You asked
When does the game end?
Abstained — no grounded answer; closest passages below counted as honest abstainAsked with: Spirit Island: Branch & Clawbg-191831-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.
The answerer couldn't ground a confident answer in the rulebook. The closest rulebook passages are below — check them before ruling.
In the Core Game, the Invaders are perfectly predictable, save for what lands they will next be interested in. But in truth, all living things - the Invaders, the Dahan, and Beasts of the wild - sometimes act in unanticipated ways. To represent this, each Invader Phase, after any Blighted Island eff
Some Fear and Event Cards refer to a 'normal' Invader Action - for instance, 'Skip the normal Build'. 'Normal' just means 'the Actions caused by Invader Cards in the normal 3 Invader Action Spaces', even if there are multiple Invader Cards in an Action Space. By contrast, Invader Actions taken due t
If you are playing with the Scenario 'Blitz', the Choice Event 'Outpaced' is guaranteed to be irrelevant, as there is no Slow Phase. If this happens - or if future Adversaries/Scenarios make a particular Event irrelevant or senseless - discard that Event Card and draw a replacement.
DEMORALIZED Defend 3 in all lands. Defend 2 in all lands. Defend 1 in all lands. 2 per player Any removed from the board returns here. If there is ever NO here, players lose. BACK AGAINST THE WALL Every Spirit Phase each Spirit gains +1 Energy and +1 Card Play. BLIGHTED ISLAND 31 MINOR POWER CARDS 2
K D Y CARDS K ED Y CARDS ARD ARD PER PLAYER Some Fear and Event Cards may change the usual orderly progression of Invader Cards along the Invader Action Track, causing there to be multiple - or zero - cards on an Action Space. If there are no cards on an Action Space, you don't perform it, just like
Strategy guides → — community tips, kept separate from rulings
🔧 debug — where this ruling came from
- outcome
- ABSTAINED — Abstained — no grounded answer; closest passages below
- corpus profile
- live
- answerer pin (deploy-frozen, attribution only)
- gemma4:26b think:false 5571076f3d70
- timings
- search 267 ms · ask 6403 ms · total 6678 ms
- single-flight (PR-5)
- queued 0 ms · generation 6678 ms
- retrieval refs (raw)
- [{"page": 8, "score": 0.030214, "snippet": "In the Core Game, the Invaders are perfectly predictable, save for what lands they will next be interested in. But in truth, all living things - the Invaders, the Dahan, and Beasts of the wild - sometimes act in unanticipated ways. To represent this, each Invader Phase, after any Blighted Island eff", "headingPath": "EVENT CARDS · BLIGHTED ISLAND", "sharpsignalDocId": 220}, {"page": 15, "score": 0.028158, "snippet": "Some Fear and Event Cards refer to a 'normal' Invader Action - for instance, 'Skip the normal Build'. 'Normal' just means 'the Actions caused by Invader Cards in the normal 3 Invader Action Spaces', even if there are multiple Invader Cards in an Action Space. By contrast, Invader Actions taken due t", "headingPath": "'NORMAL' ACTIONS VS. 'EXTRA' ACTIONS · INVADERS DOING DAMAGE MULTIPLE TIMES IN SAME LAND · 'PICK TWO' GROWTH · GROWTH THAT COSTS ENERGY", "sharpsignalDocId": 220}, {"page": 9, "score": 0.016393, "snippet": "If you are playing with the Scenario 'Blitz', the Choice Event 'Outpaced' is guaranteed to be irrelevant, as there is no Slow Phase. If this happens - or if future Adversaries/Scenarios make a particular Event irrelevant or senseless - discard that Event Card and draw a replacement.", "headingPath": "MEANINGLESS CHOICES", "sharpsignalDocId": 220}, {"page": 5, "score": 0.016129, "snippet": "DEMORALIZED Defend 3 in all lands. Defend 2 in all lands. Defend 1 in all lands. 2 per player Any removed from the board returns here. If there is ever NO here, players lose. BACK AGAINST THE WALL Every Spirit Phase each Spirit gains +1 Energy and +1 Card Play. BLIGHTED ISLAND 31 MINOR POWER CARDS 2", "headingPath": "", "sharpsignalDocId": 220}, {"page": 14, "score": 0.015873, "snippet": "K D Y CARDS K ED Y CARDS ARD ARD PER PLAYER Some Fear and Event Cards may change the usual orderly progression of Invader Cards along the Invader Action Track, causing there to be multiple - or zero - cards on an Action Space. If there are no cards on an Action Space, you don't perform it, just like", "headingPath": "STRIFE EXAMPLES · MULTIPLE/MISSING INVADER CARDS", "sharpsignalDocId": 220}, {"page": 22, "score": 0.015625, "snippet": "Placed under an individual Invader (in . target land), unlike other tokens. When an Invader with Strife deals . Damage to Dahan and/or the land, that Damage is blocked, and the . Strife is then Removed. [p. 11] Token Event: The portion of an Event Card with an effect centering . around Beasts, Disea", "headingPath": "GLOSSARY/INDEX", "sharpsignalDocId": 220}, {"page": 12, "score": 0.015385, "snippet": "For these examples, you have drawn an Event Card with 'Beasts Attack!'