v0.8.0.12 — put a player back in their seat after losing their browser storage

Gitea#33. A session token is the only identity the game has, and it lives in
exactly one place the player controls: their browser's localStorage, scoped to
the origin they joined at. Lose it — a cleared profile, a private window, a
different browser — and the seat is unreachable while the game runs on and the
session sits intact on disk. Reported from the table: of two humans in one game
the host reloaded straight back in, the joiner met an empty lobby.

Diagnosed before it was fixed, and two server-side theories of mine were
retracted on the evidence: no storage key changed in 0.8.0.11, nothing in the
app deletes the secret or name, create and join both call persistSession, that
game's sessions.json held both seats, and it resumed with 80 intents replayed.
Both players used the same URL, so it was not a second origin either.

The fix is a recovery link. An administrator mints a code for a named seat
(admin-gated: deciding somebody lost a seat is a judgement no route can make);
the player opens the link and the page trades the code for the token over a
POST, then strips it from the address bar. The link never carries the token —
lobby-and-sessions.md §1 says keep it out of URLs, and a recovery link is
exactly what gets pasted into a chat. Single use, 30-minute expiry, held in
memory because a restart dropping them is the right failure.

server/claims.ts is a pure store, so single use, lazy expiry and one identical
answer for unknown/spent/expired codes are tested rather than asserted. The
admin game listing gained seatedPlayers — the seats a human holds a token for,
read from the session map rather than guessed from player names — so the
StartOS action can offer real players instead of bot chairs.

No rule changed: `git diff v0.8.0.11..v0.8.0.12 -- src/engine/` is empty, so
games in progress resume.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017nnuCv8UodHucFfx3LWEoX

CHANGELOG.md

@@ -19,6 +19,674 @@ page as `v0.1.0 · <sha> · <date>`, so what is deployed can always be identifie
 
 ---
 
+## 0.8.0.12 — 2026-09-16
+
+A player who has lost their browser storage can be put back in their seat (Gitea#33). No rule
+changed: `git diff v0.8.0.11..v0.8.0.12 -- src/engine/` is empty, so games in progress resume.
+
+### The failure this fixes, and the four things it was not
+
+Reported from the table after the 0.8.0.11 update: of two humans in one game, the host reloaded
+straight back into it and the player who had JOINED found an empty lobby — no join secret, no display
+name, no game code. Their seat was never lost. `sessions.json` for that game held both seats, and the
+server logged it resuming with 80 intents replayed.
+
+Four explanations were ruled out with evidence before any code was written, and two of them were
+theories of mine that had to be retracted:
+
+- **Not the update.** `git diff v0.8.0.10..v0.8.0.11 -- src/web/` contains no storage change at all;
+  both tags declare identical `SECRET_KEY` and `NAME_KEY`.
+- **Not a create-vs-join asymmetry in the client.** `lb-secret` and `lb-name` sit above both doors in
+  `play.html`, so a joiner writes the same three keys a host does.
+- **Not the server forgetting joiners.** Create calls `persistSession` and so does join; it writes
+  every session for the game. Two-seat session files plainly work.
+- **Not a second origin.** Both players used the identical URL.
+
+What is left is the thing §1 has always said: the token lives in one browser's `localStorage`, scoped
+to the origin. A cleared profile, a private window or a different browser ends the seat while the game
+runs on without it. Nothing in the client can detect that — origin isolation is the point — and
+nothing in it can repair it either.
+
+### A recovery link carries a code, never the token
+
+`lobby-and-sessions.md` §1: *"Keep it out of URLs so it is not shoulder-surfed or pasted into a
+chat."* A recovery link is precisely what gets pasted into a chat, so the URL carries a **single-use
+code that expires in 30 minutes** and the page trades it for the real token over a POST, then strips
+it from the address bar. A spent code is worth nothing; a token in a chat log is the seat for the rest
+of the game.
+
+```
+POST /api/games/<id>/claim  { player }   →  { code, expiresAt, … }   admin secret
+POST /api/claim             { code }     →  { token, gameId, player, gameCode }
+```
+
+**Minting is administrative; spending is not.** Deciding that a particular person has lost a
+particular seat is a judgement no route can make safely — anyone able to mint their own code could
+take any chair at the table. Spending needs no secret because the player following the link is the one
+person in the story who holds none: the code *is* the authorisation, unguessable and one-time, which
+is the same shape as the token it returns.
+
+`server/claims.ts` is a pure store — no clock, no sockets, no disk — so its rules are actually tested
+rather than asserted: single use, lazy expiry, and one identical answer for unknown, spent and expired
+codes so it cannot be probed. The codes are held in memory on purpose. They are minted on demand and
+spent within minutes with the administrator present, so a restart dropping them is the right failure;
+persisting them would put a credential-equivalent on disk to solve a problem measured in seconds.
+
+### The administrator picks a seat, not a string
+
+The admin game listing now reports `seatedPlayers` — the seats a HUMAN holds a token for, read from
+the server's session map rather than guessed by matching "Bot 1" against a display name. That is what
+lets the StartOS side offer real players to choose from instead of chairs no token was ever issued
+for.
+
+## 0.8.0.11 — 2026-09-16
+
+Fourteen reports from the second multiplayer playtest of v0.8.0.10, the WHISTLE-6945 table. Eleven are
+fixed here; the review of the New Train phase turned into four changes of its own; and two were
+questions the engine had already answered.
+
+### A train's arrival says WHOSE Office it reached
+
+The line went to every seat reading "Train 3 ARRIVED at the Whistle Post … **You** can work it in Cargo
+now". At a table of four that is one true sentence and one false one: every seat has an Office, so the
+tier alone never said which district the train was standing in, and three of the four readers could not
+touch it. `trainArrived` now carries `owner`, and the narration resolves the name — "ARRIVED at Tom's
+Whistle Post … Tom can work it". The Mainline Phase has no actor, which is why nothing upstream could
+name the seat for it. Saves replay history rather than storing events, so the new field strands nothing.
+
+### Being five behind now looks five behind everywhere
+
+*"When player Jesse is 5 behind, it should always look like he's 5 behind."* The board, the district
+panel and the caption row have followed the animation queue since v0.8.0; the turn chart never did. So a
+player watching three bots play out a Stage saw their cards moving under a chart that had already ticked
+over to the next phase — the one part of the screen quietly insisting the game was elsewhere. The Day,
+Stage, clock, phase and Fedora now come from the step on screen (`shownTable`), as does the district
+panel's auto-open. Being behind is fine and the counter says so; being behind on *part* of the screen is
+what made it unreadable.
+
+### Pause, beside Skip
+
+Skip was the only control the catching-up row had, and it is one-way and total: the way to look harder
+at a move that had just gone past was to not be too slow about it. Pause is the opposite lever. The
+property that matters is not that it stops — any flag gives you that — but that a hold **costs the step
+nothing**: a move paused half-way through its dwell resumes with half a dwell left, rather than being
+thrown away the moment you let go. Skip lifts a hold rather than leaving a Resume button that does
+nothing.
+
+### A one-render look at another player's Office Area
+
+One button per opponent in the district header. Deliberately not a mode: the look survives exactly the
+render its own click causes, so the panel is back to following whoever is acting the next time anything
+redraws. Jesse's call, and the right one — a sticky pin has to answer what happens when the game moves
+on beneath it, and both honest answers are bad (snap home silently, or leave a player studying a stale
+board while the game waits on them). Pause already means "hold everything", so that is the lever for a
+long look and this stays a glance. Player names reach the page through `textContent`, never markup.
+
+### The Office Area summary counts the district on screen
+
+The panel has drawn somebody else's board since v0.8.0 and the heading says whose — but the summary
+line read the viewer's own frame every time. So a bot's turn showed "Bot 2's Office Area" over Bot 2's
+cards, above a line counting *your* cards, facilities and trains.
+
+### LIMITS is printed beneath the card, not through it
+
+*"The text is split by the bottom border of the limits card."* It was: the bottom row's edge lands at
+`height - 7` and the label's baseline was `height - 4`, so its 8px glyphs spanned `height - 12` to
+`height - 4` and the border ran through the middle of the word. Raising it would have put it on the
+card, over the rails, so a 14px band is added below the bottom row and the baseline moved into it.
+
+### The Mainline region divider is visible
+
+Drawn at `#4a5361`, 1.2px, on a `#161b21` card — faint enough not to be made out at all. It is the ruler
+the train is measured against and should stay quieter than the rail, but a ruler you cannot read is not
+restraint.
+
+### Only Hilly talks about FAST/SLOW
+
+Every other Mainline card ended its description with "the printed speed is scenery", which sent a player
+hunting the card for a number that is not drawn on it. Exactly one card reads the rating — Hilly, the
+only profile with `speedStarts` — and it already explains itself. Everywhere else now says nothing,
+which is the honest answer when the rating does not apply.
+
+### A passenger Modifier on a Whistle Post says it is dormant
+
+A Restaurant appeared to do nothing. It does nothing: a Whistle Post is not a Passenger Facility, so it
+allows neither direction and has zero capacity each way; `usableGrant` discards the capacity while the
+porter is granted regardless, leaving a porter with nothing to carry. The Freight Agent then refuses to
+stock the box twice over, so no coach can ever be there to board. Playing it there stays legal on
+Jesse's call — the card is not wasted, it starts working on upgrade — and the panel now says so. It had
+been claiming the facility "only receives", which is wrong in both directions at a Whistle Post.
+
+### An automatic phase says what it is doing
+
+"waiting on **nobody — the Division is running itself**" reached its answer by negation, on the one line
+whose job is to say where the game is. The phase's name is already printed directly above it, so these
+describe the work instead: "the Division is moving trains", "the Division is building this Stage's
+trains". Only a person is waited on.
+
+### The version appears once in the header
+
+The no-git fallback for the cache-bust key led with the version, and the stamp already begins with
+`v${version}` — so the header read "v0.8.0.10 · 0.8.0.10-mfq2p1 · …". It showed up on exactly the builds
+that take that path, which is **every `.s9pk`**, because the Dockerfile copies the tree in without
+`.git`. The uniqueness that fallback exists to provide comes from the timestamp, not the version, so the
+two requirements never competed. Pinned in `test/web.test.ts` beside the constant-fallback assertion
+that the same function already carries.
+
+### Save files name the game, the Stage and the day
+
+Every server game downloaded as `station-master-day1-stage5.json`, so two saves off different tables
+collided in the downloads folder and neither said which table it came from. The join code is what a
+player already says out loud to identify a game, so it leads: `whistle-6945.day1.stage5.2026.09.16.json`.
+A solitaire game has no code and falls back to its seed.
+
+### The New Train phase says what is left, and why your turn ended
+
+Reviewed before changing anything, at Jesse's request — and the phase turned out to be better served
+than it looked from a grep: the Division Yard chip IS the button, the panel already names the train
+and what its card calls for, warns about assembly order on the trains where order can lock switching,
+and carries the "no more cars" button that exists to prevent a real softlock. Four things were missing.
+
+**Still needs.** The heading says what the card CALLS FOR and goes on saying it unchanged as cars go
+on, so the one question a player actually has while clicking — what is left? — was the only thing on
+screen that had to be worked out by eye, against a consist drawn in the other column. `consistNeeds`
+counts by the same categories `acceptsCar` does, so it can never ask for a car the engine would then
+refuse, and it reads null when the consist is complete.
+
+**One car each.** §7's round — a single car, then the next player, from the Superintendent working
+left — was written down only in the turn chart's New Train chip tooltip: hovered once, early on, and
+never again. "Click a car" then reads as "build this train", and the turn ends with no explanation.
+It now says so where the clicking happens.
+
+**The train is marked on the map.** `TrainChip` gained `beingMadeUp`, set from the engine's own
+`isBeingMadeUp`, and the chip is drawn in the action amber — so the panel on the right and the train
+on the Division strip at the top left are visibly one subject rather than two.
+
+**A clickable car looks clickable.** The addable highlight was a thin blue outline, reported as "just
+a small bold and basically the same color as everything else". It now wears `#c8912f`, the border
+colour the action buttons themselves use. The loaded/empty text colours are deliberately left alone:
+green and blue-grey are what say which of the two numbers is which, and that is a different question
+from whether you may click.
+
+### Asked, and already answered by the engine
+
+**Does the bot know how to use Interlocking?** It plays it deliberately — the bot ranks it above the
+other enhancements precisely because it stops an Office collision — and a Running Track straight is the
+card's printed placement, so the play seen at the table was correct. Nobody "uses" it: `advance.ts`
+applies it automatically, turning an arrival at a full Office into a hold at the Limits. There is no
+decision to get wrong, so no change was made.
+
+**Can passengers load or unload at a Whistle Post?** No. Zero porters, zero outbound, zero inbound, and
+not a Passenger Facility — it takes a Depot to do any passenger work at all.
+
+---
+
+## 0.8.0.10 — 2026-09-15
+
+Three reports from the first multiplayer playtest of v0.8.0.9 — one player and three bots — each traced
+to its cause before it was fixed.
+
+### The Superintendent is no longer asked about a train BEHIND the one departing (Gitea#26)
+
+*"Two westbound trains: X15 is further west on Mainline cards than X18. And I still get a Superintendent
+must rule... held X15 and then collision happened. X18 destroyed, but X15 fine."*
+
+**Reproduced by replaying the exported save.** At move 78, X15 was highballing west out of seat 3's
+Office while X18, also westbound, was still crossing the card to its EAST — behind it. Every Office was a
+Whistle Post, so the Subdivision ran the whole railroad, and `evaluateClearance` counted every train in it
+without asking which side of the departing train it stood on. The ruling was meaningless; holding X15
+kept the Whistle Post's only A/D track full, and X18 arrived into it and was destroyed. Unasked, X15 —
+the lower number, so the first to move — would have left and freed the track.
+
+§8.1 asks about a train the departing one would FOLLOW and one moving TOWARDS it, and both are ahead of
+it. So a train on a card strictly behind the departing train's own card is no longer counted, in either
+pass: a following train behind is no concern, and an oncoming one behind is moving away. A train on the
+SAME card is still counted, exactly as before — which of two trains sharing a card is in front is
+`entryConflict`'s region question. The other two rulings in the save (moves 37 and 530) were correct and
+are unchanged.
+
+A test had the fault written into it: its "train ahead" of an eastbound train was the first Mainline card
+in the Division, which is west of the Office. It now stands on a card the train would follow, and three
+new cases pin the rule: a same-direction train behind is not put to the Superintendent, an opposite-
+direction train behind does not bar the departure, and a same-direction train ahead still is. The tally
+test's seed moved from 42 to 44, because the collision seed 42 was chosen for was this bug.
+
+### Games in progress — MANY WILL NOT RESUME
+
+**This release changes when the engine asks for a ruling**, so a save holding a ruling it would no longer
+ask for stops replaying at that move: `mainline.clearance` is refused with `NO_PENDING_DECISION`. The
+server then refuses to resume that game and logs the move it stopped at, leaving the save untouched —
+putting v0.8.0.9 back would resume it. **Measured:** 40 four-seat co-op games recorded by bots under
+v0.8.0.9 and replayed under this release — 28 stop, usually 5–40% of the way in, most at a ruling on a
+train behind (about ten rulings a game were being asked). The exported playtest game stops at move 78.
+**Jesse's call (2026-09-15): no games in progress need keeping, so the fix ships as it is,** with no
+per-game switch preserving the old rule.
+
+### Six more from the same playtest (Gitea#27–#32)
+
+**A Realignment says which card it converted (#27).** *"It stated Mainline card 3 converted to plains. It
+should state that the mainline card 3 curves was converted to plains."* The event carried only what the
+card became, so the line could not name what it had been; `mainlineModified` now carries `from` as well —
+events are derived by replaying a save and never stored, so widening one strands nothing — and it reads
+"Realignment: Mainline card 3, Curves, converted to Plains".
+
+**And the map flashes it (#28).** *"Is it possible to flash the mainline card when it gets changed by
+realignment?"* The one play that changes the Division itself was invisible on the map of it.
+`changedDivisionCards` compares the two public boards the animation queue already holds — the same way a
+pile is lit — so the pulse lands with the step that shows the change rather than when the intent arrived,
+and it is shown to the player who made it too, unlike a lit pile. `prefers-reduced-motion` turns it off.
+
+**The history stops running ahead of the board (#29).** *"Does history display immediately for all bot
+turns... is it possible to stall history so it stays in sync with the number behind?"* A push carries its
+narration and its steps together, so every line of a bot's turn was in the panel before the board had
+drawn any of it. The queue now reports how many lines belong to steps not yet shown, and the panel holds
+back exactly those, revealing each as its step goes up. Skip still shows everything at once.
+
+**A ruling reads as the office's (#30).** *"If a player makes a move as a superintendent instead of as
+themselves maybe it could say 'Superintendent Player Tom'."* The three moves made by holding the office
+rather than in turn — §8.1's clearance, §11's Yard Office offer, §Q's Red Flag prompt — are prefixed that
+way. `clearanceGiven` carries no player at all (the office made it, whoever holds it), so the acting seat
+is what names it.
+
+**And it names them once (#31).** *"It gives the player's name and then their player number together."*
+`record()` prefixes the acting player's NAME for every event carrying `player`, and five narration lines
+embedded `Player <index>` themselves — phase ended, actor changed, the Red Flag ruling, the extension vote
+and the Yard Office ruling. The name is the log's job; the sentence is the narration's. A test now scans a
+played game's whole log for a bare player index.
+
+**A multiplayer game can be saved as a file (#32).** *"In the StartOS actions for the save game, I get a
+string I can copy. Most of the time, I want to just save it as a JSON file."* Not possible in the action
+itself: an action result member is text only — copyable, QR or masked, with nested groups — and the SDK
+has no file member. So it is done where it can be: the play page's "Save replay" button was shown only in
+solitaire, because a server-backed session has no local save to hand it. `GET /api/save?token=…` — the
+same seat token `/api/stream` and `/api/intent` use, not the administrative secret — hands a seated player
+their own game, and the button now appears in multiplayer.
+
+### "Waiting on" follows the move on screen (Gitea#25)
+
+*"Playing against 3 bots — waiting on always says me, even when it is someone else's turn."*
+
+The server plays every bot move the moment a human's turn ends, so the LIVE game is almost always waiting
+on the human — while the screen is still replaying those bots step by step. The turn chart and the
+Division map's move marker read the live actor, and contradicted the playback row naming the bot actually
+moving. `actorOnScreen` (`web/step-queue.ts`) answers with the player of the step on screen while the
+board is catching up, and the live actor once it has; during playback the chart also leaves off a live
+"asks … ruling" note that belongs to a position the screen has not reached. Five tests pin it.
+
+### The name on the Division map is readable again (Gitea#24)
+
+*"When it shows your player, the font is unreadable... the bold font makes it look fuzzy, and the letters
+blur together."*
+
+A name takes the class `bs-turn` while it is that player's move — and `.bs-turn` is also the Division
+map's turn ARROW, which strokes its shape 2.4px grey with no fill. The name's own rule changed only the
+fill, so every letter was outlined in grey. It showed for the viewing player almost constantly because of
+Gitea#25: the map believed it was nearly always their move. `.bs-name` now sets `stroke:none`.
+
+---
+
+## 0.8.0.9 — 2026-09-15
+
+**The developer bot, re-measured decision by decision — and an engine 2.8× faster.** Across the changes
+adopted below, each paired against the bot before it, the bot went from about −0.3 revenue a game to
+about 4.8: the switching planner +2.89, taking a face-up card only if it could be played +1.52, the
+deliberate New Train fallback +0.32, and laying track by what the district can do afterwards +0.12.
+
+### Games in progress
+
+**Resume.** No rule changed, and no legality check changes its answer. The engine's speed-ups were proven
+to leave play identical — every event and intent of 32 seeded games hashed before and after each step —
+and `test/route-cache.test.ts` pins the new check/commit split. Only the developer bot plays differently,
+and a bot's past moves are already in the save.
+
+### The bot plans its switching turn instead of choosing one Move at a time
+
+The switching branch was a ladder of rules picking ONE Move, and its own comment named the gap: "a
+strong player would use the six Moves to re-order the consist — that is the game's central switching
+puzzle, and this bot does not attempt it." `sim/switch-planner.ts` attempts it, for one turn.
+
+**Why search is fair.** A switching turn draws no card and rolls no die, so trying sequences on a copy
+of the game is what a player does by looking at the board. The score reads only what a player can see
+— the district, the cars on the trains, the facilities — and never the deck. Jesse's line
+(2026-09-14): no non-player advantage for the bot.
+
+**Why not every sequence.** Measured over 30 switching turns from bot games: a median turn reaches
+229 distinct positions, but 11 of 30 passed 20,000, because **setting cars out costs no Move** and a
+crew can leave them in a great many places. So the search keeps the best 48 positions at each step
+and stops at 3,000 tried; small turns are searched completely inside that.
+
+**The score is of where the turn ends**, starting from Jesse's ruling that players deliver and pick up
+cars even when it delays trains. A car an industry can work is +1 (+0.25 past its box count); a car it
+cannot is −0.5 on its track; a finished car left on it −0.15; a wanted car aboard +0.35 or staged on
+plain track +0.2; a coach kept with its train +0.3, stranded −0.3; anything fouling the Office −3; a
+train away from the Office −0.25, −1 more if expedited; a train §8.2 would refuse (`badlyMadeUp`, now
+exported rather than copied) −0.6. A load made in this district scores nothing at a receiver here.
+
+**The copy is partial.** `forkForSwitching` copies only the arrays a switching intent writes — cars
+standing on cards, industry tracks, the district's A/D and held lists, the trays' consists, this turn
+and the tally — and shares everything else. It began as a `structuredClone` of the district, which a
+profile put at **44% of all planning time**; the targeted copy made a position about three times
+cheaper to try. `test/switch-planner.test.ts`
+proves across seeded games that planning leaves the real game byte-identical, and that every plan
+replays through `applyIntent` on a FULL copy to the exact position it promised.
+
+**Measured, 1600 paired seeds: +2.89 ± 0.18 revenue a game (t = 15.79)**, 733 seeds better, 21 worse.
+
+| | rules | planner |
+| --- | --- | --- |
+| revenue | −0.05 | 2.83 |
+| freight loads + unloads | 0.22 | 1.53 |
+| collisions | 0.12 | 0.12 |
+| Cargo phases with a car spotted | 5% | 18% |
+| … with green box, car and Laborer at one industry | 3% | 11% |
+
+**Split by source over 400 seeds**, because the biggest wins were rescues: freight **+1.21** (t = 13.9),
+passengers +0.26, expedite faults **+1.07** (19 of 400 games faulted under the rules, none under the
+planner — TODO #53, closed by this), collisions unchanged. **Without the fault rescue it is still
++1.45 ± 0.12 (t = 12.2)** — the switching itself got better, not just the catastrophes rarer. It also
+sets out half as many cars (3.6 against 7.0): the ladder was setting cars down and picking them up.
+
+**The worst seeds, traced.** Two of the three lost to "no free A/D track" collisions with both tracks
+already held by trains standing at the Office — not trains left away, which is what the score's
+`trainAway` weight would have explained. The third never had a coach train at the Office in a
+Load/Unload phase at all: a different switching turn draws different cards, and the game diverges.
+Neither is a scoring defect found; the full-Office collisions are worth watching.
+
+**How far to search.** The search size was measured rather than guessed, paired over 400 seeds against
+the 3000-position, beam-48 search the gain above was measured with:
+
+| budget / beam | revenue against 3000/48 | median / worst per turn |
+| --- | --- | --- |
+| 3000 / 48 | — | 150 ms / 445 ms |
+| **2000 / 32 (adopted)** | −0.02 ± 0.01 (t = −1.68), inside the noise | **69 ms / 222 ms** |
+| 1000 / 24 | −0.06 ± 0.02 (t = −2.65) | 53 ms / 104 ms |
+
+Times are from 51 switching turns with other simulations sharing the CPU, so read them as relative.
+
+**A switching-only legal list.** `legal.ts` now exports `legalSwitchingActions`, the switching half of
+the Local Operations candidates run through the same `check`, in the same order — so the planner stops
+paying for every draw and Freight Agent candidate at each position it tries. A test asserts it equals
+`legalActions`' switching subset across real games. It contains no rule; `check` still decides.
+
+**Nothing the simulation tests measure moved backwards.** `sim.test.ts` passes all 35 tests with the
+planner as the default, floors unchanged.
+
+**Cost.** A turn is planned once and then played a step per decision, replanned if the position is
+ever not the one expected. At a live table that is a synchronous pause inside `driveBots`, well inside
+Jesse's bar of half a second before a switch.
+
+**The price is test time.** A standard solitaire game takes ~400 ms with the planner against ~108 ms
+without, and `npm test` — which plays well over a thousand bot games in `sim.test.ts` — went from about
+150 s to **8 min 6 s** (992 of 992 passing, measured with nothing else running). What is left of a
+planned turn's time is the engine itself: applying a move is about half of it and `check` a third.
+
+**And the take rule tripled it again.** With both defaults in, `npm test` passes 992 of 992 in
+**21 min 21 s** with nothing else running — the bigger districts the take rule builds (14 → 24 cards,
+29 cards played a game) make every game longer to play and every switching turn wider to search.
+
+`noPlanSwitching=1` is the ablation.
+
+### Track goes where it lets the district do something
+
+Jesse's third area, after switching and industries: track for the run-around. With the draw no longer
+wasted, ~13 of ~16 pieces a game were still being laid by the draw turn's fallback at the first legal
+square, and both fixes to that fallback failed (holding −0.83, `bestTrackLay`'s own score +0.07). So the
+limit was the scoring: `bestTrackLay` scores the PIECE, and cannot tell one that opens an industry site
+or closes a run-around from one that fills a square.
+
+`bestValuedLay` scores the LAYOUT the piece would leave instead. Each legal lay is placed on a copy of
+the district exactly as the reducer places it — `protoCard` and `extendLimitsIfNeeded`, now exported
+rather than copied — and worth the change it makes to `layoutValue`: industry sites a crew can reach
+(`canPlaceAt`), a closed run-around (the bot's own `descendFrom` walk), ways off the main and reachable
+siding, a Running Track straight for Interlocking, and a penalty for a turnout on the main whose leg
+joins nothing. `bestTrackLay`'s slot takes the best lay that gains something; the fallback still lays as
+often as before — holding starved the district — but the best lay rather than the first.
+
+**+0.118 ± 0.029 revenue a game (t = 4.14) over 6400 paired seeds**, 1,667 better against 1,430 worse.
+It needed that many: 400 seeds read +0.17 (t = 1.48) and 1600 read +0.14 (t = 2.58). What it builds is
+the larger change — **closed run-arounds in 22 of 60 districts against 9**, industries 1.78 → 1.87, freight
+2.54 → 2.65, collisions unchanged, the district 24.2 → 21.9 cards because squares stop being filled with
+pieces that build nothing. No slower: a standard game measured faster, the smaller districts leaving
+less to search. `noValueLays=1` is the ablation.
+
+**Why so many run-arounds buy so little.** Traced over 40 games: the planner uses the loop — a move ends
+on it in 63 of 131 plans where one exists, 35 run it both ways — but its planned gain per turn is the
+same with a run-around as without (0.28 against 0.27). A run-around is for putting a train's cars in a
+different order, which pays off in the turns AFTER; a one-turn planner cannot value it. That is TODO #105.
+
+**How 6400 seeds were measured.** `compare.ts` keeps every game's full statistics for both sides, and the
+run was stopped for low memory — as was a first lean attempt, by the machine's background-task guard
+rather than by the process: a probe showed no leak (heap flat at 9 MB after GC across 300 paired games).
+The figure above comes from a paired script with the same seeds and configuration that keeps only each
+seed's revenue difference, run in resumable foreground chunks; its first 1600 seeds read +0.143, matching
+`compare.ts`'s +0.14.
+
+### The engine walks each position's routes once — 2.8× faster, every game identical
+
+Jesse's call (2026-09-15): speed up the engine's move checking and applying first, because the live
+server runs the same `legalActions` and `applyIntent` for every bot and every player. A CPU profile
+with inlining off put the route walk (`reachableDestinations`) at a third of all time and garbage
+collection at another third, and most of the walking was repeated work.
+
+- **A route cache scoped to one unchanged position** (`withRouteCache`, `apply.ts`). `legal.ts` walked a
+  tray's routes to list its moves, `check` walked them again for every one of those moves, and
+  `execute` walked the chosen move a third time. Now a legal-action listing, and the check-and-execute
+  of one intent, walk each route once. The cache is keyed to the state object and the walk's inputs
+  and dropped before `reduce` changes anything, so a hit is exactly what a fresh walk returns.
+- **`applyIntent` = `prepareIntent` + `commitEvents`.** The planner decides every candidate against one
+  position inside one cache and commits each to its own copy; deciding on the copy had re-walked every
+  route the listing had just walked.
+- **Less garbage in the walk itself** (`exploreMoves`): the blocked-square explanations are not built
+  when only destinations are wanted, the queue is read by index instead of `shift()`, and "has this route
+  been here" reads the route's own path instead of copying a Set at every step.
+
+**A standard solitaire bot game, 561 ms → 203 ms; a short one, 141 ms → 65 ms.** Proved identical by
+hashing every event and intent of 32 seeded games (20 solo standard, 6 three-player standard, 6
+two-player short) before and after each change, and pinned by `test/route-cache.test.ts`, which checks
+at every decision of a seeded game that `prepareIntent` writes nothing and that committing its events
+to a copy equals `applyIntent` in place.
+
+### The test suite is split
+
+Jesse (2026-09-15): `npm test` runs after every change, so it has to stay fast; reducing the simulations'
+game counts is not acceptable. `npm test` now runs every file except `test/sim.test.ts`, and
+`npm run test:sim` runs the bot simulations on their own, typechecked first.
+
+**Measured with nothing else running, after the engine speed-up: `npm test` 958 of 958 in 1 min 45 s;
+`npm run test:sim` 35 of 35 in 6 min 58 s** — against 21 min 21 s for the two together before it.
+
+### Where industries go was not the problem — there was nowhere, and the draw was going in circles
+
+The next thing Jesse named after switching was where industries and enhancements go, then track for the
+run-around. Measured before touching either, 30 standard games, and it turned both round:
+
+- **The bot places an industry every time it legally can.** It held an industry card at 1,441 draw
+  decisions and a legal square existed at 24 of them (2%); it played the industry at all 24 and never
+  discarded one that had somewhere to go. 0.70 industries stand on a board at game end.
+- **Why there was nowhere.** An industry is a plain east-west piece that must join existing track and
+  may not sit on the Running Track. Asked of `check` square by square, the closest a held card got was
+  NOT_CONNECTED 1,681 times, OUTSIDE_LIMITS 398, legal 28 — never locked out. Printed districts show why:
+  the row beside the main fills with 45° curves and turnouts whose east-west ends face occupied squares.
+- **And track was not the lever either — its supply was.** At 1,334 decisions holding an industry with
+  no site, a track lay was legal at only 54; a lay that would open a site existed at 22 and the bot
+  already chose one at 17. The hand had no track in it.
+- **Because the draw was cycling.** The bot took 20.1 timetabled trains and 11.9 industries a game off
+  the Departments and discarded 20.2 and 11.8: `takingRank` ranked a face-up train 2 whether or not the
+  A/D cap would let it be played, and a face-up industry 1 whether or not it had a site (one did at 0.17
+  of those takes). The same card was taken again 28.8 times a game, against 19.6 blind draws.
+
+### The bot takes a face-up card only if it could play it
+
+`takingRank` now ranks a face-up train card 0 when the A/D cap would hold it, and a face-up industry 0
+when it is locked out or has no legal site — both asked of the rules the play itself is checked by
+(`trainWouldOverfillTheOffice`, and `isLockedOut` with `canPlaceAt` in `industrySiteExists`). Nothing
+here is hidden information: the Departments are face up.
+
+**+1.52 ± 0.10 revenue a game (t = 15.59)** over 1600 paired seeds, measured as the ablation
+`noPlayableTakes=1` against the new default: 880 seeds worse without it, 211 better. (The first
+400-seed read was +1.63 ± 0.19.) Cards played 15.8 → 29.0 a game and the district grows 14.2 → 24.2
+cards, because the draws it stopped wasting now bring in track: trains scheduled 1.74 → 2.30, freight
+1.51 → 2.54, passengers 1.89 → 2.94. **Collisions rose, 0.12 → 0.24**, with the extra trains — see
+TODO #106.
+
+### Rejected: holding track the run-around scoring declined
+
+Traced first: of ~16.4 track pieces a standard game lays, only ~3.3 came from `bestTrackLay`. The
+draw turn's "play what is in hand" fallback laid the other ~13 at the FIRST legal square — pieces
+`bestTrackLay` had just declined, including 4.0 turnouts and 4.3 straights a game on the main and 1.5
+turnouts a row off it. The candidate let the fallback lay track only when nothing else could bring the
+hand under the limit. **−0.83 ± 0.16 (t = −5.17)**, 172 worse against 90 better: the district fell
+24.1 → 11.7 cards and freight 2.41 → 1.77. `bestTrackLay` declines most pieces, so holding them
+starves the district — #59's finding, again, with a fuller hand to hold them in.
+
+### Rejected: laying fallback track where it scores best instead of first
+
+The other half of the same trace: keep laying the ~13 fallback pieces a game, but at `bestTrackLay`'s
+best-scoring square (bonus or not) rather than the first legal one. **+0.07 ± 0.13 (t = 0.60), inside
+the noise**; freight 2.41 → 2.56 but the district 24.1 → 21.0 cards. With both halves measured, the
+limit is `bestTrackLay`'s SCORING — it cannot tell a piece that opens an industry site or advances a
+run-around from one that fills a square — not which branch places the piece. Code removed.
+
+### The bot stops running Second Sections by accident
+
+TODO #106 traced under today's defaults: of 20 "no free A/D track" collisions in 60 standard games, no
+train had been held before any of them and only 8 of the destroyed were Extras. Six destroyed a train of
+the same number as a Second Section run within two Stages, and **all 26 Second Sections the bot ran were
+an accident**: when no car was on offer, the New Train phase fell back to `options[0]`, and `legalActions`
+lists `newTrain.secondSection` ahead of the Extra starts — so a waiting Extra became a doubled train due
+out, into an Office that never had an A/D track to spare. The fallback now takes a car, a pass or the
+Extra's start, and never a Second Section or a Red Flag merely because it was listed first.
+
+**+0.32 ± 0.09 revenue a game (t = 3.64)** over 400 paired seeds, 30 better against 12 worse; collisions
+0.24 → 0.19. `noDeliberateNewTrain=1` is the ablation.
+
+**Found on the way, for Jesse — a rules question, not changed.** Q9 defines the Second Section as a card
+played on a train due out and `content.ts` gives `SECOND_SECTION` one copy, but `buildDeck` never deals it
+and `check` asks for no card, so any player can run one for free on every train due out.
+
+### Rejected: planning two switching turns ahead (TODO #105)
+
+Built as `planTwoTurns`: keep the six best ends of a switching turn, remove the trains that highball in
+the Mainline Phase in between (on the Office square and made up — their cars leave with them), reset the
+Moves, plan the next turn from each, and choose by the position after departures plus 0.8 of what the
+next turn adds. **−0.28 ± 0.11 (t = −2.58)**; corrected to charge the expedite fault the gap cannot undo
+and a Stage for every train left away, **−0.14 ± 0.06 (t = −2.36)** — 400 paired seeds each. A second turn
+has little to find: only 49% of switching turns keep their train for the next, a further turn could have
+spotted just 0.7 of the 6.1 wanted cars a game that leave aboard departing trains, and trains left away
+cost Stages walking crews home. Dropped at Jesse's call; the measurements are in TODO #105.
+
+### Rejected: starting an Extra the Office cannot take where its run never arrives
+
+TODO #106, re-measured under today's defaults: 20 collisions in 60 standard games (−1.67 revenue a
+game), every one "no free A/D track", and 26 Extras forced out by a full hand of them. The bot also took
+the first legal start for EVERY Extra — the western Division Point, 158 of 158 — so the candidate chose,
+when the committed trains exceeded the A/D tracks, a start whose run never reaches the Office: an Extra
+runs away from its start (`resolveExtraStart`), and at the Interchange one way may miss the Office.
+**+0.10 ± 0.05 (t = 1.82), 395 of 400 seeds identical**: such a start was on offer at 1 of 25 over-cap
+starts — most Divisions have no Interchange, and a Division Point always runs through the Office. Code
+removed.
+
+### Rejected: refusing to draw into a forced Extra
+
+With the take rule in, the bot's worst seeds all lost to "no free A/D track". Traced: every one of 23
+train plays past the A/D cap in 40 games was an EXTRA, played because the hand held four of them and
+the only legal action was `card.play` — §6.2 makes a player over the limit reduce the hand, and an Extra
+may never be discarded (`keepReason`), so the cap in `choose` yields rather than leave nothing legal.
+The candidate declined to choose the draw option with a full hand of such cards when switching or the
+Freight Agent was on offer. **−0.03 ± 0.11 (t = −0.30), inside the noise**: collisions fell
+0.26 → 0.20, and development fell with them (cards played 29.0 → 25.6) — the Stage spent avoiding the
+draw was a Stage not spent building. Code removed; the trap itself is real and filed.
+
+### Rejected: discarding the card least likely to become playable
+
+When a discard was forced, the candidate picked WHICH card by a keep-value (next Office tier, then
+Enhancements, track, trains, and industries with no site or Modifiers with no industry last), weighted
+so no pile preference could overturn it. **−1.16 ± 0.12 (t = −9.27)**, 16 better against 119 worse,
+cards played 15.7 → 10.4. Not traced; the likeliest reason is that the weight overrode `bestDiscard`'s
+pile choice, which exists to avoid burying a face-up card the bot wants. Code removed.
+
+### Flying Switch is searched, and cannot be measured
+
+The planner now considers Flying Switch alongside Moves, set-outs and sorts — its partial copy of the
+game carries the hand and the Salvage Yard, which `spendCard` writes, and spending the card costs a
+tenth of a point so it is played only when it buys something. **Over 400 paired seeds it changed
+nothing, because the card is dealt 0 copies** (not in sheet 5; Jesse, 2026-08-26): across 30 standard
+games no Flying Switch was ever drawn. TODO #58's "never fires" is therefore a deck fact, not a bot
+one. On by default, so the planner uses the card the day it is dealt again.
+
+### Rejected: letting the planned gain decide whether to switch at all
+
+Whether a Stage goes to switching is decided by `usefulSwitching`, a yes/no reading of the district.
+The candidate replaced it with the planner's own answer — plan the turn on a fork before the option is
+chosen, and switch only if it gains at least a threshold. Paired over 400 seeds against the planner:
+
+| threshold | revenue | seeds better / worse / identical |
+| --- | --- | --- |
+| 0.5 | **−0.33 ± 0.06 (t = −5.06)** | 15 / 70 / 315 |
+| 0.25 | −0.01 ± 0.05 (t = −0.21) | 21 / 22 / 357 |
+| 0.1 | +0.06 ± 0.03 (t = 1.79) | 20 / 13 / 367 |
+
+At 0.5 it refuses turns that only COLLECT — a wanted car picked up scores +0.35 — and those pickups feed
+the deliveries after them. Below that it agrees with `usefulSwitching` almost everywhere, because that
+rule already says yes exactly when there is a car to deliver or lift, which is when a plan gains. The
+decision that would matter is switching against DRAWING or the Freight Agent, and the bot has no value
+for either to compare with; that is the next piece of work, not a threshold. It also cost a full
+search at nearly every Local Operations decision. Code removed.
+
+### Rejected: discounting a car its industry cannot work yet
+
+A spotted car scored half when its shipper had no load staged or its receiver's red box was full.
+**−0.29 ± 0.05 (t = −5.52)** over 400 seeds, 47 worse against 4 better; green boxes stocked fell
+15% → 11% of Cargo phases. The reason is the order the bot works in: the Freight Agent stocks a box
+only once a car is spotted to receive it (`canStockProductively`), so an industry is "not ready"
+precisely because nothing has been delivered — the discount withheld the delivery that makes it ready.
+Code removed.
+
+---
+
+## 0.8.0.8 — 2026-09-10
+
+**A played train does not come back. Gitea#23, ruled and closed.**
+
+Jesse, on the question v0.8.0.7 filed rather than answered: *"Once you've played a regularly
+scheduled train and it's in the salvage deck, that train is already on the timetable. It does not
+make sense to put that back into a reshuffled home deck to get played again. By contrast, a regularly
+scheduled train that's in a discard pile could potentially get reused later, and so should have that
+capability. Extras run one time and then they're done — if they are in the Salvage deck, they should
+get shuffled back in so that they could get run again."*
+
+**The test is WHERE the card is, not only what it is** — which is the part worth writing down, because
+it is exactly the rule a later tidy-up would "simplify" into filtering by card kind everywhere. The
+same train card is spent in the Salvage Yard and still runnable in a Department:
+
+| card | where | on a reshuffle |
+| --- | --- | --- |
+| timetabled train | Salvage Yard — it was **played**, its number is on the timetable | stays out |
+| timetabled train | a Department — **discarded**, never played, slot still open | comes back |
+| Extra | anywhere | comes back; an Extra is one run, not a standing slot |
+| everything else | anywhere | comes back, as before |
+
+### And the duplicate that started it
+
+`trainScheduled` was pushing a synthetic `train-<number>` into the Salvage Yard **beside the real
+card `cardPlayed` had already put there** — measured: four scheduled trains left eight entries in a
+pile holding four cards. Nothing read that id. It inflated the pile's depth, displayed as "a card"
+because no such card exists, and would have been swept into the draw deck to be drawn as an id with
+nothing behind it. Removed, which retires the whole phantom-id class rather than papering over it —
+so v0.8.0.7's `cardName()` resolver for `train-<n>` is gone too, along with its test. Dead code kept
+for an id that can no longer exist is worse than no code.
+
+### Games in progress
+
+**Resume.** No predicate changed its answer — nothing that was legal became illegal, and a draw is a
+draw whatever is on top of the deck. What differs is the Salvage Yard's depth, which was
+double-counting, and what a reshuffle would recover. Reshuffles are effectively unreachable in
+ordinary play: eight games driven to 4000 moves across eight seeds produced zero.
+
+Closes #23.
+
+---
+
 ## 0.8.0.7 — 2026-09-10
 
