setup

1. Place the tracking board in the center of the table with the 2-5 player side face up.[2]

2. Each player chooses a color and collects two matching tracking markers.[2]

3. Randomly place one of each player's tracking markers on the priority track from left to right; the leftmost player has the highest priority.[2]

4. Place each player's other marker on the gold track according to their priority: Priority 1 gets Gold 0, Priority 2-3 gets Gold 1, and Priority 4-5 gets Gold 2.[2]

5. Shuffle the goal cards, reveal one, and place it on the tracking board.[2]

6. Shuffle the challenge cards and reveal a number of cards based on player count: 3 cards for 1-2 players, 4 cards for 3 players, 5 cards for 4 players, and 6 cards for 5 players.[2]

7. Shuffle the market cards and reveal a number of cards based on player count: 2 cards for 1-3 players, 4 cards for 4 players, and 8 cards for 5 players.[2]

8. Sort market tokens into piles by shape (circle, oval, and square), shuffle each pile, and place them next to the tracking board.[2]

9. Reveal market tokens for each space on the market cards and place them face up.[2]

10. Assemble dungeon frames and place one in front of each player.[3]

11. Shuffle blueprint cards and deal one face up to each player.[3]

12. Give each player two doorway markers matching their color and place one on their dungeon entrance as specified by their blueprint.[3]

13. Give each player two reference cards.[3]

14. Shuffle chamber cards and deal each player a hand of five cards.[3]

Construction Phase

The Construction Phase consists of four turns.[1] During each turn:

1. Players simultaneously select one chamber card from their hand and place it face up in their dungeon.[1] (Variant: players may choose to place cards face down and reveal them simultaneously).[1]

2. If playing with two players, each player also discards one chamber card from their hand.[1]

3. Players pass their hand of remaining chamber cards to the player on their left or right, depending on the current year:

* Year 1: Clockwise.[1]

* Year 2: Counterclockwise.[1]

* Year 3: Clockwise.[1]

* Year 4: Counterclockwise.[1]

4. If playing with two players, each player also draws one chamber card from the deck and adds it to their hand.[1]

5. Repeat steps 1-2 until players have placed four cards in their dungeon, completing a single row.[1]

6. Each player discards the remaining card from their hand.[1]

*Note: In Year 1, cards are placed below the dungeon frame; in Years 2-4, cards are placed to cover the bottom portion of the card in the row directly above.[1] *

Improvement Phase

This phase consists of two steps:

1. Count Gold: Players simultaneously total their gold for the current year by adding the gold values on all chamber cards placed in the current row and adding one gold for each treasure chest icon in any chamber of their dungeon.[4] Players move their tracking marker on the gold track to this total (up to 30).[4] Players then move their tracking marker on the priority track to the space directly below the track, maintaining order.[4]

2. Visit the Market: Players take turns either buying market tokens or passing, starting with the player who has the most gold.[4]

* Buy Tokens: A player chooses one set of market tokens on a single market card and reduces their gold by the cost of the set, then places the tokens on any combination of chambers in their dungeon.[4] The player with the most gold takes the next turn.[4]

* Pass: If a player cannot or chooses not to buy, they move their gold tracking marker to 0, select a revealed challenge card to place near their dungeon, and move their priority tracking marker to the leftmost unoccupied space on the priority track.[5] In Year 4, players do not select challenge cards when passing.[5]

* This continues until all players have passed.[5]

Cleanup Phase

The Cleanup Phase prepares for the next year:

1. Discard any remaining market tokens on market cards.[5]

2. Discard all

[1] RULEBOOK Stonespine Architects — Rulebook · p.7 · How to Play v · I. CONSTRUCTION PHASE · Placing Chamber Cards

[2] RULEBOOK Stonespine Architects — Rulebook · p.4 · Game Setup v

[3] RULEBOOK Stonespine Architects — Rulebook · p.5 · Game Setup v

[4] RULEBOOK Stonespine Architects — Rulebook · p.8 · II. IMPROVEMENT PHASE · STEP 1: COUNT GOLD · Breaking Ties · STEP 2: VISIT THE MARKET

[5] RULEBOOK Stonespine Architects — Rulebook · p.9 · STEP 2: VISIT THE MARKET · ELEMENTS · SECRET PASSAGE TOKENS · III. CLEANUP

One rule clearer, zero feelings hurt.