. During this event, each deals 2 Damage. Remove any token that Destroys a Town or City. This is a land with 1 and 1 Dahan. This land has no Invaders, so nothing happens here. This is a land with 1 . It has an Explorer and a Town,", "headingPath": "STRIFE TIMING · DISEASE EXAMPLES", "sharpsignalDocId": 220}, {"page": 9, "score": 0.015152, "snippet": "Players make any necessary decisions. When an Event says, 'On Each Board: (do this OR that)', you may select differently on each board. Y ou are permitted to read the entire Event Card before making any decisions, but you don't have to if your playgroup would rather think about one thing at a time. ", "headingPath": "BLIGHTED ISLAND · CHOICE EVENTS", "sharpsignalDocId": 220}]
- citations (raw)
- []
- boost provenance — errata + base rulebook (E2/E3)
-
no registered errata card matched this query; no base-rulebook boost — this ask EXCLUDED the base book, so its rulebook was never resolved and no id for it went on the wire (boosting it would have put that book back into a scope the reader took it out of); rulesage.errata.boost-enabled=true, rulesage.errata.supersedence-enabled=true
no base-rulebook boost — this ask EXCLUDED the base book, so its rulebook was never resolved and no id for it went on the wire (boosting it would have put that book back into a scope the reader took it out of)
- layering — route
- not layered — one pooled read, because one book was in scope and there is nothing to split
- 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 3, 2026 20:45
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 TWO-ARM fuse (full-text + vector), whose arithmetic ceiling is 0.03279 — both arms ranking the same chunk #1 (ceiling = arms/(k+1) at k=60). Measured healthy band on this fuse: 0.016–0.033. 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: Branch & Claw — Rulebook (corpus doc 220) | EVENT CARDS · BLIGHTED ISLAND | p.8 | 0.030214 (inside the measured healthy band 0.016–0.033) | 300 |
| #2 | RULEBOOK Spirit Island: Branch & Claw — Rulebook (corpus doc 220) | 'NORMAL' ACTIONS VS. 'EXTRA' ACTIONS · INVADERS DOING DAMAGE MULTIPLE TIMES IN SAME LAND · 'PICK TWO' GROWTH · GROWTH THAT COSTS ENERGY | p.15 | 0.028158 (inside the measured healthy band 0.016–0.033) | 300 |
| #3 | RULEBOOK Spirit Island: Branch & Claw — Rulebook (corpus doc 220) | MEANINGLESS CHOICES | p.9 | 0.016393 (inside the measured healthy band 0.016–0.033) | 283 |
| #4 | RULEBOOK Spirit Island: Branch & Claw — Rulebook (corpus doc 220) | p.5 | 0.016129 (inside the measured healthy band 0.016–0.033) | 300 | |
| #5 | RULEBOOK Spirit Island: Branch & Claw — Rulebook (corpus doc 220) | STRIFE EXAMPLES · MULTIPLE/MISSING INVADER CARDS | p.14 | 0.015873 (below the measured healthy band (< 0.016) — a weak fuse) | 300 |
| #6 | RULEBOOK Spirit Island: Branch & Claw — Rulebook (corpus doc 220) | GLOSSARY/INDEX | p.22 | 0.015625 (below the measured healthy band (< 0.016) — a weak fuse) | 300 |
| #7 | RULEBOOK Spirit Island: Branch & Claw — Rulebook (corpus doc 220) | STRIFE TIMING · DISEASE EXAMPLES | p.12 | 0.015385 (below the measured healthy band (< 0.016) — a weak fuse) | 300 |
| #8 | RULEBOOK Spirit Island: Branch & Claw — Rulebook (corpus doc 220) | BLIGHTED ISLAND · CHOICE EVENTS | p.9 | 0.015152 (below the measured healthy band (< 0.016) — a weak fuse) | 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 |
|---|
generation — timing and the path taken
- timing split
- retrieval 267 ms · first delta 6589 ms · total 6678 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 267 ms · ask 6403 ms · gateway work 6678 ms · total 6678 ms
- tokens in (the prompt the answerer read)
- 1095
- tokens out (what the answerer generated)
- 9
- tok/sec
- 152.5 tok/sec — 9 output tokens in 0.059 s of generation
- live seat (read from the running child)
- answering — gemma4:26b think:false · num_ctx 32768 (live-verified 38s 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.