 ### The Salvage Yard was face up and had nothing to say

README.md

@@ -56,7 +56,8 @@ deliberately no longer names one: it went stale for six releases.
 Balance is *not* where it should be, and this file no longer quotes a figure for it. It used to say
 "the developer bot averages 7.0 Revenue against a target of 20", which stopped being true the moment
 the transit rule it names was defaulted to off — that rule was worth ~5.4 of the 7.0, for traffic
-nobody had to work. Measured at the current defaults the bot means about **zero**.
+nobody had to work. Measured at the current defaults the bot meant about **zero** until it began
+planning its switching turns (2026-09-14), which put it near **2.8**.
 
 The three rates — passenger per coach, freight per load, train per transit — are **settings fixed when
 the game is dealt**, along with the opening hand and where an Extra may start, so the economy can be
@@ -102,7 +103,8 @@ separate thing: it assembles the static SITE into `dist/`.)
 
 ```sh
 npm install
-npm test          # node --test
+npm test          # node --test, everything except the bot simulations — run after every change
+npm run test:sim  # test/sim.test.ts, the bot simulations (~7 min) — run after a bot or balance change
 npm run typecheck # tsc --noEmit
 ```
 

TODO.md

@@ -81,7 +81,7 @@ Not items. Things that are true of every change, and that have gone wrong when s
 5. **Replays and saved games** — #14 #47 #48 #49 #50 #51 #52
 6. **Rules** — #12 #80 #82 #83 #85
 7. **Play balance** — #61 #62 #63 #64 #67 #68 #69 #70 #71 #72 #73 #66 #65 #74
-8. **The bot** — #41 #57 #59 #53 #54 #58 #55 #56 #60
+8. **The bot** — #104 #105 #106 #41 #57 #59 #54 #58 #55 #56 #60
 9. **Code health and housekeeping** — #46 #45 #84 #87
 10. **Documentation and assets** — #15a #86 #88
 
@@ -100,20 +100,119 @@ Everything else in this file waits behind a release; this waits behind an aftern
 **Test runs WERE made across 0.7.4 through 0.7.9** (Jesse, 2026-09-07) and produced no change
 requests — the two bugs that did come out of them are Gitea#21 and #22, fixed in v0.7.9.1. So this
 section is not "nobody has touched it since 0.7.4"; it is the narrower and still-true claim that the
-specific paths below have not been exercised at a table. **More testing is planned at the end of the
-0.7.9 series, before 0.8.0 starts** — that is the moment to close these, not a separate errand.
+specific paths below have not been exercised at a table.
+
+**The gate moved.** It was "before 0.8.0 starts"; 0.8.0 shipped anyway, through v0.8.0.8, so the
+session now runs against that build and covers what it added as well. See **Preparing the session**
+below — written 2026-09-10 because the measurement it rests on is the whole point: **three of the
+four things this section is named for do not happen by themselves.**
+
+### Preparing the session
+
+**MEASURED, 2026-09-10, across ten full competitive games driven to completion.** What a table will
+meet without trying, and what it will not:
+
+| interruption | fires in | so |
+| --- | --- | --- |
+| Superintendent clearance (§8.1) | **9/10 games** | you will meet it; just play |
+| a train held at the Limits | 7/10 | ditto |
+| Extras started and queued | 10/10 | ditto |
+| collisions | 7/10 | ditto |
+| Red Flags set / spent | 7/10, 6/10 | ditto |
+| **the Yard Office offer** | **0/10** | must be set up |
+| **the Red Flag hold and its prompt** | **0/10** | must be set up |
+| **extended play (`dayExtended`)** | **0/10** | must be set up |
+
+Those last three are exactly what #39 and #35 are NAMED for. They are not broken — they are
+conditional, and the conditions are these, read out of `advance.ts` rather than guessed:
+
+- **Yard Office** (`advance.ts` ~1290) needs the destination district to contain a card carrying the
+  `yardOffice` **enhancement**, AND an arriving train with **no coach** in its consist, AND a usable
+  route. The bot never builds one, so **somebody has to build a Yard Office and then let a freight
+  train arrive.**
+- **Red Flag hold** (`advance.ts` ~1232) needs the destination player to be **holding the Red Flags
+  maneuver card**, AND an arrival that would genuinely collide — §8.3's own two ways: no free A/D
+  track, or cars fouling the Running Track. So: **hold that card and let your A/D tracks fill.**
+- **Extended play** needs the timetable to RUN OUT, which a five-Day game does not do. Deal it with
+  **`days: 1`** — that is exactly what the 2026-08-29 API verification did, and why it got there.
+
+**What the session needs**
+
+- **Two people, two browsers, two devices.** #35's remaining gap is specifically what a SECOND
+  player sees while waiting on a first, and whether "waiting on Carol" still reads once Carol has
+  closed her laptop. That cannot be tested alone, and it is the half that has never been done.
+- **Two games, not one.** A short `days: 1` game to reach the extension vote, and an ordinary game
+  for everything else — with somebody deliberately building a Yard Office and holding Red Flags.
+- **#42a is separate and takes five minutes**, solitaire, one person: click every field on the setup
+  screen and confirm the dealt game matches what was chosen.
+
+**The caution this section exists because of.** #35's own Reference entry records that the
+2026-08-29 verification passed over the HTTP API — **which renders no dialog** — and that is exactly
+why the v0.7.9 bug survived: the vote sat underneath a modal results dialog whose only control was
+Close. What was proven was that the SERVER supports extended play, not that a player can reach it.
+Read that into every "verified on `phoenix.local`" line in this file, and into everything v0.8.0
+added, all of which is verified by test and simulation and none of it by eye.
+
+**What v0.8.0 added to this list**, none of it played by a person for a whole game and none with a
+second human: the watchable board and its ordered steps, the speed control, the pile highlighting and
+the Home Office deck tile, "Your Move" being put away while catching up, the Day-end collision line,
+and the Salvage Yard naming its top card.
+
+### The checklist
+
+Grouped by what has to be set up, with the item each observation closes. Nothing here needs a
+developer present; what it needs is somebody writing down what they saw.
+
+**Game A — `days: 1`, two humans, two browsers.** Reaches the extension vote in one Day.
+
+- [ ] The vote appears **in front of both players**, not underneath the results dialog (#35 — this is
+      the exact shape of the bug v0.7.9 fixed).
+- [ ] While one player has not voted, the other's turn chart says **who** it is waiting on (#35).
+- [ ] **Close the second laptop mid-vote.** Does the first player learn why nothing is happening, and
+      does "waiting on Carol" still read once Carol is gone? (#35 — never tested.)
+- [ ] Reopen it. The history panel comes back **populated**, not empty, and the board is current
+      (the v0.7.9.5 reconnect fix, never seen by a person).
+- [ ] Vote yes. The extra Day begins and the official result is **unchanged** from when the
+      timetable ran out (#35).
+
+**Game B — ordinary length, two humans, bots to fill.** Everything else.
+
+- [ ] Somebody **builds a Yard Office** and lets a freight train (no coach) arrive at it. The offer
+      interrupts the Mainline Phase and asks a question mid-thought — is it legible, and does it say
+      which train? (#39)
+- [ ] Somebody **holds the Red Flags card** while their A/D tracks are full, so an arrival would
+      collide. The hold is offered out of phase (#39).
+- [ ] A **loaded Extra** is made up and run (#39 — the third of its three).
+- [ ] Watch a bot take a whole turn: does the district follow it, does the lit pile catch the eye,
+      does the caption say who and what? (v0.8.0)
+- [ ] Find the speed that suits you and say what it is — it becomes the committed default.
+- [ ] Let the board fall behind, then press **Skip**. Nothing is lost; the history has it all.
+- [ ] End a Day with a collision on it: the summary reads "N on Day D, N in all" and cannot
+      contradict itself (v0.8.0.2).
+
+**Solitaire, five minutes, alone.**
+
+- [ ] Click through **every field** on the setup screen and confirm the dealt game matches what was
+      chosen (#42a).
+
+**Whatever else happens.** The two bugs that came out of the 0.7.4-0.7.9 runs were both things
+nobody set out to test. Write down anything that reads wrong, even where the rule underneath is
+right — most of this release's defects were legible-but-wrong rather than broken.
 
 - [ ] **#39** — **None of v0.7.4 has been played by a human.** The Yard Office offer, the Red Flag
       hold and its out-of-phase prompt, and the loaded-Extra make-up rules are tested end to end,
       packed, and running on `phoenix.local` — and nobody has met any of them at a board. **Two are
       interruptions that stop the Mainline Phase and put a question in front of somebody
-      mid-thought**, which is exactly the kind of thing only play reveals. See **Reference · #39**.
+      mid-thought**, which is exactly the kind of thing only play reveals. **Neither of those two
+      happens by itself — 0/10 games. See Preparing the session above for what to set up.** See
+      **Reference · #39**.
 