🔧 debug — where this ruling came from

outcome: ANSWERED — Answered
corpus profile: live,vps
answerer pin (deploy-frozen, attribution only): gemma4:26b think:false 5571076f3d70
timings: search 110 ms · ask 20042 ms · total 25434 ms
single-flight (PR-5): queued 5280 ms · generation 20154 ms
retrieval refs (raw) — what the PASSAGE SEARCH returned, not what the answerer read: [{"page": 12, "score": 0.047875, "snippet": "Queen Amalia has sent her most trusted advisor, Jael (JAY-el), to test Hortgully's students, as she is looking for talent to help design reinforcements for the defense of Sabek, the capital city. In solo play, you compete against Jael to prove your talent to the queen's advisor! Jael collects chambe", "headingPath": "Solo Play v · SOLO - SETUP CHANGES", "sharpsignalDocId": 1053}, {"page": 7, "score": 0.047643, "snippet": "Stonespine Architects is played over four rounds - called years. During a year, each player will place a single row of four chamber cards. At the end of the game, each player will have created a dungeon of 16 chambers in a 4x4 grid. Each year is divided into three phases: Construction, Improvement, ", "headingPath": "How to Play v · I. CONSTRUCTION PHASE · Placing Chamber Cards", "sharpsignalDocId": 1053}, {"page": 10, "score": 0.044796, "snippet": "At the end of year 4, each player places their remaining doorway marker at their dungeon exit, as specified at the bottom of their blueprint card. Then, using the scorepad, evaluate each player's dungeon by totaling their reputation in the six categories below. Final priority represents how much gol", "headingPath": "Final Scoring v · FINAL PRIORITY · GOAL CARD · CHALLENGE CARDS · CHAMBER REPUTATION · BLUEPRINT", "sharpsignalDocId": 1053}, {"page": 4, "score": 0.032787, "snippet": "1 Place the tracking board in the center of the table with the (2-5) player side of the board face up. 2 Have each player choose a color and collect the two matching tracking markers . Return all unused tracking markers to the game box. 3 Randomly place one of each player's tracking markers on the p", "headingPath": "Game Setup v", "sharpsignalDocId": 1053}, {"page": 5, "score": 0.032258, "snippet": "10 Assemble the dungeon frames and place one in front of each player with either side face up. Return all unused dungeon frame pieces to the game box. 11 Shuffle the blueprint cards and deal one face up to each player. Return all unused blueprint cards to the game box. 12 Give each player the two do", "headingPath": "Game Setup v", "sharpsignalDocId": 1053}, {"page": 9, "score": 0.030769, "snippet": "Pass: If a player chooses not to buy tokens or cannot afford tokens, they must pass. When passing, they perform the following steps: B Move their tracking marker on the gold track to the 0 space. Unspent gold is not carried over to the next year. Select one of the revealed challenge cards and place ", "headingPath": "STEP 2: VISIT THE MARKET · ELEMENTS · SECRET PASSAGE TOKENS · III. CLEANUP", "sharpsignalDocId": 1053}, {"page": 11, "score": 0.030303, "snippet": "Each player examines the paths in their dungeon and gains reputation for paths that connect to either their dungeon entrance or exit. To connect to an entrance or exit, there must be a door on the chamber card (or secret passage token) adjacent to the doorway marker. Gain 1 reputation for each chamb", "headingPath": "DUNGEON PATHS", "sharpsignalDocId": 1053}, {"page": 13, "score": 0.029851, "snippet": "Reveal the top card of the Jael deck and place it on the designated space of the tracking board or on top of any previously revealed Jael cards. Place a chamber card from your hand into your dungeon. Choose one card in your hand to keep for the next turn. Of the remaining chamber cards in your hand,", "headingPath": "I. Construction", "sharpsignalDocId": 1053}]
citations (raw) — what the answer cited, chosen from a retrieval this page cannot show: [{"page": 7, "quote": null, "headingPath": "How to Play v · I. CONSTRUCTION PHASE · Placing Chamber Cards", "sharpsignalDocId": 1053}, {"page": 4, "quote": null, "headingPath": "Game Setup v", "sharpsignalDocId": 1053}, {"page": 5, "quote": null, "headingPath": "Game Setup v", "sharpsignalDocId": 1053}, {"page": 8, "quote": null, "headingPath": "II. IMPROVEMENT PHASE · STEP 1: COUNT GOLD · Breaking Ties · STEP 2: VISIT THE MARKET", "sharpsignalDocId": 1053}, {"page": 9, "quote": null, "headingPath": "STEP 2: VISIT THE MARKET · ELEMENTS · SECRET PASSAGE TOKENS · III. CLEANUP", "sharpsignalDocId": 1053}]
boost provenance — errata + base rulebook (E2/E3): no registered errata card matched this query; base rulebook doc 1053 boosted; rulesage.errata.boost-enabled=true, rulesage.errata.supersedence-enabled=true
base rulebook doc 1053 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)
answer fence: passed — letter density 0.89855 of 3440 non-space chars (floor 0.45); 0 glyph-index tokens (max 2)
debug row recorded: Sep 7, 2026 03:00 UTC

