Ruling — Spyfall

You asked

How do I play?

Answered counted as answered

Asked with: Spyfallbg-166384-en-1e

To play the game, follow these steps:

1. Agree upon the number of rounds before the beginning of the game.[1]

2. Sort all cards into location decks, with each deck containing seven cards with identical illustrations and one spy card.[2]

3. Place each deck in a separate ziplock bag with the spy card on top and all cards face down.[2]

4. Choose a dealer for the round (in the first round, the most suspicious-looking player is the dealer).[1]

5. The dealer takes all card ziplock bags out of the box, turns them face down, shuffles them, and chooses a random bag.[1]

6. The dealer carefully removes the cards from the selected bag and selects a number of cards from the top of the stack that corresponds to the number of players.[1]

7. The dealer shuffles these cards and gives one to each player.[1]

8. Each player looks at their card without letting others see it, then places it face down in front of them.[1]

9. The dealer starts the stopwatch.[1]

10. The dealer asks one player a question by name, which usually pertains to the deck's location.[1]

11. The player who answers the question proceeds to ask any other player a question of their own, excluding the person who asked the previous question.[1]

12. The order of questions is established by the players based on their suspicions.[1]

13. A round ends when one of the following occurs:

* Eight minutes have passed, at which point players declare suspicions and vote.[3]

* Any player stops the game to declare a suspect and asks for a vote (if all players vote for the same suspect, the round ends).[4]

* The spy stops the game by revealing their spy card to identify the location.[4]

[1] RULEBOOK Spyfall — Rulebook · p.3 · Beginning of the Round · Spy Card · Game Flow

[2] RULEBOOK Spyfall — Rulebook · p.2 · Rulebook · Contents · Overview · Objective · Preparing for Round One

[3] RULEBOOK Spyfall — Rulebook · p.6 · End of Round · 1. Eight minutes have passed;

[4] RULEBOOK Spyfall — Rulebook · p.7 · 1. Eight minutes have passed; · 2. When a Player Gets Suspicious; · 3. At the Spy's Request.

The sage rests. The game continues.