 - [ ] **#35** — **Extended play has never been played at a real table.** It was verified over the HTTP
       API, which renders no dialog — and when a human first reached it in a browser it was unusable
       (fixed in v0.7.9). The multiplayer vote has still never been driven through two browsers: what a
       second player sees while waiting, and whether "waiting on Carol" reads once Carol has closed her
-      laptop, are unanswered. See **Reference · #35**.
+      laptop, are unanswered. **Needs a `days: 1` game — the timetable does not run out in five
+      Days, so extended play fired in 0/10 measured games.** See **Reference · #35**.
 
 - [ ] **#42a** — **Nobody has clicked through the solitaire setup screen's own fields** and confirmed
       the dealt game matches what was chosen. It took three attempts to become reachable at all —
@@ -368,21 +467,37 @@ v0.7.9's collision-floor change (#61).
 The developer bot exists to measure the game, not to be a good opponent — so a bot weakness matters
 when it stops a measurement being trustworthy. **Read #57 before tuning any weights.**
 
+**Since 2026-09-14 the bot plans its whole switching turn** (`sim/switch-planner.ts`, +2.89 revenue a
+game), **takes a face-up card only if it could play it** (+1.52), and **since 2026-09-15 lays track by
+what the district can do afterwards** (`bestValuedLay`, +0.12 over 6400 seeds, run-arounds 9/60 → 22/60). Jesse's goal for it is better decisions in simulated runs AND at a real table, with no
+non-player advantage — it reads the board, never the deck.
+
+- [ ] **#104** — Weigh a switching turn against drawing and the Freight Agent. Letting the planned gain
+      gate switching on its own measured nothing (0.1, 0.25) or worse (0.5): `usefulSwitching` already
+      says yes exactly when a plan gains. What would matter is a VALUE for the other two options to
+      compare against, which the bot does not have. See **Reference · #104**.
+
+- [ ] **#106** — The Extra trap: a full hand of Extras the A/D cap is holding back cannot be discarded,
+      so the next draw forces one into a full Office. All 23 train plays past the cap in 40 games were
+      this. Avoiding the draw measured nothing (−0.03) because it stalled development. See
+      **Reference · #106**.
+
+- [ ] **#105** — Plan across more than one turn. Jesse is in favour, one turn first to see the impact —
+      which is now measured. Deferred for a conversation, not declined. See **Reference · #105**.
+
 - [ ] **#41** — The bot never plays Red Flags — zero in 200 games since Gitea#19, and that is deck
       luck rather than unwillingness. It takes the danger prompt unconditionally; what it never does
       is plant a flag ON PURPOSE to buy a Stage for switching, which needs it to know it wants time.
       See **Reference · #41**.
 
 - [ ] **#57** — The bot's priorities are not the problem — measured across ten heuristic variations.
-      **Read this before tuning weights**; it is the argument that the ceiling is elsewhere. See
-      **Reference · #57**.
+      **Read this before tuning weights**; it is the argument that the ceiling is elsewhere. **Part of
+      "elsewhere" was choosing one Move at a time**: planning the whole switching turn was worth
+      +2.89 (t = 15.8) in the 2026-09-14 bot-tuning round. See **Reference · #57**.
 
 - [ ] **#59** — The run-around is out of reach of any bot, and the deck is why — measured five ways.
       See **Reference · #59**.
 
-- [ ] **#53** — The bot does not know to bring an expedited train back to the station. See **Reference
-      · #53**.
-
 - [ ] **#54** — The bot cannot spot a car at a stub industry, and the cut-ordering rules made that
       visible. See **Reference · #54**.
 
@@ -1424,6 +1539,12 @@ measuring deck luck rather than reachability, and its comment now says so.
 
 #### #57 — THE BOT'S PRIORITIES ARE NOT THE PROBLEM — measured.
 
+**2026-09-14 — confirmed, and one ceiling found.** Reordering priorities still moves nothing; what
+moved the bot was SEARCH. `sim/switch-planner.ts` plans the whole switching turn against a score of
+where it ends, and measured +2.89 ± 0.18 (t = 15.79) over 1600 paired seeds — freight loads and
+unloads 0.22 → 1.53, Cargo phases with a car spotted 5% → 18%. The "8% of Cargo phases" below was a
+fact about how the bot switched, not only about the deck. See `CHANGELOG.md`, 0.8.0.9.
+
 **THE BOT'S PRIORITIES ARE NOT THE PROBLEM — measured.** Ten heuristic variations, each paired
 
 over 400+ seeds. Every reordering of what the bot prefers came out inside the noise; the only
@@ -1449,6 +1570,21 @@ prioritised better. What is left is the economy itself, which is a deck question
 
 #### #59 — THE RUN-AROUND IS OUT OF REACH OF ANY BOT, AND THE DECK IS WHY — measu…
 
+**2026-09-14 — part of "the deck is why" was the bot's own draw.** It was taking ~32 face-up trains and
+industries a game it could not play and discarding them again, so the hand rarely held track. Taking
+only playable cards (now the default) grew districts 14 → 24 cards and run-arounds 3/60 → 9/60. And
+~13 of the ~16 track pieces a game were being laid by the draw turn's "play what is in hand" fallback
+at the first legal square, not by `bestTrackLay`; holding them (−0.83) and placing them by
+`bestTrackLay`'s score (+0.07, noise) both failed, so the next limit is that SCORING — it cannot tell a
+piece that opens an industry site or advances a run-around from one that fills a square. The deck
+measurements below were taken before any of this and should be re-read with it in mind.
+
+**2026-09-15 — the scoring, fixed.** `bestValuedLay` scores the layout a lay leaves (reachable industry
+sites, a closed run-around, ways off the main) instead of the piece: closed run-arounds in 22 of 60
+districts against 9, +0.118 ± 0.029 revenue (t = 4.14, 6400 seeds). The run-around is now reachable
+without changing the deck; what is left is a bot that can USE one, which needs more than one turn of
+planning (#105).
+
 **THE RUN-AROUND IS OUT OF REACH OF ANY BOT, AND THE DECK IS WHY — measured, five ways.**
 
 "Teach the bot to plan across turns" was tried properly and does not work. Every attempt is
@@ -1490,6 +1626,8 @@ the end of the game, drawing the fault **26 times**. Not an engine bug — the m
 exactly as designed — but a clear next bot heuristic: prefer ending a switching turn with any
 expedited crew back on the Office square, at least once it has finished the work it went out for.
 
+**CLOSED 2026-09-14** — see **Done · 53**.
+
 #### #54 — THE BOT CANNOT SPOT A CAR AT A STUB INDUSTRY, and the cut-ordering rul…
 
 **THE BOT CANNOT SPOT A CAR AT A STUB INDUSTRY, and the cut-ordering rules made that visible.**
@@ -1511,6 +1649,12 @@ pass should not read the drop as a deck problem.
 
 #### #58 — The bot cannot get a crew next to an industry, so Flying Switch never…
 
+**2026-09-14 — the premise is now a deck fact.** Flying Switch is dealt **0 copies** (not in sheet 5;
+Jesse, 2026-08-26), so no bot can fire it: across 30 standard games none was ever drawn. The switching
+planner searches the card by default, so it will be used the day it is dealt again. The reachability
+sweep's exemption in `sim.test.ts` stays until then.
+
+
 **The bot cannot get a crew next to an industry, so Flying Switch never fires.** Industries are
 
 now stub-only and the bot places 2.23 a game (was 3.84), in districts averaging under two rows
@@ -1558,6 +1702,114 @@ of zero" over 400 games — but at 400 games the standard error is ±0.33, so a
 have looked like nothing. They are nearly free to re-run now and at least one may have been
 discarded wrongly.
 
+#### #104 — WEIGH SWITCHING AGAINST THE OTHER TWO OPTIONS, not against a threshold.
+
+Measured 2026-09-14, paired over 400 seeds against the planner: switching only when the planned gain
+clears a threshold scored −0.33 at 0.5 (t = −5.06), −0.01 at 0.25, +0.06 at 0.1 (t = 1.79). At 0.5 it
+refuses turns that only collect cars, which feed later deliveries; below that it agrees with
+`usefulSwitching`. The choice that is still made by a fixed ladder is WHICH of §6's three options a
+Stage goes to, and the planner can now put a number on one of them. The other two need numbers of
+their own — what a draw is worth given the hand and the Departments, what stocking a box is worth given
+the cars spotted — before the three can be compared. Also: planning at every Local Operations decision
+costs a full search each time, so any version of this has to stay cheap.
+
+**2026-09-15 — measured the ceiling first: there is almost none.** At 907 real Local Operations choices
+(75 standard games, seeds outside the usual measurement range), every legal option was tried and the rest
+of the game played out by today's bot, 4 times each with the HIDDEN parts reshuffled — the Home Office
+deck order and future rolls — and the same reshuffles for every option, so the comparison is paired.
+Grouped by the rule that made the choice, the value of each alternative against it:
+
+| the ladder chose | times | switch instead | draw instead | Freight Agent instead |
+| --- | --- | --- | --- | --- |
+| draw — nothing urgent, develop | 494 | +0.08 ± 0.12 | — | +0.02 ± 0.04 |
+| Freight Agent — feed the pipeline | 147 | +0.07 ± 0.06 | −0.01 ± 0.04 | — |
+| switch — a train with work at the Office | 108 | — | −0.17 ± 0.10 | −0.25 ± 0.08 |
+| draw — an Office upgrade in hand | 100 | +0.21 ± 0.16 (6) | — | −0.08 ± 0.06 |
+| switch — walk the crew home | 47 | — | +0.22 ± 0.14 | −0.15 ± 0.08 |
+| draw — a train card in hand | 11 | — | — | −0.45 ± 0.24 |
+
+No rule has an alternative that is significantly better; where the table leans, the ladder is usually
+the one that is right. So given how the bot plays each option once chosen, the choice itself is close to
+optimal, and value functions for draw and Freight Agent have little to find. The one lean worth a look if
+this is reopened is walking a stranded crew home (+0.22, t ≈ 1.6). The rollout tool is analysis only —
+the bot never sees a rollout.
+
+#### #106 — THE EXTRA TRAP — why the cap on committed trains still lets an Office overfill.
+
+**2026-09-15 — re-measured under today's defaults, and most of it is not the Extra trap.** 20 "no free
+A/D track" collisions in 60 standard games, −1.67 revenue a game. No train was held (§8.2 or clearance) in
+the Stage before any of them, and only 8 of the 20 trains destroyed were Extras. **Six destroyed a train
+of the same number as a Second Section run within the previous two Stages — and all 26 Second Sections
+the bot ran in those games were an accident:** the New Train phase's "no car on offer" fallback takes
+`options[0]`, and `legalActions` lists `newTrain.secondSection` ahead of the Extra starts, so whenever an
+Extra was waiting to start the bot doubled the train due out instead. The Office never had an A/D track
+to spare for one.
+
+**A RULES QUESTION FOR JESSE, found on the way — not a bot matter.** Q9 (`implications.md`) defines the
+Second Section as a CARD "played on a train that is due out", and `content.ts` defines `SECOND_SECTION`
+with 1 copy — but `buildDeck` never deals it, and `check`'s `newTrain.secondSection` asks for no card in
+hand. So any player may run a Second Section for free on every train due out. Either the card should be
+dealt and required, or the free action is the intended rule and Q9's wording is stale.
+
+Also measured and removed: starting an over-cap Extra where its run never reaches the Office (+0.10,
+t = 1.82) — such a start was on offer at 1 of 25 over-cap starts.
+
+**The accident is fixed** (2026-09-15, default): the New Train fallback takes a car, a pass or the
+Extra's start, never `options[0]` — +0.32 ± 0.09 (t = 3.64), collisions 0.24 → 0.19. **Still open under
+this item:** the forced Extra itself (a full, undiscardable hand of Extras), and the Second Section card
+question above.
+
+`choose` removes train-card plays from the options when `trainWouldOverfillTheOffice`, but yields if
+that would leave nothing legal. It does leave nothing legal in one ordinary position: the hand is over
+the limit (§6.2 requires reducing it) and every card in it is an Extra, which `keepReason` forbids
+discarding. Measured 2026-09-14 after the face-up take rule: 23 of 196 train plays in 40 games were past
+the cap, every one "play what is in hand" with four Extras held, and the worst seeds each lost 3-4
+collisions to it. Declining the draw option in that position measured −0.03 ± 0.11 — collisions fell
+0.26 → 0.20 but cards played fell 29.0 → 25.6. Better answers to try: play the Extra at the least
+dangerous moment rather than the first, count WHEN each committed train is due at the Office instead of
+how many there are, or keep the hand from filling with Extras in the first place.
+
+#### #105 — PLAN ACROSS TURNS — the evidence so far, for the conversation.
+
+For: one-turn planning already reaches most switching work (Cargo phases with a car spotted 5% → 18%),
+and what it cannot do is exactly what spans a Stage — leave a car on a spur for the next crew, or start
+a run-around and finish it later. The planner already scores staged wanted cars (+0.2), which is a
+first, crude step in that direction.
+
+Against, for now: everything between two switching turns is not the player's — a Mainline Phase, trains
+arriving, a Load/Unload phase — so a second turn cannot be searched the way the first is without either
+simulating those phases (arrivals are on the public timetable, but cars on arriving trains are not
+known) or scoring the position between turns more cleverly. And search cost is already what sets the
+test suite's running time. Cheapest next step if taken up: a better score for "what the next turn can
+still reach", not a deeper search.
+
+**2026-09-15 — the run-around measurement that bears on this.** A candidate that lays track by what the
+district can do afterwards (`valueLays`) more than doubled closed run-arounds, 9/60 → 22/60, yet moved
+revenue only +0.14 ± 0.06 (t = 2.58, 1600 seeds). Traced over 40 games: the one-turn planner DOES use
+the loop — 63 of 131 plans in a district with one end a move on it, 35 run it both ways — but its planned
+gain per turn is the same with a run-around as without (0.28 against 0.27). A run-around is for putting a
+train's cars in a different order, which pays off in the turns after; a planner that looks one turn
+ahead has no way to value it.
+
+**2026-09-15 — two-turn planning, built and measured: it does not pay.** `planTwoTurns` kept the six best
+ends of a switching turn, removed the trains that would highball in the Mainline Phase in between (on the
+Office square and made up — their cars leave with them), reset the Moves, planned the next turn from each,
+and chose by the position after departures plus 0.8 of what the next turn adds. Nothing hidden is read.
+
+| version | revenue, 400 paired seeds | what went wrong |
+| --- | --- | --- |
+| first | −0.28 ± 0.11 (t = −2.58) | an expedited train left away drew 31 faults in one game: the fault is charged in the gap, which the discounted next turn "recovered"; trains left away 5.3 a game against 3.1 |
+| with the gap fault charged in full and 0.5 a Stage per train left away | −0.14 ± 0.06 (t = −2.36) | trains still left away 4.8 a game; the "crew must get back to the Office" choice 4.2 a game against 2.7 |
+
+Why a second turn has so little to find, measured over 30 standard games:
+- only **49%** of switching turns have the same train in the district at the next Local Operations choice;
+- **6.1 wanted cars a game** do leave aboard departing trains — but a further switching turn from those
+  exact positions could have spotted only **0.7** of them: most were never deliverable;
+- the one-turn planner already gains no more with a run-around than without (0.28 against 0.27).
+And what a second turn COSTS is a Stage: trains left away have to be walked home, and those choices come
+out of drawing and the Freight Agent (cards played 28.8 → 28.3). A multi-turn bot would have to weigh
+switching against the other two options — which is #104, not a deeper search.
+
 ### Code health and housekeeping
 
 #### #46 — tsc --noUnusedLocals finds 29 unused declarations across 14 files, and…
@@ -1801,6 +2053,15 @@ where it belongs, and it is still open.
 Closed items, kept because several are the only record of a ruling or a lesson. Newest first within
 each group.
 
+### Closed in the 2026-09-14 bot-tuning round (unreleased)
+
+53. ~~**The bot did not know to bring an expedited train back to the station.**~~ — done 2026-09-14,
+    not by a heuristic of its own but as a consequence of planning the switching turn: the planner's
+    score charges an expedited train left away from the Office a full Revenue point, which is what Q3
+    charges. Over 400 paired seeds, 19 games drew expedite faults under the rule ladder and **none**
+    under the planner, worth +1.07 a game (t = 3.83) — the two worst cases had drawn 45 and 42 faults
+    in a single game. See `CHANGELOG.md`, 0.8.0.9.
+
 ### Shipped through v0.7.9.8, from the queue
 
 Closed items, newest first. Kept because several of them are the only record of a ruling or a lesson;

docs/architecture/lobby-and-sessions.md

@@ -36,6 +36,34 @@ and `<ip>:<port>` are both expected — and browser storage is scoped to the ori
 at one address must come back to that address, or they are a stranger with no token. Say so in the
 UI at join time rather than letting someone discover it when they cannot get back in.
 
+**A lost token is recoverable, administratively** (Gitea#33). Everything above makes the token the
+single point of failure: it lives in one browser's storage, and a cleared profile, a private window or
+a different browser ends the seat with the game still running and the session still on disk. Seen at a
+real table — the returning player met an empty lobby while their token sat intact in `sessions.json`,
+and the only way back was an administrator reading the file off the volume and the player pasting it
+into a devtools console.
+
+So there is a supported path, in two halves that are gated differently on purpose:
+
+```
+POST /api/games/<id>/claim  { player }   →  { code, expiresAt, … }   admin secret
+POST /api/claim             { code }     →  { token, gameId, player, gameCode }
+```
+
+**The link carries the code, never the token** — which is the rule three paragraphs up, applied. A
+recovery link is exactly the sort of thing that gets pasted into a chat, so what travels in the URL is
+single-use and expires in thirty minutes (`server/claims.ts`), and the page trades it for the real
+token over a POST as it loads (`?claim=` in `web/main.ts`, which strips it from the address bar either
+way). A leaked code is worthless once spent; a leaked token is the seat for the rest of the game.
+
+**Minting is administrative; spending is not.** Deciding that a particular person has lost a
+particular seat is a judgement no route can make safely — anyone able to mint their own code could
+take any chair at the table. Spending needs no secret because the player following the link is the one
+person in the story who holds none; the code *is* the authorisation, and it is the same shape
+(unguessable, one-time) as the token it hands back. The codes are held in memory: they are minted on
+demand and spent within minutes, so a restart dropping them is the right failure, and persisting them
+would put a credential-equivalent on the volume to solve a problem measured in seconds.
+
 Real accounts can be layered on later without touching the rules engine, which is exactly why
 [`overview.md`](overview.md) keeps that boundary sharp.
 

docs/rules/as-built.md

@@ -84,14 +84,14 @@ part of the road all change the entry point rather than the card's length.
 The Mainline deck is 10 cards; the two Division Points are the fixed ends of
 the Division and are not dealt. What each card does, in the words the game uses on screen:
 
-- **Plains** — 1 region — one Stage each. · 1 Stage for every train — the printed speed is scenery. · One train at a time — anything following has to wait for it to clear.
-- **Curves** — 2 regions — one Stage each. · 2 Stages for every train — the printed speed is scenery. · One train at a time — anything following has to wait for it to clear.
+- **Plains** — 1 region — one Stage each. · 1 Stage for every train. · One train at a time — anything following has to wait for it to clear.
+- **Curves** — 2 regions — one Stage each. · 2 Stages for every train. · One train at a time — anything following has to wait for it to clear.
 - **Hilly** — 2 regions — one Stage each. · This card reads the train's FAST/SLOW rating: a fast train starts further along and crosses in 1 Stage, a slow one in 2 Stages. No other card cares which it is. · One train at a time — anything following has to wait for it to clear.
 - **Heavy Grade** — 3 regions — one Stage each. · A grade, climbing eastward. 3 Stages to climb it and 3 Stages to run down, before help. Helpers start an UPHILL train a region further on; Brakeman does the same DOWNHILL and Airbrakes another again, and Airbrakes cannot be played without Brakeman. Never less than one Stage. · One train at a time — anything following has to wait for it to clear.
-- **Double Track** — 1 region — one Stage each. · 1 Stage for every train — the printed speed is scenery. · TRAINS MAY PASS — two trains may stand on this card at once, so a following train is not held behind a slower one.
+- **Double Track** — 1 region — one Stage each. · 1 Stage for every train. · TRAINS MAY PASS — two trains may stand on this card at once, so a following train is not held behind a slower one.
 - **Uncontrolled Siding** — 2 regions — one Stage each. · A train with the card to itself starts past the back region and is across in 1 Stage. · UNCONTROLLED SIDING — arrive to find a train already here and you take the siding, a region behind it. You are not in the same place, so you do not run into it; it costs you the extra Stage instead. · One train at a time — anything following has to wait for it to clear.
-- **Tunnel** — 2 regions — one Stage each. · 2 Stages for every train — the printed speed is scenery. · One train at a time — anything following has to wait for it to clear.
-- **Trestle** — 1 region — one Stage each. · 1 Stage for every train — the printed speed is scenery. · One train at a time — anything following has to wait for it to clear.
+- **Tunnel** — 2 regions — one Stage each. · 2 Stages for every train. · One train at a time — anything following has to wait for it to clear.
+- **Trestle** — 1 region — one Stage each. · 1 Stage for every train. · One train at a time — anything following has to wait for it to clear.
 - **Interchange** — 2 regions — one Stage each. · A train with the card to itself starts past the back region and is across in 1 Stage. · An Extra beginning its run here starts in the back region and takes the extra Stage. · One train at a time — anything following has to wait for it to clear. · Cars may be sorted into any new order here.
 
