Ruling — Spirit Island

You asked

When does the game end?

Abstained — no grounded answer; closest passages below counted as honest abstain

Asked with: Spirit Islandbg-162886-en-1e

Answered before

This question didn't get an answer just now — but Spirit Island has been asked something like it before. The corpus may have changed since, so check the date and the answerer on each one before you rule from it.

When does the game end? →

Answered Aug 2, 2026 gemma4:26b

When does the game end? →

Answered Aug 1, 2026 gemma4:26b

When does the game end? →

Answered Jul 25, 2026 gemma4:26b

I looked through Spirit Island's rulebook and couldn't find a solid answer to that one — and I'd rather tell you so than guess at your table. Try asking it another way, or ask me about setup, turns, or how the game ends: those I know cold.

The closest rulebook passages are below — check them before ruling.

RULEBOOK Spirit Island — Rulebook · p.21 · GAME TOO HARD?

If you're getting crushed and it's no fun: For a slightly easier game, give all Spirits a bonus Growth at the end of Setup. For a moderately easier game, omit the Invaders' initial Explore at the end of Setup. For a much easier game, do both. Note: Two Blight rules can make the game extremely punish

RULEBOOK Spirit Island — Rulebook · p.10 · TIME PASSES

This is the wrap-up phase at the end of each turn. Discard: Players discard all Power Cards played this turn into their personal discard piles. Damage and Elements Clear: All Elements go away. All Damage done during the turn goes away; if you turned any pieces on their sides to note partial Damage,

RULEBOOK Spirit Island — Rulebook · p.8 · THE URGENCY OF WAR

Some Spirits are inclined to take a long time choosing Power Cards: perhaps several centuries in the case of slow-moving earth Spirits. But the threat of the Invaders forbids the luxury of infinite time. Any player may place a limit on the time available to complete the Spirit Phase. This can be don

RULEBOOK Spirit Island — Rulebook · p.8 · But in the rare event that agreement is not possible:

For decisions about/within a land: Each Spirit makes choices regarding the lands on their starting board. For other cases: The game owner/organizer makes the call.

RULEBOOK Spirit Island — Rulebook · p.20 · WHOOPS!

You suddenly realize that for the past 4 turns you've been using a Power Card on lands it can't actually target. What to do? The answer is: don't sweat it. Especially on your first play or two, you'll likely make minor mistakes here and there. They're unlikely to break anything: the game will just b

Strategy guides → — community tips, kept separate from rulings

🔧 debug — where this ruling came from

outcome

ABSTAINED — Abstained — no grounded answer; closest passages below

warm abstain

this ruling — W1 — generic warm abstain · receipt: none (generic)

corpus profile

live,vps

answerer pin (deploy-frozen, attribution only)

gemma4:26b think:false 5571076f3d70

timings

search 71 ms · ask 2880 ms · total 2954 ms

single-flight (PR-5)

queued 0 ms · generation 2954 ms

retrieval refs (raw) — what the PASSAGE SEARCH returned, not what the answerer read

[{"page": 21, "score": 0.046898, "snippet": "If you're getting crushed and it's no fun: For a slightly easier game, give all Spirits a bonus Growth at the end of Setup. For a moderately easier game, omit the Invaders' initial Explore at the end of Setup. For a much easier game, do both. Note: Two Blight rules can make the game extremely punish", "headingPath": "GAME TOO HARD?", "sharpsignalDocId": 58}, {"page": 10, "score": 0.032787, "snippet": "This is the wrap-up phase at the end of each turn. Discard: Players discard all Power Cards played this turn into their personal discard piles. Damage and Elements Clear: All Elements go away. All Damage done during the turn goes away; if you turned any pieces on their sides to note partial Damage, ", "headingPath": "TIME PASSES", "sharpsignalDocId": 58}, {"page": 8, "score": 0.032258, "snippet": "Some Spirits are inclined to take a long time choosing Power Cards: perhaps several centuries in the case of slow-moving earth Spirits. But the threat of the Invaders forbids the luxury of infinite time. Any player may place a limit on the time available to complete the Spirit Phase. This can be don", "headingPath": "THE URGENCY OF WAR", "sharpsignalDocId": 58}, {"page": 8, "score": 0.031746, "snippet": "For decisions about/within a land: Each Spirit makes choices regarding the lands on their starting board. For other cases: The game owner/organizer makes the call.", "headingPath": "But in the rare event that agreement is not possible:", "sharpsignalDocId": 58}, {"page": 20, "score": 0.03125, "snippet": "You suddenly realize that for the past 4 turns you've been using a Power Card on lands it can't actually target. What to do? The answer is: don't sweat it. Especially on your first play or two, you'll likely make minor mistakes here and there. They're unlikely to break anything: the game will just b", "headingPath": "WHOOPS!", "sharpsignalDocId": 58}, {"page": 13, "score": 0.030303, "snippet": "Spirit Presence marks the lands a Spirit occupies. Lands with your Presence are sometimes referred to as 'your lands'. If your Presence is ever destroyed (for example, by Invaders Blighting the land), it is removed and put next to the Island. If any Spirit ever has no Presence left on the Island, th", "headingPath": "PRESENCE AND SACRED SITES", "sharpsignalDocId": 58}, {"page": 13, "score": 0.029851, "snippet": "The game is played with one Island Board per player, laid out to make an Island (see Setup, page 6). Each Island Board is divided into eight numbered lands, with exactly two of each terrain (Jungle, Mountain, Sands, and Wetland). Most pieces only affect other pieces in the same land, unless explicit", "headingPath": "BOARDS AND LANDS", "sharpsignalDocId": 58}, {"page": 22, "score": 0.029412, "snippet": "Adversaries are specific colonizing Powers from the world of Spirit Island. Including an Adversary in a game is optional, but doing so adds an additional layer of depth and strategy to the game. Be sure to choose your foe before Setup begins, as some may change the rules of Setup. The Adversary Pane", "headingPath": "ADVERSARIES", "sharpsignalDocId": 58}]