retrieval — what the passage search returned

what this block IS: the PASSAGE SEARCH's own results — the same search that fills this page's closest-passages list — and NOT the passages the answerer read. Answering runs its OWN retrieval: a far wider candidate pool, re-ranked down to a final handful, plus floors this search never applies — the base rulebook, surviving errata corrections, and one slot per book on a multi-book read. Each of those armed only if THIS ask sent what it needs: the boost block names the ids it forwarded, and the scope line names the books it read. A document the answer CITES that no row here names is therefore EXPECTED, not a defect: two searches, two candidate pools, two orders. The answerer's own final rows are shown in their own block below, beside these and never mixed into them.

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 Stonespine Architects — Rulebook (corpus doc 1053)	Solo Play v · SOLO - SETUP CHANGES	p.12	0.047875 (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 Stonespine Architects — Rulebook (corpus doc 1053)	How to Play v · I. CONSTRUCTION PHASE · Placing Chamber Cards	p.7	0.047643 (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 Stonespine Architects — Rulebook (corpus doc 1053)	Final Scoring v · FINAL PRIORITY · GOAL CARD · CHALLENGE CARDS · CHAMBER REPUTATION · BLUEPRINT	p.10	0.044796 (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 Stonespine Architects — Rulebook (corpus doc 1053)	Game Setup v	p.4	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
#5	RULEBOOK Stonespine Architects — Rulebook (corpus doc 1053)	Game Setup v	p.5	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
#6	RULEBOOK Stonespine Architects — Rulebook (corpus doc 1053)	STEP 2: VISIT THE MARKET · ELEMENTS · SECRET PASSAGE TOKENS · III. CLEANUP	p.9	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
#7	RULEBOOK Stonespine Architects — Rulebook (corpus doc 1053)	DUNGEON PATHS	p.11	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
#8	RULEBOOK Stonespine Architects — Rulebook (corpus doc 1053)	I. Construction	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

what the answerer read — the passages it actually had in front of it

what this block IS: the passages the ANSWERING read put in front of the answerer, in the order it read them — the other half of the two reads described above, and the one that decides what the answer could possibly say. Each row names the book it came from and where in that book it sits, what it scored when the search arms were fused, which arm carried it there, and whether the re-ranker chose it or a floor guaranteed it a slot. A row here that no passage-search row names is EXPECTED: a far wider pool, re-ranked, with floors the passage search never applies.

candidate pool: the answering read scored 17 passages, out of a ceiling of 80 set gateway-side. Both figures are reported by the gateway for THIS ask — measurements of what actually happened, not constants read out of anyone's source and printed here on faith.

floors armed for this ask: the base rulebook — armed · surviving errata corrections — not armed · one slot per book on a multi-book read — not armed. Every floor is stated on every ask, armed or not, so a floor that did not apply here is never mistaken for one this app does not know about.