 ---

package.json

@@ -1,6 +1,6 @@
 {
   "name": "station-master",
-  "version": "0.8.0.7",
+  "version": "0.8.0.12",
   "private": true,
   "type": "module",
   "description": "Station Master — a railroad operations game",
@@ -10,7 +10,9 @@
   "scripts": {
     "typecheck": "tsc --noEmit",
     "pretest": "tsc --noEmit && node scripts/build-web.ts",
-    "test": "node --test test/*.test.ts test/**/*.test.ts",
+    "test": "node --test $(ls test/*.test.ts test/**/*.test.ts | grep -v '^test/sim.test.ts$')",
+    "pretest:sim": "tsc --noEmit",
+    "test:sim": "node --test test/sim.test.ts",
     "build:web": "node scripts/build-web.ts",
     "serve:web": "node scripts/build-web.ts && npx --yes http-server dist -p 8080 -c-1",
     "deploy:web": "node scripts/deploy-web.ts",

scripts/build-web.ts

@@ -70,11 +70,17 @@ function buildStamp(): string {
    * and v0.7.6's fix to it both shipped correctly to `phoenix.local` and neither reached the browser
    * that asked for them (Jesse, twice, 2026-08-29 — "setup did not work").
    *
-   * The version plus the build's own timestamp is always distinct, needs nothing from the
-   * environment, and stays honest: two builds of the same commit ARE two deploys, and a cache key
-   * that says so costs one refetch, while one that lies costs a release nobody receives.
+   * A marker plus the build's own timestamp is always distinct, needs nothing from the environment,
+   * and stays honest: two builds of the same commit ARE two deploys, and a cache key that says so
+   * costs one refetch, while one that lies costs a release nobody receives.
+   *
+   * NOT THE VERSION, which is what this used to lead with. The stamp below already begins with
+   * `v${pkg.version}`, so on exactly the builds that take this path — every `.s9pk`, which has no
+   * `.git` — the header read "v0.8.0.10 · 0.8.0.10-mfq2p1 · …" and the version appeared twice
+   * (Jesse, playtest 2026-09-16). The timestamp alone carries the uniqueness; the version is
+   * already said once, properly, at the front.
    */
-  let git = `${pkg.version}-${Date.now().toString(36)}`;
+  let git = `nogit-${Date.now().toString(36)}`;
   try {
     const sha = execFileSync('git', ['rev-parse', '--short', 'HEAD'], { cwd: root })
       .toString()

src/engine/advance.ts

@@ -94,6 +94,18 @@ function nodeIndexOfOffice(s: GameState, seat: SeatIndex): number {
 
 const step = (d: Direction): number => (d === 'east' ? 1 : -1);
 
+/** The Division node a train is standing on — its Office, its Mainline card, or its Division Point. */
+function nodeIndexOfTray(s: GameState, tray: CrewTray): number | null {
+  const at = tray.position;
+  if (at.at === 'grid') return nodeIndexOfOffice(s, at.seat);
+  if (at.at === 'mainline') return at.index;
+  if (at.at === 'divisionPoint') {
+    const side = at.side;
+    return s.division.nodes.findIndex((n) => n.kind === 'divisionPoint' && n.side === side);
+  }
+  return null;
+}
+
 // ---------------------------------------------------------------------------
 // advance
 // ---------------------------------------------------------------------------
@@ -522,8 +534,11 @@ type MoveOutcome = 'moved' | 'held' | 'needsClearance';
  *
  * This is reachable purely through switching. A train arrives made up, and only comes apart because
  * the player took cars onto the nose or picked up a cut in a run-around.
+ *
+ * Exported for the switching planner (`sim/switch-planner.ts`), which has to know whether a plan
+ * leaves a train unable to run — and must ask this rule rather than keep a copy of it.
  */
-function badlyMadeUp(tray: CrewTray): string | null {
+export function badlyMadeUp(tray: CrewTray): string | null {
   const n = tray.consist.length;
   if (n === 0) return null;
   const pulling = tray.engineAt === 0;
@@ -1064,10 +1079,27 @@ function evaluateClearance(
    * constrained. Each Office upgrade to a Control Point splits one in two and buys capacity.
    */
   const subdivision = subdivisions(s).find((group) => group.includes(targetIndex)) ?? [targetIndex];
+  /**
+   * ONLY WHAT IS AHEAD (Gitea#26). §8.1 asks about a train the considered train would FOLLOW, and one
+   * moving TOWARDS it — both of which are ahead of it. A Subdivision runs the length of every Whistle
+   * Post between two Control Points, so it can hold a train BEHIND the one departing: in the reported
+   * game X15 highballed west from an Office while X18, also westbound, was still crossing the card to its
+   * east. Counting X18 put a meaningless ruling to the Superintendent; holding X15 kept the Whistle Post's
+   * one A/D track full, and X18 arrived into it and was destroyed. A train behind and moving away is no
+   * threat at all.
+   *
+   * "Behind" is strictly behind the card the departing train stands on. A train on that same card is still
+   * counted, exactly as before: which of two trains sharing a card is in front is `entryConflict`'s region
+   * question, and this is not the place to answer it.
+   */
+  const from = nodeIndexOfTray(s, tray);
+  const behind = (onCard: number): boolean =>
+    from !== null && from >= 0 && (tray.direction === 'east' ? onCard < from : onCard > from);
   const occupants: { tray: TrayId; onCard: number }[] = [];
   for (const i of subdivision) {
     const n = s.division.nodes[i];
     if (!n || n.kind !== 'mainline') continue;
+    if (behind(i)) continue;
     for (const t of n.transits) if (t.tray) occupants.push({ tray: t.tray, onCard: i });
   }
 
@@ -1452,6 +1484,7 @@ function arriveAtOffice(
     trainNumber: tray.trainNumber ?? 0,
     consist: tray.consist.map((c) => ({ ...c })),
     office: officeProfile(area.tier).name,
+    owner: playerAtSeat(s, seat),
     expedited: isExpedited(tray),
   });
 

src/engine/apply.ts

@@ -1508,17 +1508,48 @@ export function selectDestination(
   return chosen ?? atTo[0];
 }
 
+/**
+ * ROUTES, WALKED ONCE PER POSITION.
+ *
+ * A route walk (`reachableDestinations`) was a third of all simulation time, and most of it was the
+ * same walk repeated: `legal.ts` walks a tray's routes to list its moves, then `check` walks them again
+ * for every one of those moves, and `applyIntent` walks the chosen one a third time in `execute`.
+ * Profiled 2026-09-14 with inlining off: `reachableDestinations` 34% inclusive, garbage collection 34%.
+ *
+ * So while one position is being examined — a legal-action listing, or the check and execute of one
+ * intent — a walk is kept and reused. Both scopes read the state and never write it, the key names
+ * everything the walk depends on besides that state, and the cache is keyed to the state OBJECT and
+ * cleared when the scope ends, so a hit returns exactly what a fresh walk would have. Nothing may
+ * mutate a returned route; nothing does.
+ */
+let routeCache: { state: GameState; routes: Map<string, MoveDestination[]> } | null = null;
+
+export function withRouteCache<T>(s: GameState, fn: () => T): T {
+  if (routeCache) return fn();
+  routeCache = { state: s, routes: new Map() };
+  try {
+    return fn();
+  } finally {
+    routeCache = null;
+  }
+}
+
 function destinationsFor(
   s: GameState,
   player: PlayerIndex,
   trayId: TrayId,
   from: GridCoord,
   reverse: boolean,
-) {
+): MoveDestination[] {
+  const cache = routeCache?.state === s ? routeCache.routes : null;
+  const key = cache ? `${player}|${trayId}|${from.row},${from.col}|${reverse ? 1 : 0}` : '';
+  const hit = cache?.get(key);
+  if (hit) return hit;
+
   const tray = s.trays.get(trayId)!;
   const facing = facingPort(s, trayId);
   const exit: Port = reverse ? reversePort(s, player, from, facing) : facing;
-  return reachableDestinations(
+  const routes = reachableDestinations(
     {
       area: areaOf(s, player),
       occupancy: occupancyFor(s, player, trayId),
@@ -1528,6 +1559,8 @@ function destinationsFor(
     from,
     exit,
   );
+  cache?.set(key, routes);
+  return routes;
 }
 
 /**
@@ -1856,7 +1889,15 @@ function execute(s: GameState, player: PlayerIndex, i: Intent): GameEvent[] {
           ? REALIGNMENTS.find((r) => r.from === node.card)?.to
           : undefined;
       return [
-        { type: 'mainlineModified', player, cardId: i.cardId, node: i.node, key, ...(became ? { became } : {}) },
+        {
+          type: 'mainlineModified',
+          player,
+          cardId: i.cardId,
+          node: i.node,
+          key,
+          ...(node?.kind === 'mainline' ? { from: node.card } : {}),
+          ...(became ? { became } : {}),
+        },
       ];
     }
 
@@ -2242,7 +2283,8 @@ export function reduce(s: GameState, e: GameEvent): void {
       // Everything swept comes back as ONE pile, then §4.6-4.7's opening is re-run: three cards
       // turned face up as the Departments, the rest face down as the Home Office deck. The
       // Departments start one deep again, exactly as at setup.
-      s.decks.salvageYard = [];
+      // The spent trains stay where they are; everything else in the Yard has just been swept up.
+      s.decks.salvageYard = s.decks.salvageYard.filter((id) => isSpentTimetabledTrain(s, id));
       s.decks.departments = [[], [], []];
       const order = [...e.order];
       for (const pile of s.decks.departments) {
@@ -2473,7 +2515,15 @@ export function reduce(s: GameState, e: GameEvent): void {
     case 'trainScheduled':
       s.timetable[e.slot] = e.trainNumber;
       s.rngState = e.rngState;
-      s.decks.salvageYard.push(`train-${e.trainNumber}`);
+      /**
+       * THE CARD IS ALREADY IN THE SALVAGE YARD — `cardPlayed` put it there, by its real id.
+       *
+       * This used to push a second, SYNTHETIC `train-<number>` beside it, so scheduling four trains
+       * left eight entries in a pile holding four cards. Nothing ever read that id: it inflated the
+       * pile's depth, it displayed as "a card" because no such card exists, and
+       * `reshuffleIfDepleted` would have swept it into the draw deck to be drawn as an id with
+       * nothing behind it. Removed 2026-09-10 (Gitea#23).
+       */
       break;
 
     case 'carPlacedOnTrain': {
@@ -2693,7 +2743,8 @@ export function reduce(s: GameState, e: GameEvent): void {
  * index is out of range, which `check` reports rather than silently defaulting — a wrong
  * orientation is a different card, not a detail.
  */
-function protoCard(
+/** Exported for the same reason as `extendLimitsIfNeeded`: the bot builds the card a lay would place exactly as the reducer does. */
+export function protoCard(
   kind: { kind: string; geometry?: string; facility?: string; hand?: string },
   variant: number | undefined,
 ): TrackCard | null {
@@ -2930,9 +2981,34 @@ function spendCard(s: GameState, player: PlayerIndex, cardId: CardId): void {
  * that has genuinely used every card ends on `DECK_EMPTY` rather than reshuffling an empty pile.
  * Cards played onto the board are NOT recovered: they are on the table, which is where they belong.
  */
+/**
+ * §6.2, AND THE RULING THAT SETTLES IT — Jesse, 2026-09-10 (Gitea#23).
+ *
+ * "Once you've played a regularly scheduled train and it's in the salvage deck, that train is
+ * already on the timetable. It does not make sense to put that back into a reshuffled home deck to
+ * get played again. By contrast, a regularly scheduled train that's in a discard pile could
+ * potentially get reused later, and so should have that capability. Extras run one time and then
+ * they're done — if they are in the Salvage deck, they should get shuffled back in so that they
+ * could get run again."
+ *
+ * So the test is WHERE the card is, not only what it is. A timetabled train in the SALVAGE YARD was
+ * played: its number is on the timetable and cannot be scheduled twice, so the card is spent and
+ * stays out. The same card sitting in a DEPARTMENT was discarded, never played, and its slot is
+ * still open — so it comes back with everything else. An Extra is a single run rather than a
+ * standing slot, so a played one is free to be run again.
+ */
+function isSpentTimetabledTrain(s: GameState, id: CardId): boolean {
+  return s.cards.get(id)?.kind.kind === 'timetabledTrain';
+}
+
 function reshuffleIfDepleted(s: GameState, taking: number): GameEvent | null {
   if (s.decks.homeOffice.length > taking) return null;
-  const collected = [...s.decks.salvageYard, ...s.decks.departments.flat()];
+  const collected = [
+    // The Salvage Yard, less the trains whose slots are already filled — see above.
+    ...s.decks.salvageYard.filter((id) => !isSpentTimetabledTrain(s, id)),
+    // Every Department in full: a discarded train was never played, so it is still runnable.
+    ...s.decks.departments.flat(),
+  ];
   if (collected.length === 0) return null;
   const rng = createRng(s.rngState);
   return { type: 'deckReshuffled', order: rng.shuffle(collected), rngState: rng.getState() };
@@ -3059,7 +3135,8 @@ function applyModifier(area: OfficeArea, coord: GridCoord, modifier: ModifierKin
  * §8.1 and §10 both reason about "the track between the train and the Limits", Interlocking holds
  * an arrival AT the Limits, and running past a player's Limits is what makes a collision his fault.
  */
-function extendLimitsIfNeeded(area: OfficeArea, placed: GridCoord): void {
+/** Exported so the bot can score a lay on a copy of the district by the engine's own rule, not a copy of it. */
+export function extendLimitsIfNeeded(area: OfficeArea, placed: GridCoord): void {
   if (placed.row !== area.runningRow) return;
 
   if (placed.col <= area.limitsWest.col) {
@@ -3253,16 +3330,35 @@ function limitsCard(): TrackCard {
 // Public entry point
 // ---------------------------------------------------------------------------
 
-export function applyIntent(s: GameState, player: PlayerIndex, i: Intent): ApplyResult {
-  const code = check(s, player, i);
-  if (code) return { ok: false, code, message: `${i.type} rejected: ${code}` };
+/**
+ * THE FIRST HALF OF `applyIntent`: decide, without changing anything.
+ *
+ * `check` and `execute` read the same unchanged position, so its routes are walked once between them
+ * (`withRouteCache`). Never writes `s`. Split out for a caller that decides many intents against ONE
+ * position and applies each to a COPY of it — the switching planner — which can then share that
+ * position's routes across every candidate instead of re-walking them on each copy.
+ */
+export function prepareIntent(s: GameState, player: PlayerIndex, i: Intent): ApplyResult {
+  const prepared = withRouteCache(s, (): { code: RejectionCode } | { events: GameEvent[] } => {
+    const code = check(s, player, i);
+    return code ? { code } : { events: execute(s, player, i) };
+  });
+  if ('code' in prepared) return { ok: false, code: prepared.code, message: `${i.type} rejected: ${prepared.code}` };
+  return { ok: true, events: prepared.events };
+}
 
-  const events = execute(s, player, i);
+/** THE SECOND HALF: fold events `prepareIntent` produced into a state equal to the one it read. */
+export function commitEvents(s: GameState, events: readonly GameEvent[]): void {
   for (const e of events) reduce(s, e);
   // Gitea#16 — the intent half of the fold; `advance` does the phase driver's half. See `tally.ts`
   // for why it cannot simply live inside `reduce`.
   for (const e of events) tallyEvent(s, e);
-  return { ok: true, events };
+}
+
+export function applyIntent(s: GameState, player: PlayerIndex, i: Intent): ApplyResult {
+  const r = prepareIntent(s, player, i);
+  if (r.ok) commitEvents(s, r.events);
+  return r;
 }
 
 export { isOperationalRail, destinationsFor };

src/engine/content.ts

@@ -772,7 +772,17 @@ export function mainlineDescription(kind: MainlineKind, gradeUp: Direction = 'ea
         `${stages(run({}))}.`,
     );
   } else {
-    parts.push(`${stages(run({}))} for every train — the printed speed is scenery.`);
+    /**
+     * NOT A WORD ABOUT SPEED HERE — Jesse, playtest 2026-09-16.
+     *
+     * This read "the printed speed is scenery", which sent a player hunting the card for a number
+     * that is not drawn on it. The first rewrite said "fast or slow alike", which is true but raises
+     * the question on thirteen cards in order to answer it. **Exactly one card reads FAST/SLOW**:
+     * Hilly, the only profile with `speedStarts` (see the note above it). So the explanation belongs
+     * on that card, where the branch above already gives it, and everywhere else says nothing —
+     * silence is the honest answer when the rating genuinely does not apply.
+     */
+    parts.push(`${stages(run({}))} for every train.`);
   }
 
   if (kind === 'uncontrolledSiding') {

src/engine/events.ts

@@ -87,7 +87,13 @@ export type GameEvent =
   | { type: 'deckReshuffled'; order: CardId[]; rngState: number }
   /** `variant` is the chosen orientation (Gap 11); it must be replayable, so it rides the event. */
   | { type: 'cardPlayed'; player: PlayerIndex; cardId: CardId; placement?: GridCoord; variant?: number }
-  | { type: 'mainlineModified'; player: PlayerIndex; cardId: CardId; node: number; key: string; became?: string }
+  /**
+    * `from` is the card's kind BEFORE the change, carried so the log can say what was realigned
+    * rather than only what it turned into (playtest, 2026-09-15: "it should state that the mainline
+    * card 3 curves was converted to plains"). Events are derived by replaying a save, never stored,
+    * so widening one strands nothing on disk.
+    */
+  | { type: 'mainlineModified'; player: PlayerIndex; cardId: CardId; node: number; key: string; from?: string; became?: string }
   /** §Q (Gitea#19) — a flag planted on one side of a district's Limits. */
   | { type: 'redFlagsSet'; player: PlayerIndex; cardId: CardId; seat: SeatIndex; side: Direction }
   /** §Q (Gitea#19) — the flag stopped a train and came down with it. One card, one train. */
@@ -153,7 +159,13 @@ export type GameEvent =
    * switched normally like any other arrival, but it has to be back on the Office square before the
    * next Mainline Phase begins, or `expediteFault` fires.
    */
-  | { type: 'trainArrived'; trainNumber: number; consist: RollingStock[]; office: string; expedited: boolean }
+  /**
+   * `owner` is WHOSE Office it reached — the district's player, not whoever is acting. The Mainline
+   * Phase has no actor, so nothing else in the line could name the seat, and the narration said only
+   * "ARRIVED at the Whistle Post" — every seat's Office has a tier, and at a four-seat table three of
+   * them are somebody else's (Jesse, playtest 2026-09-16).
+   */
+  | { type: 'trainArrived'; trainNumber: number; consist: RollingStock[]; office: string; owner: PlayerIndex; expedited: boolean }
   | { type: 'trainDiverted'; trainNumber: number; to: string; reason: string }
   /**
    * The train ran the length of the Division and left it. `side` is the Division Point it left by,

src/engine/legal.ts

@@ -14,7 +14,7 @@
 
 import type { CarType, Hand, TrackGeometry } from './content.ts';
 import { enhancementRule, mainlineProfile } from './content.ts';
-import { check, areaOf, destinationsFor } from './apply.ts';
+import { check, areaOf, destinationsFor, withRouteCache } from './apply.ts';
 import type { Intent } from './intents.ts';
 import type { GameState, GridCoord, PlayerIndex } from './state.ts';
 import { coordKey, seatOf } from './state.ts';
@@ -53,7 +53,80 @@ const CAR_TYPES: readonly CarType[] = ['coach', 'boxcar', 'reefer', 'hopper', 't
 
 /** Every intent `player` may legally submit right now. */
 export function legalActions(s: GameState, player: PlayerIndex): Intent[] {
-  return candidates(s, player).filter((i) => check(s, player, i) === null);
+  // One position, examined many times over: its routes are walked once (`withRouteCache`).
+  return withRouteCache(s, () => candidates(s, player).filter((i) => check(s, player, i) === null));
+}
+
+/**
+ * §6.1 — the switching half of the Local Operations candidates, in the order `legalActions` offers
+ * them. Split out so the switching planner (`sim/switch-planner.ts`) can ask for just these without
+ * `check` running over every draw and Freight Agent candidate at each of the thousands of positions it
+ * tries — that was about a quarter of all planning time. Still no rules here: `check` decides.
+ */
+function switchCandidates(s: GameState, player: PlayerIndex): Intent[] {
+  const out: Intent[] = [];
+  for (const [trayId, tray] of s.trays) {
+    if (tray.position.at !== 'grid' || tray.position.seat !== seatOf(s, player)) continue;
+    const from = tray.position.coord;
+    for (const reverse of [false, true]) {
+      const dests = destinationsFor(s, player, trayId, from, reverse);
+      // Grouped by destination square so `distinguishingVia` only ever compares routes that are
+      // actually racing for the same button — two routes to DIFFERENT squares need no `via` to
+      // tell apart, `to` already does that.
+      const byCoord = new Map<string, MoveDestination[]>();
+      for (const d of dests) {
+        const k = coordKey(d.coord);
+        (byCoord.get(k) ?? byCoord.set(k, []).get(k)!).push(d);
+      }
+      for (const group of byCoord.values()) {
+        for (const d of group) {
+          const via = group.length > 1 ? distinguishingVia(d, group) : undefined;
+          out.push({ type: 'switch.move', trayId, to: d.coord, reverse, ...(via ? { via } : {}) });
+        }
+      }
+    }
+    for (let n = 1; n <= tray.consist.length; n++) {
+      out.push({ type: 'switch.dropCars', trayId, count: n });
+      // Off the nose as well as the tail — the only way to get cars back off the front of a train
+      // that shoved a cut, and therefore the only way an engine buried mid-train reaches an end.
+      out.push({ type: 'switch.dropCars', trayId, count: n, fromNose: true });
+    }
+    // Small Yard: enumerating every permutation would explode, so offer the useful ones —
+    // bringing each car to the droppable end, plus a full reversal. `check` validates any order,
+    // so a UI may submit an arbitrary permutation.
+    const n = tray.consist.length;
+    if (n > 1) {
+      for (let k = 0; k < n; k++) {
+        const order = [...Array(n).keys()].filter((x) => x !== k);
+        order.push(k);
+        out.push({ type: 'switch.sortConsist', trayId, order });
+      }
+      out.push({ type: 'switch.sortConsist', trayId, order: [...Array(n).keys()].reverse() });
+    }
+  }
+  // Flying Switch — roll a cut into an ADJACENT industry without the engine entering it.
+  for (const cardId of s.decks.hands.get(player) ?? []) {
+    const k = s.cards.get(cardId)?.kind;
+    if (k?.kind !== 'maneuver' || k.key !== 'flyingSwitch') continue;
+    for (const [trayId, tray] of s.trays) {
+      if (tray.position.at !== 'grid' || tray.position.seat !== seatOf(s, player)) continue;
+      const from = tray.position.coord;
+      for (const reverse of [false, true]) {
+        for (const d of destinationsFor(s, player, trayId, from, reverse)) {
+          for (let count = 1; count <= tray.consist.length; count++) {
+            out.push({ type: 'maneuver.flyingSwitch', cardId, trayId, count, to: d.coord });
+          }
+        }
+      }
+    }
+  }
+  out.push({ type: 'switch.end' });
+  return out;
+}
+
+/** The switching intents `player` may legally submit right now — exactly `legalActions`' switching subset. */
+export function legalSwitchingActions(s: GameState, player: PlayerIndex): Intent[] {
+  return withRouteCache(s, () => switchCandidates(s, player).filter((i) => check(s, player, i) === null));
 }
 
 export function isLegal(s: GameState, player: PlayerIndex, i: Intent): boolean {
@@ -138,62 +211,7 @@ function localOpsCandidates(s: GameState, player: PlayerIndex): Intent[] {
   const area = areaOf(s, player);
 
   // -- switch (§6.1)
-  for (const [trayId, tray] of s.trays) {
-    if (tray.position.at !== 'grid' || tray.position.seat !== seatOf(s, player)) continue;
-    const from = tray.position.coord;
-    for (const reverse of [false, true]) {
-      const dests = destinationsFor(s, player, trayId, from, reverse);
-      // Grouped by destination square so `distinguishingVia` only ever compares routes that are
-      // actually racing for the same button — two routes to DIFFERENT squares need no `via` to
-      // tell apart, `to` already does that.
-      const byCoord = new Map<string, MoveDestination[]>();
-      for (const d of dests) {
-        const k = coordKey(d.coord);
-        (byCoord.get(k) ?? byCoord.set(k, []).get(k)!).push(d);
-      }
-      for (const group of byCoord.values()) {
-        for (const d of group) {
-          const via = group.length > 1 ? distinguishingVia(d, group) : undefined;
-          out.push({ type: 'switch.move', trayId, to: d.coord, reverse, ...(via ? { via } : {}) });
-        }
-      }
-    }
-    for (let n = 1; n <= tray.consist.length; n++) {
-      out.push({ type: 'switch.dropCars', trayId, count: n });
-      // Off the nose as well as the tail — the only way to get cars back off the front of a train
-      // that shoved a cut, and therefore the only way an engine buried mid-train reaches an end.
-      out.push({ type: 'switch.dropCars', trayId, count: n, fromNose: true });
-    }
-    // Small Yard: enumerating every permutation would explode, so offer the useful ones —
-    // bringing each car to the droppable end, plus a full reversal. `check` validates any order,
-    // so a UI may submit an arbitrary permutation.
-    const n = tray.consist.length;
-    if (n > 1) {
-      for (let k = 0; k < n; k++) {
-        const order = [...Array(n).keys()].filter((x) => x !== k);
-        order.push(k);
-        out.push({ type: 'switch.sortConsist', trayId, order });
-      }
-      out.push({ type: 'switch.sortConsist', trayId, order: [...Array(n).keys()].reverse() });
-    }
-  }
-  // Flying Switch — roll a cut into an ADJACENT industry without the engine entering it.
-  for (const cardId of s.decks.hands.get(player) ?? []) {
-    const k = s.cards.get(cardId)?.kind;
-    if (k?.kind !== 'maneuver' || k.key !== 'flyingSwitch') continue;
-    for (const [trayId, tray] of s.trays) {
-      if (tray.position.at !== 'grid' || tray.position.seat !== seatOf(s, player)) continue;
-      const from = tray.position.coord;
-      for (const reverse of [false, true]) {
-        for (const d of destinationsFor(s, player, trayId, from, reverse)) {
-          for (let count = 1; count <= tray.consist.length; count++) {
-            out.push({ type: 'maneuver.flyingSwitch', cardId, trayId, count, to: d.coord });
-          }
-        }
-      }
-    }
-  }
-  out.push({ type: 'switch.end' });
+  out.push(...switchCandidates(s, player));
 
   // -- draw (§6.2)
   out.push({ type: 'draw.fromHomeOffice' });

src/engine/track.ts

@@ -377,7 +377,7 @@ export function reachableDestinations(
   start: GridCoord,
   initialExit: Port,
 ): MoveDestination[] {
-  return exploreMoves(ctx, start, initialExit).destinations;
+  return exploreMoves(ctx, start, initialExit, false).destinations;
 }
 
 /**
@@ -434,12 +434,19 @@ export function exploreMoves(
   ctx: MoveContext,
   start: GridCoord,
   initialExit: Port,
+  /**
+   * False when only the destinations are wanted (`reachableDestinations`, every legality check): the
+   * rejections are then not recorded at all. They never change a destination, and building them was
+   * pure allocation on the hottest path in the engine.
+   */
+  collectBlocks = true,
 ): { destinations: MoveDestination[]; blocked: MoveBlock[] } {
   const { area, occupancy } = ctx;
   const results: MoveDestination[] = [];
   const blocked: MoveBlock[] = [];
   const noted = new Set<string>();
   const block = (coord: GridCoord, kind: MoveBlockKind, why: string): void => {
+    if (!collectBlocks) return;
     const k = coordKey(coord);
     if (noted.has(k)) return;
     noted.add(k);
@@ -459,10 +466,25 @@ export function exploreMoves(
     couples: RollingStock[];
     /** Coord key each entry in `couples` came off, aligned by index — see `routeOutcomeKey`. */
     origins: string[];
-    /** Cards visited on THIS route, start included. A per-path set, not a global one — see the
-     * module doc comment on `MAX_ENUMERATED_FRONTIER` for why a global one would forbid the very
-     * routes this walk exists to find. */
-    visited: Set<string>;
+  };
+
+  /**
+   * Has THIS route already used `to`? Per-path, not global — see the doc comment on
+   * `MAX_ENUMERATED_FRONTIER` for why a global set would forbid the very routes this walk exists to
+   * find.
+   *
+   * Read off the route's own `path` instead of a Set copied at every step, which was a large share of
+   * the walk's garbage. It answers exactly as that Set did: the start square, then every square
+   * enqueued along the route AFTER the first hop, including this node's own — the first hop's square
+   * was never added, and `path[0]` is that square, so the scan begins at 1.
+   */
+  const onRoute = (node: Frontier, to: GridCoord): boolean => {
+    if (sameCoord(to, start)) return true;
+    if (node.path.length > 0 && sameCoord(to, node.coord)) return true;
+    for (let k = 1; k < node.path.length; k++) {
+      if (sameCoord(node.path[k]!.coord, to)) return true;
+    }
+    return false;
   };
 
