Ruling — Codenames

You asked

How do I set up the game?

Answered counted as answered

1. Organize players into 2 teams of roughly the same size, with at least two players on each team [1]. 2. Each team chooses 1 player to be their spymaster, while the rest are operatives [1]. 3. The red and blue spymasters sit on the same side of the table, opposite their own operatives [1]. 4. Mix up the word cards, choose 25 at random, and lay them out in a 5 × 5 grid [1]. 5. The spymasters choose the key randomly and slide it into the stand between them (any side can be up) [1]. 6. Put the 8 red tiles in front of the red spymaster and the 8 blue tiles in front of the blue spymaster [1]. 7. Place the 7 bystanders and 1 assassin in between [1]. 8. Identify the starting team using the four triangles on the edge of the key card [1]. 9. The starting team receives the double agent tile, which you must flip to their color and add to their tiles [1].

[1] RULEBOOK Codenames — Rulebook · p.2 · SPLIT INTO 2 TEAMS. · CHOOSE SPYMASTERS AND OPERATIVES. · LAY OUT 25 RANDOM WORDS. · SPYMASTERS DRAW A RANDOM KEY CARD. · SPYMASTERS TAKE THE TILES. · GIVE THE DOUBLE AGENT TO THE STARTING TEAM.

🔧 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 190 ms · ask 4551 ms · total 4753 ms
single-flight (PR-5): queued 0 ms · generation 4753 ms
retrieval refs (raw): [{"page": 8, "score": 0.032266, "snippet": "Be patient. The spymaster's job is hard. Finding the first clue of the game may take a couple minutes. A/fter that, things usually start to flow. You are allowed to give a clue for only 1 word , but clues for more words are more fun. Don't give any extra information with your clue. Don't say, 'This ", "headingPath": "REMINDERS AND TIPS", "sharpsignalDocId": 117}, {"page": 2, "score": 0.016393, "snippet": "Organize players into 2 teams of roughly the same size, with at least two players on each team. Each team chooses 1 player to be their spymaster. The rest are operatives. The red and blue spymasters sit on the same side of the table, opposite their own operatives. Mix up the word cards, choose 25 at", "headingPath": "SPLIT INTO 2 TEAMS. · CHOOSE SPYMASTERS AND OPERATIVES. · LAY OUT 25 RANDOM WORDS. · SPYMASTERS DRAW A RANDOM KEY CARD. · SPYMASTERS TAKE THE TILES. · GIVE THE DOUBLE AGENT TO THE STARTING TEAM.", "sharpsignalDocId": 117}, {"page": 4, "score": 0.016129, "snippet": "The operatives should take a moment to discuss the clue. (Or, if you are the only operative, you may think out loud.) Then, one of the operatives must make a guess by touching one of the cards . The spymaster covers the word with a tile matching the color shown on the key. If you guess a word in you", "headingPath": "THE OPERATIVES MAY MAKE MULTIPLE GUESSES.", "sharpsignalDocId": 117}, {"page": 8, "score": 0.015625, "snippet": "I've personally spent the past few years working on the app with our team, filling it with tons of content, cool twists, new ways to play (I am especially proud of the system of daily challenges), and interesting features - you'd be surprised just how many! :) ' Iʼm writing this because I genuinely ", "headingPath": "REMINDERS AND TIPS", "sharpsignalDocId": 117}, {"page": 5, "score": 0.015385, "snippet": "Three things can go wrong: If you guess an innocent bystander, your turn ends . If your turn does not end with a wrong guess, you can end it voluntarily. For example, if you are given the clue cold: 3 , you want to find 3 words related to cold and then end your turn. Or, if you find only 2 good word", "headingPath": "A WRONG GUESS ENDS YOUR TURN. · YOU CHOOSE TO END YOUR TURN. · YOU USE THE 'PLUS ONE' RULE. · WHEN YOUR TURN ENDS, IT IS THE OTHER TEAM'S TURN.", "sharpsignalDocId": 117}, {"page": 7, "score": 0.015152, "snippet": "Instead of a number, you can say your clue is unlimited. For example, tiny: unlimited . This allows your operatives to make as many guesses as they want, as long as they keep guessing right. Operatives do not know how many words were intended by the unlimited clue, but sometimes they can figure it o", "headingPath": "UNLIMITED CLUE · PENALTY FOR INVALID CLUE · VARIANTS · CODENAMES", "sharpsignalDocId": 117}, {"page": 6, "score": 0.014925, "snippet": "Spymasters must give clues that are in the spirit of the game. If you aren't sure your clue is valid, consult with the opposing spymaster. (Quietly, so the operatives can't hear.) Some clues are invalid because they violate the spirit of the game. Your clue must be about the meaning of the words. Y ", "headingPath": "VALID CLUES · SPIRIT OF THE GAME · DON'T BE TOO STRICT", "sharpsignalDocId": 117}, {"page": 3, "score": 0.014706, "snippet": "BLUE OPERATIVES 1 key card (from a deck of 40) 1 assassin 8 blue agents BLUE SPYMASTER ONLY SPYMASTERS SEE THE KEY! The key card shows the secret identities of the 25 words. APPLE is red. IRON is an innocent bystander. DEGREE is blue. HORN is the assassin. Heart is Blue, Iron is Red, Tube n ocent, a", "headingPath": "GIVE THE DOUBLE AGENT TO THE STARTING TEAM.", "sharpsignalDocId": 117}]
citations (raw): [{"page": 2, "quote": null, "headingPath": "SPLIT INTO 2 TEAMS. · CHOOSE SPYMASTERS AND OPERATIVES. · LAY OUT 25 RANDOM WORDS. · SPYMASTERS DRAW A RANDOM KEY CARD. · SPYMASTERS TAKE THE TILES. · GIVE THE DOUBLE AGENT TO THE STARTING TEAM.", "sharpsignalDocId": 117}]
boost provenance — errata + base rulebook (E2/E3): no registered errata card matched this query; base-rulebook boost: not recorded (this ruling predates it); rulesage.errata.boost-enabled=true, rulesage.errata.supersedence-enabled=true
base-rulebook boost: not recorded (this ruling predates it)
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: Oct 31, 2026 22:30