arms fused for this ask: vector + full-text + the doc-priority boost arm. Stated separately from the rows because an arm that ran and carried nothing is a different fact from an arm that never ran.

read at	document	heading	page	fused rank	fused RRF score	carried by	how it got its slot
#1	RULEBOOK Stonespine Architects — Rulebook (corpus doc 1053)	How to Play v · I. CONSTRUCTION PHASE · Placing Chamber Cards	p.7	#2	0.047643 (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))	vector + full-text + the doc-priority boost arm	the re-ranker chose it
#2	RULEBOOK Stonespine Architects — Rulebook (corpus doc 1053)	Game Setup v	p.5	#5	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))	vector + the doc-priority boost arm	the re-ranker chose it
#3	RULEBOOK Stonespine Architects — Rulebook (corpus doc 1053)	DUNGEON PATHS	p.11	#7	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))	vector + the doc-priority boost arm	the re-ranker chose it
#4	RULEBOOK Stonespine Architects — Rulebook (corpus doc 1053)	II. IMPROVEMENT PHASE · STEP 1: COUNT GOLD · Breaking Ties · STEP 2: VISIT THE MARKET	p.8	#12	0.014493 (below 0.016, the low end of the measured band — a weak fuse (that band was taken on two-arm asks; only its low end, one arm's #1 vote, carries over to a 3-arm one))	vector	the re-ranker chose it
#5	RULEBOOK Stonespine Architects — Rulebook (corpus doc 1053)	STEP 2: VISIT THE MARKET · ELEMENTS · SECRET PASSAGE TOKENS · III. CLEANUP	p.9	#6	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))	vector + the doc-priority boost arm	the re-ranker chose it
#6	RULEBOOK Stonespine Architects — Rulebook (corpus doc 1053)	I. Construction	p.13	#8	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))	vector + the doc-priority boost arm	the re-ranker chose it
#7	RULEBOOK Stonespine Architects — Rulebook (corpus doc 1053)	Solo Play v · SOLO - SETUP CHANGES	p.12	#1	0.047875 (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))	vector + full-text + the doc-priority boost arm	the re-ranker chose it
#8	RULEBOOK Stonespine Architects — Rulebook (corpus doc 1053)	Game Setup v	p.4	#4	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))	vector + the doc-priority boost arm	the re-ranker chose it

chosen chunks — what the answer actually cited

#	document	heading	page	quote chars
[1]	RULEBOOK Stonespine Architects — Rulebook (corpus doc 1053)	How to Play v · I. CONSTRUCTION PHASE · Placing Chamber Cards	p.7	no quote on the wire
[2]	RULEBOOK Stonespine Architects — Rulebook (corpus doc 1053)	Game Setup v	p.4	no quote on the wire
[3]	RULEBOOK Stonespine Architects — Rulebook (corpus doc 1053)	Game Setup v	p.5	no quote on the wire
[4]	RULEBOOK Stonespine Architects — Rulebook (corpus doc 1053)	II. IMPROVEMENT PHASE · STEP 1: COUNT GOLD · Breaking Ties · STEP 2: VISIT THE MARKET	p.8	no quote on the wire
[5]	RULEBOOK Stonespine Architects — Rulebook (corpus doc 1053)	STEP 2: VISIT THE MARKET · ELEMENTS · SECRET PASSAGE TOKENS · III. CLEANUP	p.9	no quote on the wire

generation — timing and the path taken

timing split: retrieval 110 ms · first delta 1494 ms · total 25434 ms (retrieval and first-delta both measured from the START of gateway work, so first-delta INCLUDES retrieval — do not subtract)
full breakdown: queued 5280 ms · retrieval 110 ms · ask 20042 ms · gateway work 20154 ms · total 25434 ms
tokens in (the prompt the answerer read): 2880
tokens out (what the answerer generated): 1024
tok/sec: 55.4 tok/sec — 1024 output tokens in 18.498 s of generation
live seat (read from the running child): answering — gemma4:26b think:false · num_ctx 32768 (live-verified 40s ago)
answerer host (measured by rulesage): reachable (for 24h, checked 2s ago)
answerer node (the seat proxy's own header — this probe, not this ask): node1
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 983), 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.