   // The very first hop is checked here because `start`'s card is not itself enqueued; every later
@@ -493,13 +515,14 @@ export function exploreMoves(
       path: [],
       couples: ownCut,
       origins: ownCut.map(() => startKey),
-      visited: new Set([startKey]),
     },
   ];
   let enumerated = 1;
 
-  while (queue.length > 0) {
-    const node = queue.shift()!;
+  // FIFO by index rather than `shift()`, which re-packs the array on every pop. Same order.
+  let head = 0;
+  while (head < queue.length) {
+    const node = queue[head++]!;
     const card = cardAt(area, node.coord);
     if (!card) continue;
 
@@ -592,17 +615,15 @@ export function exploreMoves(
       // direction; it never says without repeating ground, but a train cannot occupy the same
       // track twice at once either). Per-path, not global — a DIFFERENT route may legitimately
       // pass through a card this one already used.
-      const toKey = coordKey(to);
-      if (node.visited.has(toKey)) continue;
+      if (onRoute(node, to)) continue;
       if (enumerated >= MAX_ENUMERATED_FRONTIER) break;
       enumerated++;
       const step: MoveStep = { coord: node.coord, entry: node.entry, exit };
-      const visited = new Set(node.visited);
-      visited.add(toKey);
-      queue.push({ coord: to, entry: opposite(exit), path: [...node.path, step], couples, origins, visited });
+      queue.push({ coord: to, entry: opposite(exit), path: [...node.path, step], couples, origins });
     }
   }
 
+  if (!collectBlocks) return { destinations: results, blocked };
   // A card that turned out to be reachable after all is not a blocker: the walk may meet a square
   // from a bad angle first and a good one later.
   const reached = new Set(results.map((r) => coordKey(r.coord)));

src/server/claims.ts

@@ -0,0 +1,83 @@
+/**
+ * SEAT RECOVERY CODES — Gitea#33.
+ *
+ * A session token is the only identity the game has (`lobby-and-sessions.md` §1) and it lives in
+ * exactly one place the player controls: their browser's `localStorage`, scoped to the origin they
+ * joined at. Lose that — a different browser, a cleared profile, a private window — and the seat is
+ * unreachable, because there is nothing else on the server that will accept a claim to it. Seen at a
+ * real table on 2026-09-16: the joining player came back to an empty lobby while their token sat
+ * intact in `sessions.json`, and the only way in was an administrator reading the file off the data
+ * volume and the player pasting it into a devtools console.
+ *
+ * THE CODE IS NOT THE TOKEN, AND THAT IS THE WHOLE POINT. §1 says to keep the token out of URLs so it
+ * is not shoulder-surfed or pasted into a chat — and a recovery link is exactly the kind of thing
+ * that gets pasted into a chat. So an administrator mints a SHORT-LIVED, SINGLE-USE code, the player
+ * opens a link carrying that, and the page trades it for the real token over the same connection it
+ * would have used anyway. A code that leaks after it is spent is worth nothing; a token that leaks is
+ * worth the seat for the rest of the game.
+ *
+ * PURE ON PURPOSE, like `lobby.ts` beside it: no sockets, no filesystem, no clock of its own. `now`
+ * is passed in so expiry is testable without faking timers, which is the only reason this file can be
+ * tested at all — nothing in this repo stands an HTTP server up to make requests against it.
+ *
+ * IN MEMORY, NOT ON DISK, which is a deliberate limit rather than an oversight. A restart drops every
+ * outstanding code, and that is the right failure: the codes are minted on demand and spent within
+ * minutes, the administrator is by definition present, and persisting them would put a credential-
+ * equivalent on the volume to solve a problem measured in seconds.
+ */
+
+import { randomUUID } from 'node:crypto';
+
+/**
+ * Long enough to walk to the other room and read it out; short enough that a link left in a chat
+ * window is useless by the time anyone scrolls back to it.
+ */
+export const CLAIM_TTL_MS = 30 * 60 * 1000;
+
+export type ClaimStore = {
+  /** Mint a code for one seat's token. Returns the code and when it stops working. */
+  mint(token: string, gameId: string, now: number, ttlMs?: number): { code: string; expiresAt: number };
+  /**
+   * Spend a code. Returns the seat it names, or null when the code is unknown, already spent or
+   * expired — deliberately one answer for all three, so a caller cannot probe which it was.
+   */
+  redeem(code: string, now: number): { token: string; gameId: string } | null;
+  /** Outstanding, unexpired codes. For tests and for anything that wants to report the store's size. */
+  outstanding(now: number): number;
+};
+
+export function createClaimStore(): ClaimStore {
+  const claims = new Map<string, { token: string; gameId: string; expiresAt: number }>();
+
+  /** Expiry is lazy: there is no timer to own, start, stop or leak across a server's lifetime. */
+  const prune = (now: number): void => {
+    for (const [code, claim] of claims) if (claim.expiresAt <= now) claims.delete(code);
+  };
+
+  return {
+    mint(token, gameId, now, ttlMs = CLAIM_TTL_MS) {
+      prune(now);
+      // The same primitive the session tokens themselves use (`lobby.ts`), for the same reason: it
+      // has to be unguessable, and inventing a second scheme here would be inventing a weaker one.
+      const code = randomUUID();
+      const expiresAt = now + ttlMs;
+      claims.set(code, { token, gameId, expiresAt });
+      return { code, expiresAt };
+    },
+
+    redeem(code, now) {
+      prune(now);
+      const claim = claims.get(code);
+      if (!claim) return null;
+      // SINGLE USE. Deleted before the caller can do anything with it, so two browsers racing on the
+      // same link cannot both be seated — and a link that stays in someone's history is spent.
+      claims.delete(code);
+      return { token: claim.token, gameId: claim.gameId };
+    },
+
+    outstanding(now) {
+      prune(now);
+      return claims.size;
+    },
+  };
+}

src/server/http.ts

@@ -37,6 +37,7 @@ import {
   writeLobby,
   writeSessions,
 } from './persistence.ts';
+import { createClaimStore } from './claims.ts';
 import { createSession } from './session.ts';
 import type { GameSession, Push } from './session.ts';
 import {
@@ -170,6 +171,15 @@ export function startServer(opts: ServerOptions): void {
   const games = opts.initialGames;
   const lobbies = opts.initialLobbies;
   const sessions = opts.initialSessions;
+  /**
+   * Outstanding seat recovery codes — Gitea#33, `claims.ts`.
+   *
+   * In memory and not on the volume, deliberately: a code is minted on demand and spent within
+   * minutes with the administrator standing right there, so a restart dropping them all is the right
+   * failure. Persisting them would put a credential-equivalent on disk to solve a problem measured
+   * in seconds.
+   */
+  const claims = createClaimStore();
   const gameCodes = new Map<string, string>(); // gameCode -> gameId, for /api/lobby/join
   for (const [gameId, lobby] of lobbies) gameCodes.set(lobby.gameCode, gameId);
 
@@ -322,6 +332,17 @@ export function startServer(opts: ServerOptions): void {
             gameId,
             gameCode: codes.get(gameId) ?? null,
             state: 'running' as const,
+            /**
+             * WHICH SEATS A PERSON IS SITTING IN — Gitea#33.
+             *
+             * `playerNames` cannot answer it: a bot's name is just a name, and telling the two apart
+             * by matching "Bot 1" would be guessing at a label. `sessions` holds humans and only
+             * humans, so this is the fact rather than an inference — and it is what lets the seat
+             * recovery action offer real players instead of chairs no token was ever issued for.
+             */
+            seatedPlayers: [...sessions.values()]
+              .filter((s) => s.gameId === gameId)
+              .map((s) => s.player),
             ...g.summary(),
           }));
           // A lobby has no game to summarize yet — it is reported as what it is, so an
@@ -340,14 +361,48 @@ export function startServer(opts: ServerOptions): void {
           return;
         }
 
-        const match = /^\/api\/games\/([^/]+)(\/save)?$/.exec(url.pathname);
+        const match = /^\/api\/games\/([^/]+)(\/save|\/claim)?$/.exec(url.pathname);
         const gameId = match?.[1];
+        // Compared explicitly rather than tested for truthiness: with two suffixes in the group, a
+        // bare `match?.[2]` would let a GET on `/claim` fall into the `/save` branch below.
+        const suffix = match?.[2];
         if (!gameId) {
           sendJson(res, 404, { error: 'no such route' });
           return;
         }
 
-        if (match?.[2] && req.method === 'GET') {
+        /**
+         * MINT A SEAT RECOVERY CODE FOR ONE PLAYER — Gitea#33.
+         *
+         * The token is the only identity this game has and it lives in one browser's `localStorage`;
+         * lose it and the seat is unreachable, because nothing else here will accept a claim to it.
+         * This is the supported way back, and it is administrative on purpose: whoever runs the
+         * server decides that a particular player has lost their seat, which is a judgement no
+         * automated route can make safely.
+         *
+         * IT HANDS BACK A CODE, NOT THE TOKEN. §1 says keep the token out of URLs, and the code is
+         * going into one. Short-lived and single-use (`claims.ts`), so a link left in a chat window
+         * is worth nothing by the time anyone finds it.
+         */
+        if (suffix === '/claim' && req.method === 'POST') {
+          const body = (await readJson(req)) as { player?: number };
+          const ps = [...sessions.values()].find((s) => s.gameId === gameId && s.player === body.player);
+          if (!ps) {
+            sendJson(res, 404, { error: 'no such seat' });
+            return;
+          }
+          const { code, expiresAt } = claims.mint(ps.token, gameId, Date.now());
+          sendJson(res, 200, {
+            code,
+            expiresAt,
+            player: ps.player,
+            displayName: ps.displayName,
+            gameCode: codes.get(gameId) ?? null,
+          });
+          return;
+        }
+
+        if (suffix === '/save' && req.method === 'GET') {
           const session = games.get(gameId);
           if (!session) {
             sendJson(res, 404, { error: 'no such game' });
@@ -656,6 +711,57 @@ export function startServer(opts: ServerOptions): void {
         return;
       }
 
+      /**
+       * SPEND A SEAT RECOVERY CODE — Gitea#33, the other half of `/api/games/<id>/claim`.
+       *
+       * NOT GATED BY THE ADMIN SECRET, and it must not be: the player following the link is the one
+       * person in this story who holds no secret at all. The code IS the authorisation — unguessable,
+       * single-use and short-lived — which is the same shape as the session token it hands back, and
+       * why minting one is the administrative act rather than spending one.
+       *
+       * The token travels in the response BODY of a POST, never in a URL (`lobby-and-sessions.md`
+       * §1). One answer for unknown, spent and expired codes, so this cannot be used to probe which.
+       */
+      if (url.pathname === '/api/claim' && req.method === 'POST') {
+        const body = (await readJson(req)) as { code?: string };
+        const claimed = typeof body.code === 'string' ? claims.redeem(body.code, Date.now()) : null;
+        const ps = claimed ? sessions.get(claimed.token) : undefined;
+        const live = ps ? games.get(ps.gameId) : undefined;
+        if (!claimed || !ps || !live) {
+          sendJson(res, 404, { error: 'no such claim' });
+          return;
+        }
+        const codes = new Map((await readIndex(opts.dataDir)).map((e) => [e.gameId, e.gameCode]));
+        sendJson(res, 200, {
+          token: ps.token,
+          gameId: ps.gameId,
+          player: ps.player,
+          gameCode: codes.get(ps.gameId) ?? '',
+        });
+        return;
+      }
+
+      /**
+       * THIS SEAT'S OWN GAME, AS A SAVE (playtest, 2026-09-15: "most of the time, I want to go ahead and
+       * just save it as a JSON file in my Downloads folder").
+       *
+       * The administrative export at `/api/games/<id>/save` is gated on the admin secret, which a player
+       * does not have and should not need: a save is the seed and the moves, and every one of those moves
+       * is already on this player's screen. So the seat's own session token is the gate, exactly as it is
+       * for `/api/stream` and `/api/intent` — it proves which game and which chair, and nothing else is
+       * disclosed. The page turns the JSON into a file (`main.ts`'s `downloadSave`).
+       */
+      if (url.pathname === '/api/save' && req.method === 'GET') {
+        const ps = sessions.get(url.searchParams.get('token') ?? '');
+        const session = ps ? games.get(ps.gameId) : undefined;
+        if (!ps || !session) {
+          sendJson(res, 404, { error: 'no such game' });
+          return;
+        }
+        sendJson(res, 200, { gameId: ps.gameId, save: session.exportSave() });
+        return;
+      }
+
       if (url.pathname === '/api/stream' && req.method === 'GET') {
         const token = url.searchParams.get('token') ?? '';
         const ps = sessions.get(token);

src/sim/board-svg.ts

@@ -42,6 +42,12 @@ export type DivisionRoster = {
   actor: number | null;
   /** The player this map is being drawn for. */
   viewer: number;
+  /**
+   * Division nodes to flash — a Mainline card that has just become a different card (Realignment).
+   * Playtest, 2026-09-15: the log said a card had been converted and the map said nothing, so the one
+   * play that changes the Division itself was invisible on the map of it.
+   */
+  flash?: readonly number[];
 };
 
 export function divisionSvg(nodes: DivisionView[], roster?: DivisionRoster | null): string {
@@ -131,9 +137,13 @@ export function divisionSvg(nodes: DivisionView[], roster?: DivisionRoster | nul
       region?: number;
       direction?: string;
       stagesLeft?: number;
+      /** Being made up at a Division Point right now, so the map can mark the train you are loading. */
+      beingMadeUp?: boolean;
     }[];
     cap: number | null;
     tip: string;
+    /** Set on a card that has just changed under the players' feet — drawn with a brief pulse. */
+    flash?: boolean;
     /** Which SEAT's district this cell belongs to, or null for Mainline and Division Points. */
     seat: number | null;
     /** Set on an Office cell when a roster was supplied: whose district this is. */
@@ -168,7 +178,11 @@ export function divisionSvg(nodes: DivisionView[], roster?: DivisionRoster | nul
     cells.push({ ...c, x: 0, y: 0 });
   };
 
+  // The node's own index, so a cell can be matched against `roster.flash`. `continue` below skips the
+  // rest of the body, never this.
+  let nodeIndex = -1;
   for (const n of nodes) {
+    nodeIndex++;
     if (n.kind === 'office') {
       const cap = n.capacity;
       const ad = n.trains.flat();
@@ -261,6 +275,7 @@ export function divisionSvg(nodes: DivisionView[], roster?: DivisionRoster | nul
     push({
       kind: dp ? 'dp' : 'ml',
       label: n.label,
+      ...(roster?.flash?.includes(nodeIndex) ? { flash: true } : {}),
       sub: n.capacity === null
         ? 'no limit — trains queue'
         : [free, inYard.length > 0 ? `${inYard.length} in the yard` : ''].filter(Boolean).join(' · '),
@@ -356,7 +371,7 @@ export function divisionSvg(nodes: DivisionView[], roster?: DivisionRoster | nul
 
   cells.forEach((c) => {
     const full = c.cap !== null && c.trains.length >= c.cap;
-    out += `<g class="bs-dcell bs-d${c.kind}${full ? ' bs-full' : ''}" data-tip="${esc(c.tip)}">`;
+    out += `<g class="bs-dcell bs-d${c.kind}${full ? ' bs-full' : ''}${c.flash ? ' bs-changed' : ''}" data-tip="${esc(c.tip)}">`;
     out += `<rect x="${c.x}" y="${c.y}" width="${c.w}" height="${CH}" rx="5"/>`;
     /**
      * WHOSE IS IT, IS IT THEIR MOVE, AND IS IT MINE — answered by colour and one suffix rather
@@ -502,13 +517,18 @@ export function divisionSvg(nodes: DivisionView[], roster?: DivisionRoster | nul
       const arrow = t.facing === 'w' ? '\u25c0' : '\u25b6';
       const loaded = cars.filter((x) => /^loaded/.test(x) || /caboose/.test(x)).length;
       const label = cars.length === 0 ? `${t.label} ${arrow}` : `${t.label} ${arrow}${cars.length}`;
+      // THE TRAIN THE MAKE-UP PANEL IS TALKING ABOUT. Amber, because that is what the rest of the
+      // page uses for "this is the thing you are acting on" (Jesse, playtest 2026-09-16).
+      const building = t.beingMadeUp === true;
       const inRegion = c.regions > 1 && typeof t.region === 'number';
       const dir = t.direction === 'west' ? ' \u25c0 west' : t.direction === 'east' ? ' east \u25b6' : '';
       const stages =
         typeof t.stagesLeft === 'number'
           ? ` \u00b7 ${t.stagesLeft} Stage${t.stagesLeft === 1 ? '' : 's'} still to run across this card`
           : '';
-      out += `<g class="bs-train" data-tip="${esc(t.label)} \u2014 carrying ${esc(cars.join(', ') || 'no cars')}${
+      out += `<g class="bs-train${building ? ' bs-building' : ''}" data-tip="${esc(t.label)}${
+        building ? ' \u2014 BEING MADE UP NOW: add cars from the Division Yard' : ''
+      } \u2014 carrying ${esc(cars.join(', ') || 'no cars')}${
         cars.length ? ` (${loaded} loaded)` : ''
       }${inRegion ? ` \u00b7 region ${(t.region ?? 0) + 1} of ${c.regions}, counted west to east${dir}` : ''}${esc(stages)}${
         // What the card prints. A train on the Mainline is exactly where "why did that leave without
@@ -669,7 +689,17 @@ export function officeSvg(
   const c0 = Math.min(...cols);
   const c1 = Math.max(...cols);
   const width = (c1 - c0 + 1) * (W + PAD);
-  const height = (r1 - r0 + 1) * (H + PAD) + 4;
+  /**
+   * A BAND BENEATH THE BOTTOM ROW FOR THE LIMITS LABELS, and only when there are labels to put in it.
+   *
+   * The bottom card's lower edge lands at `height - 7`, and the label's baseline was `height - 4` —
+   * so its 8px glyphs spanned `height - 12` to `height - 4` and the card's own border ran straight
+   * through the middle of the word (Jesse, playtest 2026-09-16: *"the text is split by the bottom
+   * border of the limits card… it should be printed directly beneath the card"*). Raising the text
+   * instead would have pushed it onto the card, over the rails; the room has to be made below.
+   */
+  const limitBand = limits ? 14 : 0;
+  const height = (r1 - r0 + 1) * (H + PAD) + 4 + limitBand;
 
   // Screen position of a card. Rows count DOWN from the top row, so the Running Track sits highest
   // and the district hangs beneath it, as the rules describe it.
@@ -1205,7 +1235,9 @@ export function officeSvg(
   if (limits) {
     const edge = (x: number, side: string): string =>
       `<line class="bs-limitline" x1="${x}" y1="0" x2="${x}" y2="${height}"/>` +
-      `<text class="bs-limitlab" x="${x + (side === 'w' ? 4 : -4)}" y="${height - 4}" ` +
+      // Baseline inside the band below the cards: the glyphs run from `height - 19` to `height - 11`
+      // and the bottom row's edge is at `height - 21`, so the whole word clears the card border.
+      `<text class="bs-limitlab" x="${x + (side === 'w' ? 4 : -4)}" y="${height - 11}" ` +
       `text-anchor="${side === 'w' ? 'start' : 'end'}">LIMITS</text>`;
     out += edge(px(limits.west) - PAD / 2, 'w') + edge(px(limits.east) + W + PAD / 2, 'e');
   }
@@ -1248,9 +1280,18 @@ export const BOARD_CSS = `
    and leave at the other, and a seated layout must not be read as a ring. */
 .bs-stop line{stroke:#e0a060;stroke-width:2.6;stroke-linecap:round}
 .bs-end{fill:#e0a060;font:10px ui-monospace,monospace;letter-spacing:.03em}
-/* The vertical bars a Mainline card is divided into (§2.1). Drawn faint: they are the ruler the
-   train is measured against, not something to look at instead of the train. */
-.bs-region{stroke:#4a5361;stroke-width:1.2;stroke-dasharray:3 3}
+/* A card that has just BECOME a different card (Realignment). The same amber the rest of the page
+   spends on "it is happening here", pulsing only while the step that did it is on screen — so the
+   change is seen on the map rather than only read in the log. */
+.bs-dcell.bs-changed rect{stroke:#e0a060;stroke-width:2.4;animation:bs-changed-pulse 1.1s ease-in-out infinite}
+@keyframes bs-changed-pulse{0%,100%{stroke-opacity:1}50%{stroke-opacity:.35}}
+@media (prefers-reduced-motion: reduce){.bs-dcell.bs-changed rect{animation:none}}
+/* The vertical bars a Mainline card is divided into (§2.1). They are the ruler the train is measured
+   against, not something to look at instead of the train — but they were drawn so faint they could
+   not be made out at all (Jesse, playtest 2026-09-16: "the dividing line is barely visible"). A
+   ruler you cannot read is not restraint, so this is lifted to the tie colour and given a longer
+   dash: still quieter than the rail, and now actually there. */
+.bs-region{stroke:#98a3b2;stroke-width:1.6;stroke-dasharray:4 2}
 /* #94 — the one red mark on the Division map, so it reads as a stop rather than as decoration. */
 .bs-flag line{stroke:#9aa3b0;stroke-width:1.6}
 .bs-flag polygon{fill:#d2453f;stroke:#7d211d;stroke-width:0.8}
@@ -1283,6 +1324,10 @@ export const BOARD_CSS = `
 .bs-slot.bs-car-cch.bs-loaded{fill:rgba(90,169,230,.85)}
 .bs-slot.bs-car-cab.bs-loaded{fill:rgba(192,90,90,.85)}
 .bs-train rect{fill:#2f6b3d;stroke:#8fd6a0;stroke-width:1.2}
+/* The train the New Train phase is loading, in the page's action amber, so the make-up panel on the
+   right and the train on the map at the top left are visibly the same subject. */
+.bs-train.bs-building rect{fill:#4a3a1c;stroke:#c8912f;stroke-width:2}
+.bs-train.bs-building .bs-tlab{fill:#f2d49a}
 .bs-crew rect{fill:#8a6d1f;stroke:#e0c060;stroke-width:1.2}
 /* Each car in the train, in the order it is seated. Loaded is solid, empty is hollow, and the
    engine is the one that carries the arrow — which is what makes "reverse" mean something. */
@@ -1329,7 +1374,10 @@ export const BOARD_CSS = `
 .bs-arrow{fill:#5f6b7a;font:10px ui-monospace,monospace}
 .bs-cn{fill:#e6e9ee;font:600 11px ui-monospace,monospace}
 .bs-coord{fill:#5f6b7a;font:9px ui-monospace,monospace}
-.bs-name{fill:#e6e9ee;font:600 11px ui-monospace,monospace}
+/* stroke:none (Gitea#24). A name takes the class \`bs-turn\` while it is that player's move, and \`.bs-turn\` is
+   also the turn ARROW's rule, which strokes its shape 2.4px grey. Declared after it, this keeps that
+   outline off the letters, which it smeared into an unreadable blur. */
+.bs-name{fill:#e6e9ee;stroke:none;font:600 11px ui-monospace,monospace}
 .bs-name.bs-you{fill:#5aa9e6}
 /* Their move — wins over .bs-you when both apply, because whose turn it is changes every few
    seconds and which railroad is yours never does.

src/sim/bot.ts

@@ -22,6 +22,9 @@
 import {
   applyIntent,
   areaOf,
+  extendLimitsIfNeeded,
+  isLockedOut,
+  protoCard,
   canAdvanceLoad,
   destinationsFor,
   facilityCarTypes,
@@ -29,14 +32,15 @@ import {
   ownCutFor,
 } from '../engine/apply.ts';
 import { MAX_CONSIST, nextOfficeTier, officeProfile } from '../engine/content.ts';
-import type { CarType, Hand, TrackGeometry } from '../engine/content.ts';
+import type { CarType, FreightKind, Hand, TrackGeometry } from '../engine/content.ts';
 import type { GameEvent } from '../engine/events.ts';
 import type { Intent } from '../engine/intents.ts';
 import { legalActions } from '../engine/legal.ts';
-import { connectionsFor, exitsFrom, facilityVariants, hasPort, joins, neighbour, opposite, variantsFor } from '../engine/track.ts';
+import { canPlaceAt, connectionsFor, exitsFrom, facilityVariants, hasPort, joins, neighbour, opposite, variantsFor } from '../engine/track.ts';
 import type { Port } from '../engine/track.ts';
 import { actingPlayer, coordKey, turnOf } from '../engine/state.ts';
 import type { Facility, GameState, GridCoord, OfficeArea, PlayerIndex, RollingStock, TrackCard } from '../engine/state.ts';
+import { planSwitchingTurn, switchFingerprint } from './switch-planner.ts';
 
 export type BotPolicy = {
   name: string;
@@ -106,7 +110,8 @@ function because(reason: string, intent: Intent): Intent {
  * Every flag here turns something OFF. That is the opposite of how this started — the tweaks were
  * candidates to switch on — and it is the right shape once a candidate has been adopted: what a
  * measured heuristic needs afterwards is a way to ask "is this still worth it?" when the deck or
- * the rules move under it. Both of these were worth about +1.5 revenue together when adopted; if a
+ * the rules move under it. The first two were worth about +1.5 revenue together when adopted, and
+ * planning the switching turn (`noPlanSwitching`) +2.89 on its own; if a
  * rebalance changes the economy, that is a claim to re-test rather than to assume.
  *
  * The candidates that did NOT survive are gone rather than left switched off: preferring coaches at
@@ -121,9 +126,78 @@ export type BotTweaks = {
   noTrainCap?: boolean;
   /** Draw whenever nothing is urgent, as the bot did before it preferred operating. */
   noOperateFirst?: boolean;
-
+  /**
+   * Choose switching Moves one at a time from the rule ladder, as the bot did before it planned the
+   * whole turn (`switch-planner.ts`). Measured at adoption, 2026-09-14: planning was worth
+   * +2.89 ± 0.18 revenue a game (t = 15.79) over 1600 paired seeds, 733 better against 21 worse.
+   */
+  noPlanSwitching?: boolean;
+  /**
+   * Take a face-up train or industry card whether or not it could be played, as the bot did before
+   * 2026-09-14. It then took 20.1 trains and 11.9 industries a game off the Departments and discarded
+   * 20.2 and 11.8, retaking the same card 28.8 times a game. Asking first measured +1.52 ± 0.10
+   * (t = 15.59) over 1600 paired seeds — this ablation was worse on 880 of them and better on 211.
+   */
+  noPlayableTakes?: boolean;
+  /**
+   * Choose where track goes by `bestTrackLay`'s piece rules and the fallback's first legal square, as the
+   * bot did before 2026-09-15, instead of by what the district can DO afterwards (`bestValuedLay`).
+   * Scoring the layout measured +0.118 ± 0.029 (t = 4.14) over 6400 paired seeds, and closed run-arounds
+   * in 22 of 60 districts against 9.
+   */
+  noValueLays?: boolean;
+  /**
+   * Let the New Train phase's fallback take `options[0]`, as the bot did before 2026-09-15. Because
+   * `legalActions` lists `newTrain.secondSection` ahead of the Extra starts, all 26 Second Sections the
+   * bot ran in 60 games were that accident, and 6 of the 20 collisions followed one. Taking a car, a pass
+   * or the Extra's start instead measured +0.32 ± 0.09 (t = 3.64) over 400 paired seeds.
+   */
+  noDeliberateNewTrain?: boolean;
 };
 