🔧 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 94 ms · ask 3051 ms · total 36580 ms
single-flight (PR-5): queued 33433 ms · generation 3147 ms
retrieval refs (raw) — what the PASSAGE SEARCH returned, not what the answerer read: [{"page": 3, "score": 0.048916, "snippet": "Each game is a series of short rounds, the number of which is agreed upon by the the players before the beginning of the game. We recommend playing five rounds for your first session (this should take about an hour). A new dealer is chosen for each round. The dealer participates in the game as a reg", "headingPath": "Beginning of the Round · Spy Card · Game Flow", "sharpsignalDocId": 1027}, {"page": 7, "score": 0.048131, "snippet": "Note: The players can agree on a different round length. Make sure you decide how long you want rounds to be before you start the game. First-time players might want to play longer rounds - say, 12-15 minutes. Any player can stop the game at any time (but only once per round) to declare one of the p", "headingPath": "1.\tEight minutes have passed; · 2.\t When a Player Gets Suspicious; · 3.\t At the Spy's Request.", "sharpsignalDocId": 1027}, {"page": 6, "score": 0.047643, "snippet": "Note: if you are not the spy, your card will also specify your status in the given location. Before the beginning of the game, the players should agree on whether they will adhere to the status on their card and play the corresponding role or not. We strongly recommend doing this in order to make th", "headingPath": "Game Flow · Sample Round", "sharpsignalDocId": 1027}, {"page": 2, "score": 0.031746, "snippet": "'Loose lips sink ships' is a saying that dates back to the days of WWII. It also represents a vital wartime principle. There were lots of things that could give spies away - rust-free staples in their documents, square-headed nails in the soles of their boots, and so on... This game offers everyone ", "headingPath": "Rulebook · Contents · Overview · Objective · Preparing for Round One", "sharpsignalDocId": 1027}, {"page": 8, "score": 0.030769, "snippet": "The spy's objective is to listen as carefully as possible to what the other players say and do their best to avoid blowing their cover while also trying to identify the location before eight minutes have passed. A spy who waits until the end of the round is taking a risk - it is entirely possible th", "headingPath": "Objectives and Strategies · Scoring · If the Spy Wins · If the Non-Spies Win · End of the Game · Credits · …", "sharpsignalDocId": 1027}, {"page": 7, "score": 0.030303, "snippet": "The objectives of the non-spy players are to identify the spy and avoid revealing their location. Therefore, the non-spies should refrain from being too explicit in their questions: (for example, a bank teller asks a security guard, 'How much cash did the robbers steal yesterday?' The spy will then ", "headingPath": "Objectives and Strategies", "sharpsignalDocId": 1027}, {"page": 6, "score": 0.029851, "snippet": "A round ends when one of the following three things happen. When time runs out, the players must declare their suspicions as to the spy's identity. The dealer asks each player to vote, starting with themselves and proceeding clockwise. If need be, the players can discuss their suspicions and then vo", "headingPath": "End of Round · 1.\tEight minutes have passed;", "sharpsignalDocId": 1027}]
citations (raw) — what the answer cited, chosen from a retrieval this page cannot show: [{"page": 3, "quote": null, "headingPath": "Beginning of the Round · Spy Card · Game Flow", "sharpsignalDocId": 1027}, {"page": 2, "quote": null, "headingPath": "Rulebook · Contents · Overview · Objective · Preparing for Round One", "sharpsignalDocId": 1027}, {"page": 6, "quote": null, "headingPath": "End of Round · 1.\tEight minutes have passed;", "sharpsignalDocId": 1027}, {"page": 7, "quote": null, "headingPath": "1.\tEight minutes have passed; · 2.\t When a Player Gets Suspicious; · 3.\t At the Spy's Request.", "sharpsignalDocId": 1027}]
boost provenance — errata + base rulebook (E2/E3): no registered errata card matched this query; base rulebook doc 1027 boosted; rulesage.errata.boost-enabled=true, rulesage.errata.supersedence-enabled=true
base rulebook doc 1027 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 0.91722 of 1353 non-space chars (floor 0.45); 0 glyph-index tokens (max 2)
debug row recorded: Sep 7, 2026 02:57 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 Spyfall — Rulebook (corpus doc 1027)	Beginning of the Round · Spy Card · Game Flow	p.3	0.048916 (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 Spyfall — Rulebook (corpus doc 1027)	1. Eight minutes have passed; · 2. When a Player Gets Suspicious; · 3. At the Spy's Request.	p.7	0.048131 (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 Spyfall — Rulebook (corpus doc 1027)	Game Flow · Sample Round	p.6	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
#4	RULEBOOK Spyfall — Rulebook (corpus doc 1027)	Rulebook · Contents · Overview · Objective · Preparing for Round One	p.2	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
#5	RULEBOOK Spyfall — Rulebook (corpus doc 1027)	Objectives and Strategies · Scoring · If the Spy Wins · If the Non-Spies Win · End of the Game · Credits · …	p.8	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 Spyfall — Rulebook (corpus doc 1027)	Objectives and Strategies	p.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))	300
#7	RULEBOOK Spyfall — Rulebook (corpus doc 1027)	End of Round · 1. Eight minutes have passed;	p.6	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 7 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 Spyfall — Rulebook (corpus doc 1027)	Beginning of the Round · Spy Card · Game Flow	p.3	#1	0.048916 (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 Spyfall — Rulebook (corpus doc 1027)	Objectives and Strategies	p.7	#6	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
#3	RULEBOOK Spyfall — Rulebook (corpus doc 1027)	End of Round · 1. Eight minutes have passed;	p.6	#7	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
#4	RULEBOOK Spyfall — Rulebook (corpus doc 1027)	Rulebook · Contents · Overview · Objective · Preparing for Round One	p.2	#4	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))	vector + the doc-priority boost arm	the re-ranker chose it
#5	RULEBOOK Spyfall — Rulebook (corpus doc 1027)	Objectives and Strategies · Scoring · If the Spy Wins · If the Non-Spies Win · End of the Game · Credits · …	p.8	#5	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 Spyfall — Rulebook (corpus doc 1027)	1. Eight minutes have passed; · 2. When a Player Gets Suspicious; · 3. At the Spy's Request.	p.7	#2	0.048131 (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
#7	RULEBOOK Spyfall — Rulebook (corpus doc 1027)	Game Flow · Sample Round	p.6	#3	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

chosen chunks — what the answer actually cited

#	document	heading	page	quote chars
[1]	RULEBOOK Spyfall — Rulebook (corpus doc 1027)	Beginning of the Round · Spy Card · Game Flow	p.3	no quote on the wire
[2]	RULEBOOK Spyfall — Rulebook (corpus doc 1027)	Rulebook · Contents · Overview · Objective · Preparing for Round One	p.2	no quote on the wire
[3]	RULEBOOK Spyfall — Rulebook (corpus doc 1027)	End of Round · 1. Eight minutes have passed;	p.6	no quote on the wire
[4]	RULEBOOK Spyfall — Rulebook (corpus doc 1027)	1. Eight minutes have passed; · 2. When a Player Gets Suspicious; · 3. At the Spy's Request.	p.7	no quote on the wire

generation — timing and the path taken

timing split: retrieval 94 ms · first delta 1091 ms · total 36580 ms (retrieval and first-delta both measured from the START of gateway work, so first-delta INCLUDES retrieval — do not subtract)
full breakdown: queued 33433 ms · retrieval 94 ms · ask 3051 ms · gateway work 3147 ms · total 36580 ms
tokens in (the prompt the answerer read): 2731
tokens out (what the answerer generated): 400
tok/sec: 200.1 tok/sec — 400 output tokens in 1.999 s of generation
live seat (read from the running child): answering — gemma4:26b think:false · num_ctx 32768 (live-verified 37s ago)
answerer host (measured by rulesage): reachable (for 22h, checked 7s 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 957), 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.