retrieval — what the search arm returned

score = the fused retrieval score (reciprocal-rank fusion over the search arms) — NOT a cosine similarity. How many arms fused this ruling was NOT RECORDED (it predates the boost trail, ran on the path that stored none, or stored one that could not be read), and the ceiling depends on that count (ceiling = arms/(k+1) at k=60): 2 arms top out at 0.03279; 3 arms top out at 0.04918. So a score above 0.03279 here is not an anomaly: it is arithmetic proof a boost arm voted, and the measured band 0.016–0.033 (taken on two-arm asks) is what does not apply. 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 Codenames — Rulebook (corpus doc 117)	REMINDERS AND TIPS	p.8	0.032266 (inside the measured healthy band 0.016–0.033 (measured on two-arm asks; this ruling's arm count was never recorded))	300
#2	RULEBOOK Codenames — Rulebook (corpus doc 117)	SPLIT INTO 2 TEAMS. · CHOOSE SPYMASTERS AND OPERATIVES. · LAY OUT 25 RANDOM WORDS. · SPYMASTERS DRAW A RANDOM KEY CARD. · SPYMASTERS TAKE THE TILES. · GIVE THE DOUBLE AGENT TO THE STARTING TEAM.	p.2	0.016393 (inside the measured healthy band 0.016–0.033 (measured on two-arm asks; this ruling's arm count was never recorded))	300
#3	RULEBOOK Codenames — Rulebook (corpus doc 117)	THE OPERATIVES MAY MAKE MULTIPLE GUESSES.	p.4	0.016129 (inside the measured healthy band 0.016–0.033 (measured on two-arm asks; this ruling's arm count was never recorded))	300
#4	RULEBOOK Codenames — Rulebook (corpus doc 117)	REMINDERS AND TIPS	p.8	0.015625 (below the measured healthy band (< 0.016) — a weak fuse on a two-arm ask, and this ruling's arm count was never recorded)	300
#5	RULEBOOK Codenames — Rulebook (corpus doc 117)	A WRONG GUESS ENDS YOUR TURN. · YOU CHOOSE TO END YOUR TURN. · YOU USE THE 'PLUS ONE' RULE. · WHEN YOUR TURN ENDS, IT IS THE OTHER TEAM'S TURN.	p.5	0.015385 (below the measured healthy band (< 0.016) — a weak fuse on a two-arm ask, and this ruling's arm count was never recorded)	300
#6	RULEBOOK Codenames — Rulebook (corpus doc 117)	UNLIMITED CLUE · PENALTY FOR INVALID CLUE · VARIANTS · CODENAMES	p.7	0.015152 (below the measured healthy band (< 0.016) — a weak fuse on a two-arm ask, and this ruling's arm count was never recorded)	300
#7	RULEBOOK Codenames — Rulebook (corpus doc 117)	VALID CLUES · SPIRIT OF THE GAME · DON'T BE TOO STRICT	p.6	0.014925 (below the measured healthy band (< 0.016) — a weak fuse on a two-arm ask, and this ruling's arm count was never recorded)	300
#8	RULEBOOK Codenames — Rulebook (corpus doc 117)	GIVE THE DOUBLE AGENT TO THE STARTING TEAM.	p.3	0.014706 (below the measured healthy band (< 0.016) — a weak fuse on a two-arm ask, and this ruling's arm count was never recorded)	300

chosen chunks — what the answer actually cited

#	document	heading	page	quote chars
[1]	RULEBOOK Codenames — Rulebook (corpus doc 117)	SPLIT INTO 2 TEAMS. · CHOOSE SPYMASTERS AND OPERATIVES. · LAY OUT 25 RANDOM WORDS. · SPYMASTERS DRAW A RANDOM KEY CARD. · SPYMASTERS TAKE THE TILES. · GIVE THE DOUBLE AGENT TO THE STARTING TEAM.	p.2	no quote on the wire

generation — timing and the path taken

timing split: retrieval 190 ms · first delta 3112 ms · total 4753 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 190 ms · ask 4551 ms · gateway work 4753 ms · total 4753 ms
tokens in (the prompt the answerer read): 2783
tokens out (what the answerer generated): 235
tok/sec: 148.2 tok/sec — 235 output tokens in 1.586 s of generation
live seat (read from the running child): answering — gemma4:26b think:false · num_ctx 32768 (live-verified 36s 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: NO_APPROVED_CHECKLIST — setup-shaped, but no approved setup checklist exists for this game
pattern that matched: SETUP — "setup" / "set up"
matched words (in the normalized question): set up

errata registry — what the gate above had to match against

this game registers NO errata cards (edition 39), 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.