+/**
+ * A switching turn planned once and then played a step per decision.
+ *
+ * Keyed by the tweaks object, because that is what one policy owns — the server shares a single
+ * `developerBot` across every bot seat, so the plan inside it is kept per player. Each step is
+ * submitted only while the position still matches the fingerprint the plan expected there; anything
+ * else replans. A switching turn has no randomness, so in practice a plan is made once a turn.
+ */
+type ActivePlan = { steps: Intent[]; keys: string[]; next: number; summary: string };
+const activePlans = new WeakMap<BotTweaks, Map<PlayerIndex, ActivePlan>>();
+
+function plannedSwitch(s: GameState, player: PlayerIndex, options: Intent[], tweaks: BotTweaks): Intent | null {
+  let mine = activePlans.get(tweaks);
+  if (!mine) activePlans.set(tweaks, (mine = new Map()));
+  const here = switchFingerprint(s, player);
+  let active = mine.get(player);
+  if (!active || active.keys[active.next] !== here) {
+    const p = planSwitchingTurn(s, player);
+    active = {
+      steps: p.steps,
+      keys: p.keys,
+      next: 0,
+      summary:
+        `position ${p.rootScore.toFixed(2)} → ${p.score.toFixed(2)} over ${p.expanded} positions` +
+        (p.complete ? '' : ', search budget reached'),
+    };
+    mine.set(player, active);
+  }
+  if (active.next >= active.steps.length) {
+    mine.delete(player);
+    const end = options.find((i) => i.type === 'switch.end');
+    return end ? because(`planned switching turn complete — ${active.summary}`, end) : null;
+  }
+  const want = JSON.stringify(active.steps[active.next]);
+  const match = options.find((i) => JSON.stringify(i) === want);
+  if (!match) {
+    mine.delete(player);
+    return null;
+  }
+  active.next++;
+  return because(`step ${active.next} of ${active.steps.length} of a planned switching turn — ${active.summary}`, match);
+}
+
 /** The bot as it plays today. Every knob off. */
 export const developerBot: BotPolicy = makeDeveloperBot({});
 
@@ -209,6 +283,13 @@ export function makeDeveloperBot(tweaks: BotTweaks): BotPolicy {
         );
         if (match) return because(`the ${w.loaded ? 'loaded' : 'empty'} ${w.type} is what a facility is short of`, match);
       }
+      if (!tweaks.noDeliberateNewTrain) {
+        const move =
+          pickFirst(options, 'newTrain.placeCar', 'newTrain.passCar', 'newTrain.startExtra') ??
+          options.find((i) => i.type !== 'newTrain.secondSection' && i.type !== 'maneuver.redFlags') ??
+          options[0]!;
+        return because('no car on offer is one our facilities need', move);
+      }
       return because('no car on offer is one our facilities need', pickFirst(options, 'newTrain.placeCar', 'newTrain.passCar') ?? options[0]!);
     }
 
@@ -438,7 +519,7 @@ function topOfDepartment(s: GameState, slot: number): string | undefined {
  * In a competitive game the same call reads the other way round — burying a card a rival wants is an
  * attack — which is why the choice belongs to the discarding player and not to the rules.
  */
-function bestDiscard(s: GameState, player: PlayerIndex, options: Intent[]): Intent | null {
+function bestDiscard(s: GameState, player: PlayerIndex, options: Intent[], tweaks: BotTweaks): Intent | null {
   let best: Intent | null = null;
   let bestScore = -Infinity;
   for (const i of options) {
@@ -448,7 +529,7 @@ function bestDiscard(s: GameState, player: PlayerIndex, options: Intent[]): Inte
     // two showing whatever they happened to start with. Measured over 100 games — spreading 2.87
     // revenue, concentrating on the deepest 2.67, indifferent 2.67.
     const top = topOfDepartment(s, i.toSlot);
-    const wanted = isWorthTaking(s, player, i.toSlot);
+    const wanted = isWorthTaking(s, player, i.toSlot, tweaks);
     const depth = s.decks.departments[i.toSlot]?.length ?? 0;
     const score = (top === undefined ? 6 : wanted ? -10 : 2) - Math.min(depth, 4) * 0.5;
     if (score > bestScore) {
@@ -460,10 +541,39 @@ function bestDiscard(s: GameState, player: PlayerIndex, options: Intent[]): Inte
 }
 
 /** A face-up card worth spending the draw on rather than gambling on the deck. */
-function isWorthTaking(s: GameState, player: PlayerIndex, slot: number): boolean {
-  return takingRank(s, player, slot) > 0;
+function isWorthTaking(s: GameState, player: PlayerIndex, slot: number, tweaks: BotTweaks): boolean {
+  return takingRank(s, player, slot, tweaks) > 0;
 }
 
+/**
+ * Could an industry of this kind be laid anywhere right now? Asked of the engine's own placement
+ * rule (`canPlaceAt`) and lockout (`isLockedOut`) rather than a copy: an industry is plain east-west
+ * track, so the only squares worth asking about are empty ones east or west of a card already down.
+ */
+function industrySiteExists(s: GameState, player: PlayerIndex, kind: FreightKind): boolean {
+  const area = areaOf(s, player);
+  if (isLockedOut(area, kind)) return false;
+  const probe = {
+    geometry: { kind: 'facility', facility: kind },
+    baseOperationalRail: true,
+    standing: [],
+    standingWest: 0,
+    facility: null,
+    modifiers: [],
+    enhancements: [],
+  } as unknown as TrackCard;
+  for (const key of area.grid.keys()) {
+    const [row, col] = key.split(',').map(Number);
+    for (const dc of [1, -1]) {
+      const at = { row: row!, col: col! + dc };
+      if (at.row === area.runningRow || area.grid.has(coordKey(at))) continue;
+      if (canPlaceAt(area, at, probe)) return true;
+    }
+  }
+  return false;
+}
+
+
 /**
  * HOW BADLY the face-up card is wanted. 0 means not worth the draw.
  *
@@ -471,14 +581,18 @@ function isWorthTaking(s: GameState, player: PlayerIndex, slot: number): boolean
  * happened to be scanned first — a coin flip on the card that decides whether the district ever
  * becomes a Passenger Facility at all.
  */
-function takingRank(s: GameState, player: PlayerIndex, slot: number): number {
+function takingRank(s: GameState, player: PlayerIndex, slot: number, tweaks: BotTweaks): number {
   const id = topOfDepartment(s, slot);
   if (!id) return 0;
   const k = s.cards.get(id)?.kind;
   if (!k) return 0;
   if (k.kind === 'office') return nextOfficeTier(areaOf(s, player).tier) === k.tier ? 3 : 0;
-  if (k.kind === 'timetabledTrain' || k.kind === 'extraTrain') return 2;
-  if (k.kind === 'freightFacility') return 1;
+  if (k.kind === 'timetabledTrain' || k.kind === 'extraTrain') {
+    return !tweaks.noPlayableTakes && trainWouldOverfillTheOffice(s, player, tweaks) ? 0 : 2;
+  }
+  if (k.kind === 'freightFacility') {
+    return !tweaks.noPlayableTakes && !industrySiteExists(s, player, k.facility) ? 0 : 1;
+  }
   return 0;
 }
 
@@ -741,6 +855,104 @@ function bestFacilityPlay(s: GameState, player: PlayerIndex, options: Intent[]):
   return best;
 }
 
+/**
+ * WHAT A DISTRICT'S TRACK IS WORTH FOR WHAT IT LETS HAPPEN NEXT — the default since 2026-09-15;
+ * `noValueLays` turns it off.
+ *
+ * `bestTrackLay` scores the PIECE — its shape and where it sits — and cannot tell one that opens an
+ * industry site or closes a run-around from one that merely fills a square. This scores the LAYOUT the
+ * piece would leave, so a lay is worth the difference it makes. Every term is something the rules turn
+ * into play: a site is somewhere a held industry can go; a run-around lets a crew pass its own cars
+ * (§A.5); a way off the main is the only road to either; a Running Track straight is what Interlocking
+ * needs. The weights are a starting point to measure, not a result.
+ */
+function layoutValue(area: OfficeArea): number {
+  let v = 0;
+  const reachable = reachableOffMain(area);
+  v += Math.min(reachable.size, 12) * 0.2;
+
+  let ways = 0;
+  let loops = 0;
+  for (const side of SIDES) {
+    for (const col of waysOff(area, side)) {
+      ways++;
+      if (descendFrom(area, col, side).rejoins.size > 0) loops++;
+    }
+  }
+  v += [0, 1.5, 2, 2.5][Math.min(ways, 3)]!;
+  if (loops > 0) v += 6 + Math.min(loops - 1, 1) * 2;
+
+  // Squares an industry could legally be laid on, joined to track a crew can reach.
+  const probe = protoCard({ kind: 'freightFacility', facility: 'mineTipple' }, 0)!;
+  const tried = new Set<string>();
+  let sites = 0;
+  for (const key of reachable) {
+    const [row, col] = key.split(',').map(Number);
+    for (const dc of [1, -1]) {
+      const at = { row: row!, col: col! + dc };
+      const k = coordKey(at);
+      if (tried.has(k) || area.grid.has(k) || at.row === area.runningRow) continue;
+      tried.add(k);
+      if (canPlaceAt(area, at, probe)) sites++;
+    }
+  }
+  v += [0, 2, 3, 3.5][Math.min(sites, 3)]!;
+
+  let mainStraight = false;
+  for (const [key, card] of area.grid) {
+    if (Number(key.split(',')[0]) !== area.runningRow || card.geometry.kind !== 'track') continue;
+    if (card.geometry.geometry === 'straight') mainStraight = true;
+    // A turnout on the main whose leg joins nothing is a hole in the Running Track with no road behind it.
+    if (card.geometry.geometry === 'turnout') {
+      const col = Number(key.split(',')[1]);
+      for (const side of SIDES) {
+        if (!hasPort(card, legPort(side))) continue;
+        const beyond = area.grid.get(`${area.runningRow + side},${col}`);
+        if (!beyond || !joins(card, legPort(side), beyond)) v -= 0.5;
+      }
+    }
+  }
+  if (mainStraight) v += 1.5;
+  return v;
+}
+
+/**
+ * The track lay worth most by `layoutValue`, placed on a copy of the district exactly as the reducer
+ * places it (`protoCard`, `extendLimitsIfNeeded`). With `mustBuild`, only a lay that gains something is
+ * offered, which is the slot `bestTrackLay` fills; without it, the best of whatever is legal, which is
+ * the slot the "play what is in hand" fallback fills. Ties go to the square nearer the Office.
+ */
+function bestValuedLay(s: GameState, player: PlayerIndex, options: Intent[], mustBuild: boolean): Intent | null {
+  const area = areaOf(s, player);
+  const base = layoutValue(area);
+  let best: Intent | null = null;
+  let bestScore = -Infinity;
+  for (const i of options) {
+    if (i.type !== 'card.play' || i.placement === undefined) continue;
+    const kind = s.cards.get(i.cardId)?.kind;
+    if (kind?.kind !== 'track') continue;
+    const built = protoCard(kind, i.variant);
+    if (!built) continue;
+    const after: OfficeArea = {
+      ...area,
+      grid: new Map(area.grid),
+      limitsWest: { ...area.limitsWest },
+      limitsEast: { ...area.limitsEast },
+    };
+    after.grid.set(coordKey(i.placement), built);
+    extendLimitsIfNeeded(after, i.placement);
+    const gain = layoutValue(after) - base;
+    if (mustBuild && gain <= 0.1) continue;
+    const distance = Math.abs(i.placement.row - area.officeCoord.row) * 2 + Math.abs(i.placement.col - area.officeCoord.col);
+    const score = gain - distance * 0.01;
+    if (score > bestScore) {
+      bestScore = score;
+      best = i;
+    }
+  }
+  return best;
+}
+
 function bestTrackLay(s: GameState, player: PlayerIndex, options: Intent[]): Intent | null {
   const area = areaOf(s, player);
 
@@ -1213,7 +1425,7 @@ function followThrough(
       if (!turnOf(s, player).drawnThisTurn) {
         const piles = options.filter(
           (i): i is Extract<Intent, { type: 'draw.fromDepartment' }> =>
-            i.type === 'draw.fromDepartment' && isWorthTaking(s, player, i.slot),
+            i.type === 'draw.fromDepartment' && isWorthTaking(s, player, i.slot, tweaks),
         );
         // Best-ranked pile rather than the first that qualifies: an Office card and a train card
         // both "qualify", and only one of them stops the collisions.
@@ -1221,7 +1433,7 @@ function followThrough(
         // and a train card are face up together 1.6 decisions a game — but ranking them is what the
         // ranking function is for, and a coin flip on the card that decides whether the district
         // ever becomes a Passenger Facility is not worth keeping for its own sake.
-        const useful = piles.sort((a, b) => takingRank(s, player, b.slot) - takingRank(s, player, a.slot))[0];
+        const useful = piles.sort((a, b) => takingRank(s, player, b.slot, tweaks) - takingRank(s, player, a.slot, tweaks))[0];
         if (useful) return because('a face-up card is worth more than a blind draw right now', useful);
         const blind = options.find((i) => i.type === 'draw.fromHomeOffice');
         if (blind) return because('no face-up card is worth taking — gamble on the deck', blind);
@@ -1282,7 +1494,7 @@ function followThrough(
       // STRAIGHTS that Enhancements require, and no Freight Facility has anywhere to go until a
       // district exists. Measured with track absent, the hand held a playable Enhancement on 4,778
       // turns and could legally place one on 33.
-      const track = bestTrackLay(s, player, options);
+      const track = !tweaks.noValueLays ? bestValuedLay(s, player, options, true) : bestTrackLay(s, player, options);
       if (track) return because('lay track — nothing else creates the straights Enhancements need or the spurs freight needs', track);
 
       // Then real development: a card actually laid into the grid. Freight facilities are scored —
@@ -1307,17 +1519,27 @@ function followThrough(
           s.cards.get(i.cardId)?.kind.kind !== 'track',
       );
       if (placed) return because('develop the district with a card that goes on the board', placed);
-      const play = options.find((i) => i.type === 'card.play');
+      const play = !tweaks.noValueLays
+        ? options.find((i) => i.type === 'card.play' && s.cards.get(i.cardId)?.kind.kind !== 'track') ??
+          bestValuedLay(s, player, options, false)
+        : options.find((i) => i.type === 'card.play');
       if (play) return because('play what is in hand', play);
       const end = options.find((i) => i.type === 'draw.end');
       if (end) return because('nothing in hand can be played anywhere legal', end);
       return because(
         'nothing playable — discard onto the Department whose face-up card is least worth keeping reachable',
-        bestDiscard(s, player, options) ?? pickFirst(options, 'card.discard') ?? options[0]!,
+        bestDiscard(s, player, options, tweaks) ?? pickFirst(options, 'card.discard') ?? options[0]!,
       );
     }
 
     case 'switch': {
+      // The planner decides the whole turn; the rules below are its fallback if the position is ever
+      // not the one it planned for, and the whole of switching under `noPlanSwitching`.
+      if (!tweaks.noPlanSwitching) {
+        const planned = plannedSwitch(s, player, options, tweaks);
+        if (planned) return planned;
+      }
+
       /**
        * A MOVE THAT DRAGS THE CREW'S OWN CUT BACK ON IS A WASTED MOVE, so take those off the table
        * before any heuristic gets to choose one.

src/sim/compare.ts

@@ -23,6 +23,7 @@
  * Run with:
  *   node src/sim/compare.ts 1600 noTrainCap=1        — what the A/D cap is worth today
  *   node src/sim/compare.ts 1600 noOperateFirst=1    — what operating before drawing is worth
+ *   node src/sim/compare.ts 1600 noPlanSwitching=1   — what planning the switching turn is worth
  *
  * The flags are ABLATIONS: they turn off heuristics the bot already plays, so a negative delta is
  * the heuristic earning its place. That is what a measured bot needs going forward — the question
@@ -221,7 +222,7 @@ export function formatPaired(r: PairedResult): string {
  * against itself and report a confident zero, which is the most expensive way this tool could fail.
  */
 export const NUMERIC_TWEAKS = new Set<string>([]);
-export const BOOLEAN_TWEAKS = new Set(['noTrainCap', 'noOperateFirst']);
+export const BOOLEAN_TWEAKS = new Set(['noTrainCap', 'noOperateFirst', 'noPlanSwitching', 'noPlayableTakes', 'noValueLays', 'noDeliberateNewTrain']);
 
 export function parseTweaks(args: string[]): BotTweaks {
   const tweaks: Record<string, number | boolean> = {};

src/sim/narrate.ts

@@ -15,7 +15,7 @@
  * panel cannot drift from the rules.
  */
 
-import { MAX_CONSIST, crewTrayCount } from '../engine/content.ts';
+import { MAINLINE_PROFILES, MAX_CONSIST, crewTrayCount } from '../engine/content.ts';
 import { adTrackCount, coordKey, seatOf, turnOf } from '../engine/state.ts';
 import type { GameState, GridCoord, PlayerIndex, RollingStock, SeatIndex, TrayId } from '../engine/state.ts';
 import { areaOf, canAdvanceLoad, canBoard, canDetrain, canStartLoad, facilityCarType, facilityCarTypes, freightRuleSpentHere, isFreight, laborersLeft, movesFor, passengerRefusal, portersLeft } from '../engine/apply.ts';
@@ -134,7 +134,7 @@ export function narrate(e: GameEvent, ctx: NarrateContext = {}): Narration {
     case 'actorChanged':
       return {
         tone: 'quiet',
-        text: e.player === null ? 'No player acts — automatic phase' : `Player ${e.player} to act`,
+        text: e.player === null ? 'No player acts — automatic phase' : 'to act',
       };
 
     // -- local operations
@@ -211,13 +211,19 @@ export function narrate(e: GameEvent, ctx: NarrateContext = {}): Narration {
           ? `Played ${card(e.cardId)} onto ${at(e.placement)}`
           : `Played ${card(e.cardId)}`,
       };
-    case 'mainlineModified':
+    case 'mainlineModified': {
+      // WHICH CARD, NOT JUST WHICH WAY IT WENT. "Mainline card 3 converted to plains" left the reader
+      // to remember what card 3 had been (playtest, 2026-09-15), and the card it WAS is the half that
+      // says what the play was worth.
+      const kindName = (k: string | undefined): string =>
+        MAINLINE_PROFILES.find((m) => m.kind === k)?.name ?? k ?? 'that card';
       return {
         tone: 'plain',
         text: e.became
-          ? `Realignment: Mainline card ${e.node} converted to ${e.became}`
-          : `Played ${e.key} on Mainline card ${e.node}`,
+          ? `Realignment: Mainline card ${e.node}, ${kindName(e.from)}, converted to ${kindName(e.became)}`
+          : `Played ${e.key} on Mainline card ${e.node}${e.from ? ` (${kindName(e.from)})` : ''}`,
       };
+    }
     case 'redFlagSpent':
       return {
         tone: 'good',
@@ -225,8 +231,8 @@ export function narrate(e: GameEvent, ctx: NarrateContext = {}): Narration {
       };
     case 'redFlagRuled':
       return e.flag
-        ? { tone: 'plain', text: `Player ${e.player} flagged the approaching train` }
-        : { tone: 'plain', text: `Player ${e.player} waved the train through` };
+        ? { tone: 'plain', text: 'Flagged the approaching train' }
+        : { tone: 'plain', text: 'Waved the train through' };
     case 'redFlagsSet':
       return {
         tone: 'good',
@@ -337,12 +343,25 @@ export function narrate(e: GameEvent, ctx: NarrateContext = {}): Narration {
        * arrival; the only difference is what happens if it is left on Secondary Track when the next
        * Mainline Phase begins (`expediteFault`).
        */
-      return {
-        tone: 'plain',
-        text: e.expedited
-          ? `Train ${e.trainNumber} ARRIVED at the ${e.office} carrying ${carsLabel(e.consist)} — its card prints EXPEDITE, so keep it on the Office square: parked anywhere else in the district when the next Mainline Phase begins costs a Revenue point. It works and switches normally in the meantime.`
-          : `Train ${e.trainNumber} ARRIVED at the ${e.office} carrying ${carsLabel(e.consist)} — it stands here for the rest of this Stage. You can work it in Cargo now, switch it in the NEXT Stage's Local Operations, and it departs in that Stage's Mainline Phase.`,
-      };
+      /**
+       * WHOSE OFFICE, AND WHOSE TRAIN TO WORK — Jesse, playtest 2026-09-16.
+       *
+       * This said "ARRIVED at the Whistle Post" and then "You can work it in Cargo now". Both halves
+       * are wrong at a table of four: every seat has an Office, so the tier alone does not say which
+       * district the train is standing in, and the reader is usually NOT its Station Master — the
+       * line was telling three players they could work a train they cannot touch.
+       */
+      {
+        const name = ctx.playerName?.(e.owner) ?? null;
+        const whose = name === null ? `the ${e.office}` : `${name}'s ${e.office}`;
+        const worker = name === null ? 'Its Station Master' : name;
+        return {
+          tone: 'plain',
+          text: e.expedited
+            ? `Train ${e.trainNumber} ARRIVED at ${whose} carrying ${carsLabel(e.consist)} — its card prints EXPEDITE, so it must stay on the Office square: parked anywhere else in that district when the next Mainline Phase begins costs a Revenue point. It works and switches normally in the meantime.`
+            : `Train ${e.trainNumber} ARRIVED at ${whose} carrying ${carsLabel(e.consist)} — it stands there for the rest of this Stage. ${worker} can work it in Cargo now and switch it in the NEXT Stage's Local Operations, and it departs in that Stage's Mainline Phase.`,
+        };
+      }
     case 'expediteFault':
       return {
         tone: 'bad',
@@ -496,13 +515,13 @@ export function narrate(e: GameEvent, ctx: NarrateContext = {}): Narration {
         ? { tone: 'bad', text: `${e.reason.toUpperCase()} · ${e.delta} Revenue (now ${e.total})` }
         : { tone: 'good', text: `+${e.delta} Revenue (now ${e.total}) — ${e.reason}` };
     case 'phaseEnded':
-      return { tone: 'quiet', text: `Player ${e.player} finished ${phaseLabel(e.phase)}` };
+      return { tone: 'quiet', text: `Finished ${phaseLabel(e.phase)}` };
 