citations (raw) — what the answer cited, chosen from a retrieval this page cannot show

[]

boost provenance — errata + base rulebook (E2/E3)

no registered errata card matched this query; base rulebook doc 58 boosted; rulesage.errata.boost-enabled=true, rulesage.errata.supersedence-enabled=true

base rulebook doc 58 boosted

answered-before recall

past answered rulings were found and offered

similarity threshold in force: 0.55 (rulesage.answered-before.similarity-threshold; the exact-question tier is not gated by it)

ruling 657 — matched by the same question (normalized, punctuation folded) — measured similarity 1.0
ruling 602 — matched by the same question (normalized, punctuation folded) — measured similarity 1.0
ruling 420 — matched by the same question (normalized, punctuation folded) — measured similarity 1.0

debug row retained until

Nov 12, 2026 07:23

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 Spirit Island — Rulebook (corpus doc 58)	GAME TOO HARD?	p.21	0.046898 (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 Spirit Island — Rulebook (corpus doc 58)	TIME PASSES	p.10	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
#3	RULEBOOK Spirit Island — Rulebook (corpus doc 58)	THE URGENCY OF WAR	p.8	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
#4	RULEBOOK Spirit Island — Rulebook (corpus doc 58)	But in the rare event that agreement is not possible:	p.8	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))	163
#5	RULEBOOK Spirit Island — Rulebook (corpus doc 58)	WHOOPS!	p.20	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
#6	RULEBOOK Spirit Island — Rulebook (corpus doc 58)	PRESENCE AND SACRED SITES	p.13	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 Spirit Island — Rulebook (corpus doc 58)	BOARDS AND LANDS	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 Spirit Island — Rulebook (corpus doc 58)	ADVERSARIES	p.22	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

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 80 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 Spirit Island — Rulebook (corpus doc 58)	GAME TOO HARD?	p.21	#1	0.046898 (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 Spirit Island — Rulebook (corpus doc 58)	WHOOPS!	p.20	#5	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))	vector + the doc-priority boost arm	the re-ranker chose it
#3	RULEBOOK Spirit Island — Rulebook (corpus doc 58)	STREAMLINING GAMEPLAY	p.20	#9	0.029206 (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 re-ranker chose it
#4	RULEBOOK Spirit Island — Rulebook (corpus doc 58)	THE KINGDOM OF BRANDENBURG-PRUSSIA	p.22	#15	0.013699 (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 Spirit Island — Rulebook (corpus doc 58)	ADVERSARIES	p.22	#8	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))	vector + the doc-priority boost arm	the re-ranker chose it
#6	RULEBOOK Spirit Island — Rulebook (corpus doc 58)	STAGE II ESCALATION	p.10	#11	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
#7	RULEBOOK Spirit Island — Rulebook (corpus doc 58)	WHO DECIDES?	p.8	#20	0.012658 (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
#8	RULEBOOK Spirit Island — Rulebook (corpus doc 58)	TIME PASSES	p.10	#2	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

generation — timing and the path taken

timing split: retrieval 71 ms · first delta 2874 ms · total 2954 ms (retrieval and first-delta both measured from the START of gateway work, so first-delta INCLUDES retrieval — do not subtract)
full breakdown: queued 0 ms · retrieval 71 ms · ask 2880 ms · gateway work 2954 ms · total 2954 ms
tokens in (the prompt the answerer read): 1546
tokens out (what the answerer generated): 9
tok/sec: 155.2 tok/sec — 9 output tokens in 0.058 s of generation
live seat (read from the running child): answering — gemma4:26b think:false · num_ctx 32768 (live-verified 12s 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: ABSTAINED — Abstained — no grounded answer; closest passages below
abstain rationale: I don't have enough information.
the answerer's own words, verbatim — hostile corpus/LLM text, rendered escaped

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 7), 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.