     // -- §3.3, extended play (Gitea#11)
     case 'extensionVoted':
       return e.agree
-        ? { tone: 'plain', text: `Player ${e.player} would play one more Day` }
-        : { tone: 'plain', text: `Player ${e.player} called time — the game ends here` };
+        ? { tone: 'plain', text: 'Would play one more Day' }
+        : { tone: 'plain', text: 'Called time — the game ends here' };
     case 'dayExtended':
       return { tone: 'clock', text: `── The table plays on: Day ${e.day} is added to the timetable ──` };
     case 'playConcluded':
@@ -511,8 +530,8 @@ export function narrate(e: GameEvent, ctx: NarrateContext = {}): Narration {
     // -- §11, the Yard Office (Gitea#5)
     case 'yardOfficeRuled':
       return e.take
-        ? { tone: 'plain', text: `Player ${e.player} sent ${train(e.trainId)} into the Yard Office` }
-        : { tone: 'plain', text: `Player ${e.player} kept ${train(e.trainId)} at the Train Order Office` };
+        ? { tone: 'plain', text: `Sent ${train(e.trainId)} into the Yard Office` }
+        : { tone: 'plain', text: `Kept ${train(e.trainId)} at the Train Order Office` };
   }
 }
 

src/sim/public-delta.ts

@@ -157,6 +157,25 @@ export type PileKey = 'home' | 'salvage' | `dept${number}`;
  * | `card.play`             | `salvage`, or nothing here when it lands on the board      |
  * | switching, new trains   | nothing here — those show on the board itself             |
  */
+/**
+ * Mainline cards that became a different card between two public boards — a Realignment, which is the
+ * one play that changes the Division itself.
+ *
+ * Playtest, 2026-09-15: *"is it possible to flash the mainline card when it gets changed by realignment?
+ * This would be more obvious to see what's happening on the map."* Detected the same way `changedPiles`
+ * detects a pile moving — by comparing the two boards the queue already holds — rather than by reading
+ * the event, so the flash lands with the step that shows it and not when the intent arrived.
+ */
+export function changedDivisionCards(before: PublicFrame | null, after: PublicFrame): number[] {
+  if (before === null) return [];
+  const out: number[] = [];
+  after.division.forEach((node, i) => {
+    const was = before.division[i];
+    if (was && was.kind === 'ml' && node.kind === 'ml' && was.label !== node.label) out.push(i);
+  });
+  return out;
+}
+
 export function changedPiles(before: PublicFrame | null, after: PublicFrame): PileKey[] {
   if (before === null) return [];
   const out: PileKey[] = [];

src/sim/replay.ts

@@ -79,6 +79,8 @@ export function record(seed: number, length: GameLength, maxSteps = 100_000): Re
   const narrateCtx = {
     cardName: (id: string) => cardName(s, id),
     trainName: (id: string) => trainName(s, id),
+    // A player index is not a seat index, so the fallback names no number at all — see `web/game.ts`.
+    playerName: (p: number) => s.players[p]?.name ?? 'another player',
   };
 
   // Tracks whether a phase did anything, so an empty one can say so rather than ending silently.

src/sim/switch-planner.ts

@@ -0,0 +1,347 @@
+/**
+ * Component 17b — planning a whole switching turn before making the first Move.
+ *
+ * Dev-side, like the rest of the bot. The developer bot's switching branch chooses ONE move at a time
+ * from a ladder of rules, and its own comment names what that cannot do: "a strong player would use
+ * the six Moves to re-order the consist — that is the game's central switching puzzle, and this bot
+ * does not attempt it." This attempts it, for one turn at a time.
+ *
+ * WHY SEARCH IS FAIR HERE. A switching turn draws no card and rolls no die, so trying sequences on a
+ * copy of the game is exactly what a player does by looking at the board. The score below reads only
+ * what a player can see — the district, the cars on the trains, the facilities — and never the deck.
+ *
+ * WHY NOT EVERY SEQUENCE. Measured 2026-09-14 over 30 switching turns from bot games: a median turn
+ * reaches 229 distinct positions, but 11 of 30 passed 20,000, because setting cars out is free and a
+ * crew can leave them in a great many places. So the search keeps the best `beam` positions at each
+ * step and stops at `budget` positions tried. Small turns are searched completely inside that.
+ *
+ * THE SCORE IS OF WHERE THE TURN ENDS, not of what it did, and it starts from Jesse's ruling
+ * (2026-09-14): "players will attempt to deliver / pick up cars even if it delays trains." So a car
+ * put where it can be worked is worth a point, and a train left away from the Office costs a quarter
+ * of one. The weights are a starting point to measure, not a result.
+ */
+
+import { areaAtSeat, areaOf, commitEvents, facilityCarTypes, prepareIntent, withRouteCache } from '../engine/apply.ts';
+import { badlyMadeUp, isExpedited } from '../engine/advance.ts';
+import { MAX_CONSIST } from '../engine/content.ts';
+import type { Intent } from '../engine/intents.ts';
+import { legalSwitchingActions } from '../engine/legal.ts';
+import { cloneTally, coordKey, seatOf, turnOf } from '../engine/state.ts';
+import type { Facility, GameState, GridCoord, PlayerIndex, RollingStock, TrackCard } from '../engine/state.ts';
+
+export const SWITCH_WEIGHTS = {
+  /** A car standing where its industry can load or unload it — the point of switching. */
+  spot: 1.0,
+  /** The same, past what the industry's boxes can work at once. */
+  spotBeyondCapacity: 0.25,
+  /** A car the industry cannot work, taking room on its track. */
+  junkOnIndustry: -0.5,
+  /** A finished car — loaded at a shipper, emptied at a receiver — still waiting to be lifted. */
+  finishedLeft: -0.15,
+  /** A car on one of this district's trains that some industry here would work. */
+  carriedWanted: 0.35,
+  /** The same car left on ordinary track, where a later turn can fetch it. */
+  stagedWanted: 0.2,
+  /** A coach kept with its train, or parked at the Office where §A.4 allows it. */
+  coachWithTrain: 0.3,
+  /** A coach left anywhere else, where no Porter can work it. */
+  coachStranded: -0.3,
+  /** Anything but a coach standing on the Office square — the next arrival collides (§8.3). */
+  fouling: -3,
+  /** A train that ends the turn away from the Office and so cannot highball next Mainline Phase. */
+  trainAway: -0.25,
+  /** On top of that, an expedited train — Q3 charges a Revenue point every Phase it is away. */
+  expeditedAway: -1.0,
+  /** A train that could not leave even from the Office — engine buried, caboose mid-train (§8.2). */
+  notMadeUp: -0.6,
+  /** Tie-breaks, so equal outcomes prefer the plan that does less. */
+  perMove: -0.02,
+  perSetOut: -0.005,
+  /** A maneuver card spent — Flying Switch — so the planner plays one only when it buys something. */
+  cardSpent: -0.1,
+} as const;
+
+export type PlanOptions = {
+  budget: number;
+  beam: number;
+  /**
+   * Search Flying Switch alongside Moves, set-outs and sorts. On by default but UNMEASURED: the card
+   * is dealt 0 copies (Jesse, 2026-08-26), so over 400 paired seeds turning it on changed nothing —
+   * it is here so the planner can use the card the day it is dealt again.
+   */
+  flyingSwitch?: boolean;
+};
+/**
+ * Measured 2026-09-14, paired over 400 seeds against 3000/48: 2000/32 cost −0.02 ± 0.01 (t = −1.68,
+ * inside the noise) at half the time per turn; 1000/24 cost −0.06 ± 0.02 (t = −2.65) for little more.
+ */
+export const DEFAULT_PLAN: PlanOptions = { budget: 2000, beam: 32, flyingSwitch: true };
+
+export type SwitchPlan = {
+  /** The intents to submit, in order. Empty when nothing beats stopping where the crew stands. */
+  steps: Intent[];
+  /** `switchFingerprint` before each step, and after the last — so a caller can tell it is on plan. */
+  keys: string[];
+  rootScore: number;
+  score: number;
+  /** Positions tried. */
+  expanded: number;
+  /** False when the budget ran out before the search did. */
+  complete: boolean;
+};
+
+/**
+ * A copy of the game that a switching intent can be applied to without touching the original.
+ *
+ * NOT `structuredClone`, of the state or even of the district. A switching intent writes only the
+ * cars standing on cards, the industry tracks, the district's A/D and held lists, the consist and
+ * position of the trays standing in it, this player's turn and the tally — so exactly those arrays are
+ * copied and everything else is shared by reference. Measured 2026-09-14, deep-cloning the district
+ * was 44% of all planning time.
+ *
+ * `test/switch-planner.test.ts` proves across real games that planning leaves the original
+ * byte-identical — which is what fails first if a reducer ever starts writing somewhere new, or
+ * starts mutating a car or a card in place instead of replacing it.
+ */
+export function forkForSwitching(s: GameState, player: PlayerIndex): GameState {
+  const seat = seatOf(s, player);
+  const area = areaAtSeat(s, seat);
+  const grid = new Map<string, TrackCard>();
+  for (const [key, card] of area.grid) {
+    const f = card.facility;
+    grid.set(key, {
+      ...card,
+      standing: [...card.standing],
+      facility: f ? { ...f, industryTrack: { cars: [...f.industryTrack.cars] } } : f,
+    });
+  }
+  const officeAreas = new Map(s.officeAreas);
+  officeAreas.set(seat, {
+    ...area,
+    grid,
+    adOccupancy: [...area.adOccupancy],
+    heldAtLimits: [...area.heldAtLimits],
+    dispatchUsedToday: [...area.dispatchUsedToday],
+  });
+  const trays = new Map(s.trays);
+  for (const [id, t] of s.trays) {
+    if (t.position.at === 'grid' && t.position.seat === seat) trays.set(id, { ...t, consist: [...t.consist] });
+  }
+  const turns = new Map(s.turns);
+  const turn = s.turns.get(player)!;
+  turns.set(player, { ...turn, freightWorked: { ...turn.freightWorked } });
+  // Flying Switch spends its card (`spendCard`): the hand map is rewritten and the Salvage Yard grows.
+  const decks = { ...s.decks, hands: new Map(s.decks.hands), salvageYard: [...s.decks.salvageYard] };
+  return { ...s, officeAreas, trays, turns, decks, tally: cloneTally(s.tally) };
+}
+
+const carList = (xs: readonly RollingStock[]): string =>
+  xs.map((c) => `${c.type}${c.loaded ? '+' : '-'}${c.origin ?? ''}`).join(',');
+
+/**
+ * Everything a switching intent can change, as a string — two positions with the same fingerprint
+ * are the same position as far as the rest of the turn is concerned. Identical cars are not told
+ * apart, which is right: no intent names a car.
+ */
+export function switchFingerprint(s: GameState, player: PlayerIndex): string {
+  const seat = seatOf(s, player);
+  const parts: string[] = [];
+  for (const [id, t] of s.trays) {
+    if (t.position.at !== 'grid' || t.position.seat !== seat) continue;
+    const { row, col } = t.position.coord;
+    parts.push(`${id}@${row},${col}/${t.facing}/${t.railFacing ?? ''}/${t.engineAt}:${carList(t.consist)}`);
+  }
+  for (const [key, card] of areaOf(s, player).grid) {
+    const track = card.facility?.kind === 'freight' ? card.facility.industryTrack.cars : null;
+    if (card.standing.length === 0 && card.standingWest === 0 && (track?.length ?? 0) === 0) continue;
+    parts.push(`${key}=${carList(card.standing)}|${card.standingWest}|${track ? carList(track) : ''}`);
+  }
+  const turn = turnOf(s, player);
+  parts.push(`m${turn.movesRemaining}`, JSON.stringify(turn.freightWorked), `h${(s.decks.hands.get(player) ?? []).join(',')}`);
+  return parts.join(';');
+}
+
+/**
+ * §9.3 — an outbound industry loads EMPTY cars of its commodity, an inbound one unloads LOADED ones —
+ * but never a load that was made in this same district (v0.4.9e, `LOADED_IN_THIS_DISTRICT`).
+ */
+function works(f: Facility, c: RollingStock, seat: number): boolean {
+  if (f.kind !== 'freight' || !facilityCarTypes(f).includes(c.type)) return false;
+  return (!c.loaded && f.allows.outbound) || (c.loaded && f.allows.inbound && c.origin !== seat);
+}
+
+/**
+ * The car an industry has finished with. Only decidable at a one-way industry: at one that both
+ * ships and receives, a loaded car may be a delivery still waiting to be unloaded.
+ */
+function finished(f: Facility, c: RollingStock): boolean {
+  if (f.kind !== 'freight' || !facilityCarTypes(f).includes(c.type)) return false;
+  if (f.allows.outbound && !f.allows.inbound) return c.loaded;
+  if (f.allows.inbound && !f.allows.outbound) return !c.loaded;
+  return false;
+}
+
+const same = (a: GridCoord, b: GridCoord): boolean => a.row === b.row && a.col === b.col;
+
+/** How good this district's position is for the rest of the game, in rough Revenue points. */
+export function evaluateSwitching(s: GameState, player: PlayerIndex): number {
+  const W = SWITCH_WEIGHTS;
+  const area = areaOf(s, player);
+  const seat = seatOf(s, player);
+  const officeKey = coordKey(area.officeCoord);
+  const passengerOffice = area.grid.get(officeKey)?.facility?.kind === 'passenger';
+  let v = 0;
+
+  const withRoom: Facility[] = [];
+  for (const card of area.grid.values()) {
+    const f = card.facility;
+    if (!f || f.kind !== 'freight') continue;
+    if (f.industryTrack.cars.length < MAX_CONSIST) withRoom.push(f);
+    const cap = Math.max(1, f.capacity.outbound + f.capacity.inbound);
+    let working = 0;
+    for (const c of f.industryTrack.cars) {
+      if (works(f, c, seat)) v += ++working <= cap ? W.spot : W.spotBeyondCapacity;
+      else if (finished(f, c)) v += W.finishedLeft;
+      else v += W.junkOnIndustry;
+    }
+  }
+  const wanted = (c: RollingStock): boolean => withRoom.some((f) => works(f, c, seat));
+
+  for (const [key, card] of area.grid) {
+    if (card.facility?.kind === 'freight') continue;
+    const atOffice = key === officeKey;
+    for (const c of card.standing) {
+      if (c.type === 'coach') v += atOffice && passengerOffice ? W.coachWithTrain : W.coachStranded;
+      else if (atOffice) v += W.fouling;
+      else if (wanted(c)) v += W.stagedWanted;
+    }
+  }
+
+  for (const t of s.trays.values()) {
+    if (t.position.at !== 'grid' || t.position.seat !== seat) continue;
+    for (const c of t.consist) {
+      if (c.type === 'coach') v += passengerOffice ? W.coachWithTrain : 0;
+      else if (wanted(c)) v += W.carriedWanted;
+    }
+    if (t.trainNumber === null) continue;
+    if (!same(t.position.coord, area.officeCoord)) {
+      v += W.trainAway;
+      if (isExpedited(t)) v += W.expeditedAway;
+    }
+    if (badlyMadeUp(t)) v += W.notMadeUp;
+  }
+  return v;
+}
+
+/**
+ * For ORDERING the beam only, never for choosing the plan: a Move toward an industry changes nothing
+ * the score can see until the car is set out, so without this the beam would drop the approach in
+ * favour of positions that merely look tidy.
+ */
+function approach(s: GameState, player: PlayerIndex): number {
+  const area = areaOf(s, player);
+  const seat = seatOf(s, player);
+  const targets: { at: GridCoord; f: Facility }[] = [];
+  for (const [key, card] of area.grid) {
+    const f = card.facility;
+    if (!f || f.kind !== 'freight' || f.industryTrack.cars.length >= MAX_CONSIST) continue;
+    const [row, col] = key.split(',').map(Number);
+    targets.push({ at: { row: row!, col: col! }, f });
+  }
+  let bonus = 0;
+  for (const t of s.trays.values()) {
+    if (t.position.at !== 'grid' || t.position.seat !== seat) continue;
+    const here = t.position.coord;
+    for (const c of t.consist) {
+      let nearest = Infinity;
+      for (const { at, f } of targets) {
+        if (works(f, c, seat)) nearest = Math.min(nearest, Math.abs(at.row - here.row) + Math.abs(at.col - here.col));
+      }
+      if (nearest !== Infinity) bonus += 0.1 / (1 + nearest);
+    }
+  }
+  return bonus;
+}
+
+const SEARCHED = new Set<Intent['type']>(['switch.move', 'switch.dropCars', 'switch.sortConsist']);
+
+type Node = {
+  s: GameState;
+  steps: Intent[];
+  keys: string[];
+  moves: number;
+  setOuts: number;
+  cards: number;
+  score: number;
+  rank: number;
+};
+
+/** The best way found to spend what is left of this switching turn. Never mutates `s`. */
+export function planSwitchingTurn(
+  s: GameState,
+  player: PlayerIndex,
+  opts: PlanOptions = DEFAULT_PLAN,
+): SwitchPlan {
+  const W = SWITCH_WEIGHTS;
+  const scoreOf = (st: GameState, moves: number, setOuts: number, cards: number): number =>
+    evaluateSwitching(st, player) + moves * W.perMove + setOuts * W.perSetOut + cards * W.cardSpent;
+  const searched = (type: Intent['type']): boolean =>
+    SEARCHED.has(type) || (opts.flyingSwitch === true && type === 'maneuver.flyingSwitch');
+
+  const rootKey = switchFingerprint(s, player);
+  const rootScore = scoreOf(s, 0, 0, 0);
+  const root: Node = { s, steps: [], keys: [rootKey], moves: 0, setOuts: 0, cards: 0, score: rootScore, rank: rootScore };
+  let best = root;
+  const seen = new Set([rootKey]);
+  let frontier: Node[] = [root];
+  let expanded = 0;
+  let complete = true;
+
+  search: while (frontier.length > 0) {
+    const next: Node[] = [];
+    for (const node of frontier) {
+      const movesLeft = turnOf(node.s, player).movesRemaining;
+      // Every candidate is decided against THIS position, inside one route cache, and only then applied
+      // to its own copy: deciding on the copy would re-walk routes the listing had just walked.
+      const decided = withRouteCache(node.s, () =>
+        legalSwitchingActions(node.s, player)
+          .filter((i) => searched(i.type) && (i.type === 'switch.dropCars' || movesLeft >= 1))
+          .map((i) => ({ i, r: prepareIntent(node.s, player, i) })),
+      );
+      for (const { i, r } of decided) {
+        if (expanded >= opts.budget) {
+          complete = false;
+          break search;
+        }
+        expanded++;
+        if (!r.ok) continue;
+        const f = forkForSwitching(node.s, player);
+        commitEvents(f, r.events);
+        const key = switchFingerprint(f, player);
+        if (seen.has(key)) continue;
+        seen.add(key);
+        const setOut = i.type === 'switch.dropCars';
+        const moves = node.moves + (setOut ? 0 : 1);
+        const setOuts = node.setOuts + (setOut ? 1 : 0);
+        const cards = node.cards + (i.type === 'maneuver.flyingSwitch' ? 1 : 0);
+        const score = scoreOf(f, moves, setOuts, cards);
+        const child: Node = {
+          s: f,
+          steps: [...node.steps, i],
+          keys: [...node.keys, key],
+          moves,
+          setOuts,
+          cards,
+          score,
+          rank: score + approach(f, player),
+        };
+        if (score > best.score + 1e-9) best = child;
+        next.push(child);
+      }
+    }
+    // A stable sort, so equal ranks keep `legalActions` order and the bot stays deterministic.
+    frontier = next.length > opts.beam ? next.sort((a, b) => b.rank - a.rank).slice(0, opts.beam) : next;
+  }
+
+  return { steps: best.steps, keys: best.keys, rootScore, score: best.score, expanded, complete };
+}

src/sim/turnchart.ts

@@ -106,7 +106,25 @@ export function turnChartHtml(f: TurnChartFrame, actorName: string | null, super
    * and answers who; `awaiting` says what, because "waiting on Bob" with no more than that is a
    * game that looks stuck to everyone except Bob.
    */
-  const who = actorName ?? 'nobody — the Division is running itself';
+  /**
+   * AN AUTOMATIC PHASE WAITS ON NOBODY, so it says what it is DOING instead of apologising.
+   *
+   * "nobody — the Division is running itself" reached the answer by negation, and left a player
+   * reading a line whose subject was an absence (Jesse, playtest 2026-09-16: it should say "waiting
+   * on <player>", or describe what the engine is doing — "the Division is moving trains during the
+   * mainline phase"). The phase's NAME is already printed on the line directly above this one, so
+   * these describe the work rather than repeating the label.
+   */
+  const DOING: Record<string, string> = {
+    mainline: 'the Division is moving trains',
+    newTrain: "the Division is building this Stage's trains",
+    loadUnload: 'the Division is working cargo',
+    shiftChange: 'the Division is changing shifts',
+  };
+  // Local Operations always has an actor, so its entry is the fallback rather than a case.
+  const who = actorName ?? DOING[f.phaseKey] ?? 'the Division is running itself';
+  // Only a person is WAITED ON. The Division is not waiting; it is working.
+  const waiting = actorName === null ? '' : 'waiting on ';
   const asked = f.awaiting
     ? ` <span class="tc-asks">${esc(f.awaiting.asks)} · ${esc(f.awaiting.train)}</span>`
     : '';
@@ -120,7 +138,7 @@ export function turnChartHtml(f: TurnChartFrame, actorName: string | null, super
     `<div class="tc-when"><b>Day ${f.day}</b><span>Stage ${f.stage} of 12</span>` +
     `<span class="dim">${esc(f.clock)}</span></div>` +
     `<div class="tc-now">phase <b>${esc(f.phase)}</b></div>` +
-    `<div class="tc-who">waiting on <b>${esc(who)}</b>${asked}</div>` +
+    `<div class="tc-who">${waiting}<b>${esc(who)}</b>${asked}</div>` +
     // THE FEDORA RIDES AT THE END OF THE PHASE ROW (`TODO.md` #29, Jesse). It sat on its own line
     // between the phases and everything above them, which put a thing that changes every third
     // Stage in the middle of the things that change every Stage. The row it belongs beside is the

src/sim/view.ts

@@ -16,6 +16,7 @@ import {
   areaOf,
   destinationsFor,
   facilityCarType,
+  isBeingMadeUp,
   laborersLeft,
   movesFor,
   ownCutFor,
@@ -276,6 +277,15 @@ export type TrainChip = {
    * it belongs in the tooltip, where there is room to say which it is.
    */
   stagesLeft?: number;
+  /**
+   * BEING MADE UP RIGHT NOW — §7's round, one car at a time, at a Division Point.
+   *
+   * The make-up panel names the train and the yard chips load it, and both are in the right-hand
+   * column; the train itself is drawn on the Division strip at the top left, looking exactly like
+   * every other chip on the map. So the two halves of the same activity never pointed at each other
+   * (Jesse, playtest 2026-09-16). Absent rather than false everywhere else, like `region` above.
+   */
+  beingMadeUp?: true;
 };
 /**
  * One card of a player's Running Track, as the Division sees it.
@@ -720,6 +730,26 @@ function suppressedGrants(modifiers: string[], f: Facility): string[] {
   for (const key of modifiers) {
     const m = MODIFIER_PROFILES.find((p) => p.kind === key);
     if (!m) continue;
+    /**
+     * A WHISTLE POST TAKES NOTHING AT ALL, AND SAYING "IT ONLY RECEIVES" WOULD BE A LIE.
+     *
+     * Jesse, playtest 2026-09-16: a Restaurant appeared to do nothing. It does nothing — a Whistle
+     * Post is not a Passenger Facility, so it allows neither direction and has 0 capacity each way;
+     * the engine's `usableGrant` discards the capacity while the porter is granted regardless, which
+     * leaves a porter with nothing to carry. Playing it there stays LEGAL on Jesse's call, so the
+     * card is not wasted — it starts working the moment the Office is upgraded — but the panel has
+     * to say so, or the player is left believing the card is broken.
+     *
+     * Only a Whistle Post can reach this: every freight flow allows at least one direction, and
+     * every Office above the first allows both.
+     */
+    if (f.kind === 'passenger' && !f.allows.outbound && !f.allows.inbound) {
+      out.push(
+        `${m.name}: DORMANT — a Whistle Post works no passengers at all, so nothing this card ` +
+          `grants is in use yet. It all starts working when the Office is upgraded to a Depot.`,
+      );
+      continue;
+    }
     if (m.addOut > 0 && !f.allows.outbound) {
       out.push(`${m.name}: +${m.addOut} outbound has no effect here — this facility only receives`);
     }
@@ -1791,23 +1821,6 @@ export function snapshot(
 
 /** A card id turned into something a person can read. */
 export function cardName(s: GameState, id: string): string {
-  /**
-   * A SCHEDULED TRAIN IS IN THE SALVAGE YARD UNDER A SYNTHETIC ID, and without this the pile that is
-   * supposed to be face up reads "a card".
-   *
-   * `trainScheduled` pushes `train-<number>` rather than the id of the card that was played
-   * (`apply.ts`), so there is nothing in `s.cards` to look up — and since a train is scheduled
-   * several times a Day, that synthetic id is on top of the Salvage Yard most of the time. Reported
-   * by Jesse 2026-09-10 as "why is salvage deck not face up. I should see the card played onto
-   * salvage": it was face up all along and simply had nothing to say.
-   *
-   * Resolved here rather than in the engine because this is a NAME, which is this file's job. Whether
-   * the engine should be pushing a real card id instead is a separate question with a separate
-   * consequence — the Salvage Yard is swept back into the draw deck when it runs out — and is filed
-   * rather than answered in passing.
-   */
-  const scheduled = /^train-(\d+)$/.exec(id);
-  if (scheduled) return `Train ${scheduled[1]}`;
   const k = s.cards.get(id)?.kind;
   if (!k) return 'a card';
   switch (k.kind) {
@@ -2384,6 +2397,9 @@ function trainChip(s: GameState, id: string): TrainChip {
     engineAt: at,
     facing: railFacingOf(t),
     what: t.trainNumber === null ? 'A local crew — no timetable, no card, no special rules.' : trainRules(t),
+    // Conditional spread, not `beingMadeUp: isBeingMadeUp(t)`: the field is optional-and-true, and
+    // `exactOptionalPropertyTypes` refuses an explicit `false` for it.
+    ...(isBeingMadeUp(t) ? { beingMadeUp: true as const } : {}),
   };
 }
 

src/web/game.ts

@@ -663,6 +663,14 @@ export type Menu = {
   makeUp: {
     trayId: string;
     title: string;
+    /**
+     * What the card still wants, after what is already coupled up — "1 boxcar/hopper + 1 caboose".
+     *
+     * The title says what the card CALLS FOR and never changes as cars go on, so a player had to
+     * diff it against the consist drawn on the Division map, in the other column. Null when the
+     * train is complete and only the send-it-out button is left.
+     */
+    needs: string | null;
     cars: MakeUpAction[];
     pass: number | null;
     /**
@@ -803,6 +811,7 @@ export function actionMenu(game: Game, seat: PlayerIndex = 0): Menu {
       ? {
           trayId: filling,
           title: consistTitle(game, filling) ?? 'Making up the train',
+          needs: consistNeeds(game, filling),
           cars: makeUpCars,
           pass,
           advice: makeUpAdvice(game, filling, makeUpCars),
@@ -955,6 +964,40 @@ function consistTitle(game: Game, trayId: string): string | null {
   return trainCardTitle(tray.trainNumber ?? 0, tray.trainIsExtra);
 }
 
+/**
+ * WHAT THE TRAIN STILL WANTS — the card's demand minus what is already on it.
+ *
+ * The heading says "its card calls for 3 boxcar/hopper + 1 caboose" and goes on saying it whether
+ * you have added none or three; the cars themselves are drawn on the Division map, in the other
+ * column. So the one question a player actually has while clicking — what is left? — was the one
+ * thing on screen that had to be worked out by eye, across two panels (Jesse, playtest 2026-09-16).
+ *
+ * BY CATEGORY, exactly as `acceptsCar` counts them, so this cannot promise a car the engine would
+ * then refuse. Null when nothing is outstanding.
+ */
+function consistNeeds(game: Game, trayId: string): string | null {
+  const tray = game.state.trays.get(trayId);
+  if (!tray) return null;
+  const p = trainProfile(tray.trainNumber ?? 0, tray.trainIsExtra);
+  if (!p) return null;
+
+  const cat = (t: string): 'coach' | 'caboose' | 'freight' =>
+    t === 'coach' ? 'coach' : t === 'caboose' ? 'caboose' : 'freight';
+  const have = (k: 'coach' | 'caboose' | 'freight'): number =>
+    tray.consist.filter((c) => cat(c.type) === k).length;
+
+  const parts: string[] = [];
+  const freight = p.consist.freight - have('freight');
+  const coach = p.consist.coach - have('coach');
+  const caboose = p.consist.caboose - have('caboose');
+  if (freight > 0) {
+    parts.push(`${freight} ${p.consist.freightTypes?.join('/') ?? 'freight'}${p.consist.emptiesOnly ? ' (empties only)' : ''}`);
+  }
+  if (coach > 0) parts.push(`${coach} coach${coach > 1 ? 'es' : ''}`);
+  if (caboose > 0) parts.push(`${caboose} caboose`);
+  return parts.length > 0 ? parts.join(' + ') : null;
+}
+
 /**
  * "Making up Extra X22 “Pee-Dee”: its card calls for 1 caboose — Per-diem train…"
  *
@@ -1234,6 +1277,13 @@ function record(game: Game, events: GameEvent[], actor: PlayerIndex | null = nul
     const n = narrate(e, {
       cardName: (id) => cardName(game.state, id),
       trainName: (id) => trainName(game.state, id),
+      // Whose district a train reached is not the actor — the Mainline Phase has none — so the
+      // narration resolves the name itself rather than being prefixed with one by the code below.
+      // NO NUMBER IN THE FALLBACK. This is a PLAYER index, and a player is not a seat — seats rotate
+      // under Employee Rotation, which is why `seatOf` exists — so "Seat 3" here would be a wrong
+      // number dressed as a right one, and `session.test.ts` rightly refuses any raw index shown to
+      // a person. Every caller passes real names; an unnamed player is anonymous rather than mislabelled.
+      playerName: (p) => game.state.players[p]?.name ?? 'another player',
     });
     /**
      * A BLIND DRAW IS PUBLIC; WHICH CARD CAME UP IS NOT (Gitea#20 step 1).
@@ -1255,9 +1305,22 @@ function record(game: Game, events: GameEvent[], actor: PlayerIndex | null = nul
     const said = blindDraw ? 'Drew a card from the Home Office deck' : n.text;
     // "Chose to DRAW a card" does not say WHO, which is unreadable the moment there is more than
     // one seat. Only events the player caused are attributed; the Division running itself is not.
+    /**
+     * A RULING IS MADE AS SUPERINTENDENT, NOT AS YOURSELF (playtest, 2026-09-15: "maybe it could say
+     * 'Superintendent Player Tom', so it's clear they got the move because they're Superintendent").
+     * These three are the only moves a player makes out of turn, by holding the office: §8.1's
+     * clearance, §11's Yard Office offer and §Q's Red Flag prompt. `clearanceGiven` carries no
+     * player at all — the office made it, whoever holds it — so the actor is what names it.
+     */
+    const RULINGS = ['clearanceGiven', 'yardOfficeRuled', 'redFlagRuled'];
+    const ruling = RULINGS.includes(e.type) && who !== null;
     const mine = who !== null && 'player' in e;
-    const text = mine ? `Player ${who} ${uncapitalise(said)}` : said;
-    game.log.push({ text, tone: mine ? 'act' : n.tone });
+    const text = ruling
+      ? `Superintendent Player ${who} ${uncapitalise(said)}`
+      : mine
+        ? `Player ${who} ${uncapitalise(said)}`
+        : said;
+    game.log.push({ text, tone: mine || ruling ? 'act' : n.tone });
 
   }
   game.cues.push(...cuesFor(events));

src/web/lobby.ts

@@ -107,7 +107,8 @@ function explain(code: unknown, fallback: string): string {
   return messages[key] ?? (key !== '' ? key : fallback);
 }
 
-async function postJson(path: string, body: unknown): Promise<{ status: number; body: Record<string, unknown> }> {
+/** Exported for `main.ts`'s seat-recovery path (Gitea#33), so there is one JSON POST on this page. */
+export async function postJson(path: string, body: unknown): Promise<{ status: number; body: Record<string, unknown> }> {
   const res = await fetch(path, {
     method: 'POST',
     headers: { 'Content-Type': 'application/json' },

src/web/main.ts

@@ -25,9 +25,9 @@ import type { LocalSession, Session } from './session.ts';
 import { createLocalSession, createRemoteSession } from './session.ts';
 import type { PlayerIndex } from '../engine/state.ts';
 import type { PublicDistrict } from '../sim/view.ts';
-import { createStepQueue } from './step-queue.ts';
+import { actorOnScreen, createStepQueue } from './step-queue.ts';
 import { PACE_LEVELS } from '../sim/pacing.ts';
-import { notice, prefillCode, runLobby } from './lobby.ts';
+import { notice, postJson, prefillCode, runLobby } from './lobby.ts';
 import type { LobbyReady } from './lobby.ts';
 import {
   closestPreset,
@@ -231,7 +231,9 @@ function renderWatching(f?: Frame): void {
     return;
   }
   row.hidden = false;
-  $('watching-behind').textContent = behind === 0 ? 'catching up' : `${behind} behind`;
+  // A held queue stops counting down, so the counter has to say why rather than look stuck.
+  $('watching-behind').textContent =
+    (behind === 0 ? 'catching up' : `${behind} behind`) + (stepQueue.paused() ? ' · paused' : '');
   /**
    * THE CAPTION IS #15, and this is where that item lands rather than as a line of its own.
    *
@@ -273,6 +275,56 @@ function renderWatching(f?: Frame): void {
   $('watching-skip').onclick = () => {
     if (stepQueue.skip()) render();
   };
+  /**
+   * PAUSE IS SKIP'S OPPOSITE, and shares its row for that reason.
+   *
+   * The label says what pressing it DOES, so it flips to Resume while held — the same rule the
+   * district's three-mode control settled on, for the same reason: a label that reports state reads
+   * as a status line and gets skipped over.
+   *
+   * `performance.now()` because that is the clock `requestAnimationFrame` hands `advance()`; mixing
+   * in `Date.now()` would shift the deadline by the page's whole lifetime. Guarded because the
+   * static build is loaded head-first against a DOM stub with no `performance`.
+   */
+  const pauseBtn = $('watching-pause');
+  pauseBtn.textContent = stepQueue.paused() ? 'Resume' : 'Pause';
+  pauseBtn.onclick = () => {
+    const now = typeof performance !== 'undefined' ? performance.now() : Date.now();
+    if (stepQueue.paused()) {
+      stepQueue.resume(now);
+      // The loop exits whenever the queue stops being busy; restart it rather than assume it survived.
+      startAnimationLoop();
+    } else {
+      stepQueue.pause(now);
+    }
+    render();
+  };
+}
+
+/**
+ * THE TABLE AS IT IS ON SCREEN — the Day, the Stage, the clock, the phase and the Fedora.
+ *
+ * "WHEN PLAYER JESSE IS 5 BEHIND, IT SHOULD ALWAYS LOOK LIKE HE'S 5 BEHIND" (playtest, 2026-09-16).
+ * The board, the district panel and the caption row have followed the queue since v0.8.0; the turn
+ * chart never did. So a player watching three bots play out a Stage saw their cards moving under a
+ * chart that had already ticked over to the next phase — the one part of the screen quietly
+ * insisting the game was somewhere else. Being behind is fine and is stated plainly by the counter;
+ * being behind on some of the screen and level on the rest is what makes it unreadable.
+ *
+ * Only while the queue is actually behind. At rest this IS the live frame, so nothing downstream
+ * needs to know which of the two it was handed.
+ */
+function shownTable(f: Frame): Pick<Frame, 'day' | 'stage' | 'clock' | 'phase' | 'phaseKey' | 'superintendent'> {
+  const pub = stepQueue.current();
+  if (!pub || !stepQueue.busy()) return f;
+  return {
+    day: pub.day,
+    stage: pub.stage,
+    clock: pub.clock,
+    phase: pub.phase,
+    phaseKey: pub.phaseKey,
+    superintendent: pub.superintendent,
+  };
 }
 
 function watchedDistrict(f: Frame): PublicDistrict | null {
@@ -285,6 +337,10 @@ function watchedDistrict(f: Frame): PublicDistrict | null {
    * A PHASE STEP NAMES NOBODY — the Mainline advances itself — so it falls through to the actor,
    * which keeps the board where it was instead of snapping home mid-sequence.
    */
+  // A deliberate look wins over whoever happens to be acting, for this one render (see `peekPlayer`).
+  if (peekPlayer !== null && peekPlayer !== f.viewer) {
+    return pub.districts.find((d) => d.player === peekPlayer) ?? null;
+  }
   let player: PlayerIndex | null = f.actor;
   if (stepQueue.busy()) {
     const acting = stepQueue.showing()?.player;
@@ -430,6 +486,22 @@ let lastDay: number | null = null;
  * ones. Cleared whenever the named crew stops being one of the choices.
  */
 let selectedCrew: string | null = null;
+
+/**
+ * A ONE-RENDER LOOK AT SOMEBODY ELSE'S OFFICE AREA — Jesse, playtest 2026-09-16.
+ *
+ * Deliberately NOT a mode. It survives exactly the render its own click causes and is cleared at the
+ * end of `renderDistrict`, so the panel is back to following whoever is acting the next time
+ * anything redraws. That is the whole design, in his words: *"if you want to study someone else's
+ * office area, you should do it while it's your turn to move, or put the backlog on pause, then look
+ * at their area, and when you're done looking, resume."*
+ *
+ * A sticky pin would have to answer what happens when the game moves on beneath it — and the honest
+ * answers are all bad: silently snap home, or leave a player staring at a stale board with the game
+ * waiting on them. Pause already means "hold everything", so it is the right lever for a long look,
+ * and this stays a glance.
+ */
+let peekPlayer: PlayerIndex | null = null;
 const FOCUS_PHASES = new Set(['localOps', 'loadUnload']);
 
 /** The crew whose squares the board is drawing: the chosen one, or the only one there is. */
@@ -493,12 +565,20 @@ function piecePreview(links: string[], label: string): string {
  * are in the Day the same way and with the same violet highlight. It used to live here alone.
  */
 function renderTurnChart(f: Frame): void {
-  const actorName = f.actor === null ? null : (f.players[f.actor]?.name ?? null);
+  // The move on screen, not the live one, while the board is still catching up (Gitea#25).
+  const { actor, replaying } = actorOnScreen(stepQueue, f.actor);
+  const actorName = actor === null ? null : (f.players[actor]?.name ?? null);
+  // The Day, Stage and phase of the step being shown, so the whole screen reports one moment.
+  const table = shownTable(f);
   // Named only at a table with more than one seat: in solitaire the Fedora is always yours, and a
   // chip that can never change is a chip to read past.
   const superName =
-    f.players.length > 1 ? (f.players.find((p) => p.index === f.superintendent)?.name ?? null) : null;
-  $('turnchart').innerHTML = turnChartHtml(f, actorName, superName);
+    f.players.length > 1 ? (f.players.find((p) => p.index === table.superintendent)?.name ?? null) : null;
+  $('turnchart').innerHTML = turnChartHtml(
+    replaying ? { ...f, ...table, awaiting: null } : f,
+    actorName,
+    superName,
+  );
 }
 
 /**
@@ -1000,6 +1080,7 @@ function beginRemote(ready: LobbyReady, rejoining = false): void {
   // banner (`#presence`), and it holds a beat so the game visibly begins.
   openHandoff();
   session = createRemoteSession(ready.token, ready.seat, abandonRemote);
+  remoteToken = ready.token;
   rejoiningRemote = rejoining;
   applyCapabilities();
   // A LocalSession has data the instant it is constructed; a RemoteSession does not — its first
@@ -1045,9 +1126,66 @@ function abandonRemote(): void {
  * into a remembered multiplayer game, straight into solitaire (the zero-friction default, D11 — the
  * common case and the only one a bare page load has ever needed a decision for), or the lobby.
  */
+/**
+ * A SEAT RECOVERY LINK — Gitea#33.
+ *
+ * The token is the only identity this game has, and it lives in one browser's `localStorage`. Lose
+ * that and the seat is unreachable: nothing else on the server will accept a claim to it. This is the
+ * supported way back — an administrator mints a short-lived, single-use code (`server/claims.ts`) and
+ * the player opens a link carrying it.
+ *
+ * THE LINK CARRIES A CODE, NEVER THE TOKEN. `lobby-and-sessions.md` §1 says to keep the token out of
+ * URLs so it is not shoulder-surfed or pasted into a chat — and a recovery link is precisely the sort
+ * of thing that ends up in a chat. So the code is traded for the token here, over the connection the
+ * page was going to open anyway, and is dead the moment it is spent.
+ *
+ * THE CODE IS STRIPPED FROM THE URL EITHER WAY, so a reload does not re-spend a code that is already
+ * gone and the address bar stops carrying a credential-shaped string. `replaceState` rather than
+ * assigning `location.search`, which everywhere else on this page means "navigate" — it reloads, and
+ * reloading is exactly what must not happen to the session we have just been handed. Guarded like
+ * `requestAnimationFrame` and `performance` are, because the static build is imported head-first by
+ * `test/web.test.ts` against a DOM stub that provides neither.
+ */
+async function claimSeat(code: string): Promise<void> {
+  showScreen('lobby');
+  const { status, body } = await postJson('/api/claim', { code });
+  if (typeof history !== 'undefined' && typeof history.replaceState === 'function') {
+    history.replaceState(null, '', location.pathname);
+  }
+  if (status !== 200) {
+    runLobby(lobbyHandlers);
+    notice(
+      'That restore link has already been used, or it has expired. Ask whoever runs the server for a ' +
+        'fresh one — each link works once.',
+    );
+    return;
+  }
+  // `beginRemote` writes the seat into this browser's storage itself, which is the whole point of
+  // the exercise: the next ordinary reload finds it and goes straight back into the game.
+  beginRemote(
+    {
+      token: body['token'] as string,
+      gameId: body['gameId'] as string,
+      gameCode: (body['gameCode'] as string | undefined) ?? '',
+      seat: body['player'] as PlayerIndex,
+    },
+    true,
+  );
+}
+
 function start(): void {
   const params = new URLSearchParams(location.search);
 
+  /**
+   * A RECOVERY LINK OUTRANKS EVERYTHING, including a game this browser already remembers: someone
+   * arriving on one is being handed a seat deliberately, and that is never the load to second-guess.
+   */
+  const claimCode = params.get('claim');
+  if (claimCode !== null && claimCode !== '') {
+    void claimSeat(claimCode);
+    return;
+  }
+
   /**
    * ASKING FOR THE LOBBY BEATS RESUMING A GAME.
    *
@@ -1167,6 +1305,14 @@ function start(): void {
  * the lobby's job. Hidden rather than disabled: a greyed-out Undo in a multiplayer game invites the
  * question "why not?" every turn, and the honest answer is that the control does not belong there.
  */
+/**
+ * The session token of a server-backed game, or null in solitaire (playtest, 2026-09-15: "most of the
+ * time, I want to go ahead and just save it as a JSON file"). It is the seat's proof of identity to
+ * `/api/save`, exactly as it is to `/api/stream` — a save is the seed and the moves, every one of which
+ * is already on this player's screen.
+ */
+let remoteToken: string | null = null;
+
 function applyCapabilities(): void {
   const c = session.capabilities;
   const hide = (id: string, on: boolean): void => {
@@ -1174,7 +1320,7 @@ function applyCapabilities(): void {
     if (el) el.hidden = !on;
   };
   hide('undo', c.undo);
-  hide('savefile', c.saveLocal);
+  hide('savefile', c.saveLocal || remoteToken !== null);
   hide('newgame', c.newGame);
   // Creating or joining ANOTHER multiplayer game from inside a running one is not a thing this
   // page offers — same reasoning as `newgame`, and the same capability answers both.
@@ -1347,8 +1493,10 @@ function render(): void {
   // -- division
   $('division').innerHTML = divisionSvg(f.division, {
     players: f.players,
-    actor: f.actor,
+    actor: actorOnScreen(stepQueue, f.actor).actor,
     viewer: f.viewer,
+    // A Realignment changes the Division under everyone; flashed only while the step that did it is up.
+    flash: stepQueue.busy() ? stepQueue.flashing() : [],
   });
   renderSeatingChain(f);
   applyZoom($('division'));
@@ -1633,7 +1781,16 @@ function render(): void {
 
   // -- log
   const log = $('log');
-  const allLines = session.lines();
+  /**
+   * THE LOG IS HELD BACK WITH THE BOARD (playtest, 2026-09-15).
+   *
+   * A push carries its narration and its display steps together, so every line of a bot's turn was in
+   * this panel before the board had drawn a single move of it — the history ran ahead of the "N behind"
+   * counter it is meant to match. Those lines are the TAIL of the log, so exactly the ones belonging to
+   * steps still queued are withheld, and each appears as its step goes up.
+   */
+  const heldBack = stepQueue.pendingLines();
+  const allLines = heldBack > 0 ? session.lines().slice(0, -heldBack) : session.lines();
   const shownLines = allLines.slice(-60);
   /**
    * WHERE THE GAME BEGAN. In a multiplayer game the bots move the instant the host presses Start, so
@@ -1771,17 +1928,34 @@ function renderYards(f: Frame): void {
 }
 
 function renderDistrict(f: Frame): void {
-  const open = districtMode === 'auto' ? FOCUS_PHASES.has(f.phaseKey) : districtMode === 'open';
+  // The phase ON SCREEN, so the panel opens for the Local Operations being WATCHED rather than for
+  // one the game has already moved past — same rule as the turn chart, see `shownTable`.
+  const open =
+    districtMode === 'auto' ? FOCUS_PHASES.has(shownTable(f).phaseKey) : districtMode === 'open';
   const sec = $('district');
   if (open) sec.classList.remove('folded');
   else sec.classList.add('folded');
 
-  const cars = f.cells.reduce((n, c) => n + c.cars.length, 0);
+  /**
+   * THE SUMMARY COUNTS THE BOARD ON SCREEN, WHICH IS NOT ALWAYS YOUR OWN.
+   *
+   * The panel has drawn somebody else's district since v0.8.0 — `watchedDistrict` follows whoever is
+   * acting — and the heading beside this line says whose it is. The counts were read from `f`, the
+   * viewer's own Frame, every time: so while a bot's turn played out, the header read "Bot 2's Office
+   * Area" over a board of Bot 2's cards, with a summary counting YOUR cards, facilities and trains
+   * (Jesse, playtest 2026-09-16 — "the hidden office summary line describes my district, not the one
+   * being shown"). One source for the drawing and the counting, so the two cannot disagree again.
+   */
+  const watched = watchedDistrict(f);
+  const cells = watched?.cells ?? f.cells;
+  const facilityCount = watched ? watched.facilities.length : f.facilities.length;
+
+  const cars = cells.reduce((n, c) => n + c.cars.length, 0);
   // Trains, not cards-with-a-crew: the Office is the one card that may hold more than one, and a
   // card-count silently read "1 crew on the board" with two trains standing at a busy Station.
-  const crew = f.cells.reduce((n, c) => n + c.trains.length, 0);
+  const crew = cells.reduce((n, c) => n + c.trains.length, 0);
   $('districtsummary').textContent =
-    `${f.cells.length} cards · ${f.facilities.length} facilities · ${cars} cars standing` +
+    `${cells.length} cards · ${facilityCount} facilities · ${cars} cars standing` +
     (crew > 0 ? ` · ${crew} crew on the board` : '');
 
   /**
@@ -1813,6 +1987,35 @@ function renderDistrict(f: Frame): void {
       render();
     };
   }
+
+  /**
+   * ONE BUTTON PER OPPONENT, and nothing at a table of one.
+   *
+   * Built with `createElement` and `textContent` rather than interpolated into `innerHTML`, because
+   * a player's NAME is whatever they typed in the lobby — the one string on this page that comes
+   * from another person, and so the one that must never be pasted into markup.
+   */
+  const peek = $('districtpeek');
+  const others = f.players.filter((p) => p.index !== f.viewer);
+  peek.innerHTML = '';
+  peek.hidden = others.length === 0;
+  for (const p of others) {
+    const b = document.createElement('button');
+    b.type = 'button';
+    b.className = 'ghost';
+    b.textContent = p.name;
+    b.title =
+      `Look at ${p.name}'s Office Area. It is read-only, and it reverts as soon as the board next ` +
+      `redraws — press Pause first if you want to study it.`;
+    b.onclick = () => {
+      peekPlayer = p.index;
+      render();
+    };
+    peek.appendChild(b);
+  }
+
+  // SPENT. The look lasted the render it asked for; the next one follows the game again.
+  peekPlayer = null;
 }
 
 /**
@@ -2181,9 +2384,32 @@ function renderActions(
     const addable = menu.makeUp.cars.length;
     html +=
       `<div class="grp"><h3>${esc(menu.makeUp.title)}</h3>` +
+      /**
+       * WHAT IS STILL WANTED, which is not what the heading says.
+       *
+       * The heading names what the card CALLS FOR and goes on saying it unchanged as cars go on, so
+       * the one question a player has while clicking — what is left? — was the only thing on screen
+       * that had to be worked out by eye, against a consist drawn in the other column (Jesse,
+       * playtest 2026-09-16). `consistNeeds` counts by the same categories `acceptsCar` does, so it
+       * can never ask for a car the engine would then refuse.
+       */
+      (menu.makeUp.needs !== null
+        ? `<div class="makeup-needs">Still needs <b>${esc(menu.makeUp.needs)}</b></div>`
+        : `<div class="makeup-needs done">Its card's consist is complete — nothing further may be added.</div>`) +
       `<div class="dim makeup-note">` +
       (addable > 0
-        ? `Click a car in the Division Yard below to add it — ${addable} kind${addable === 1 ? '' : 's'} it may take are highlighted there.`
+        ? `Click a car in the Division Yard below to add it — the ${addable} kind${addable === 1 ? '' : 's'} it may take ` +
+          `${addable === 1 ? 'is' : 'are'} highlighted in amber there. ` +
+          /**
+           * WHY YOUR TURN ENDS AFTER ONE CAR — §7's round, said where the clicking happens.
+           *
+           * It was written down only in the turn chart's New Train chip tooltip: hovered once, early
+           * on, and never again. "Click a car" then reads as "build this train", so a player adds one
+           * and the turn moves on with no explanation (Jesse, playtest 2026-09-16).
+           */
+          `<b>One car each:</b> you add a single car, then the round passes to the next player — ` +
+          `starting from the Superintendent and working left, coming round again until the train is ` +
+          `full or the Division Yard holds nothing it can take.`
         : menu.makeUp.pass !== null
           ? 'The Division Yard is bare, so there is nothing to add. Send the train out as it stands.'
           : 'Nothing in the Division Yard may join this train, and passing is not allowed while the yard holds cars.') +
@@ -2317,21 +2543,60 @@ function renderActions(
  * hundred bytes, so a finished game can be emailed or dropped on the site's replay directory —
  * where a rendered page would have been megabytes.
  */
-function downloadSave(): void {
-  // The button this fires from is hidden by `applyCapabilities()` for any session that cannot save
-  // (`#savefile`), but nothing stops this function being called directly, so the guard is repeated
-  // here rather than only trusted to the DOM.
-  if (!isLocal(session)) return;
-  const data = JSON.stringify(session.save(), null, 1);
+function writeFile(name: string, data: string): void {
   const blob = new Blob([data], { type: 'application/json' });
   const url = URL.createObjectURL(blob);
   const a = document.createElement('a');
   a.href = url;
-  a.download = `station-master-seed${session.seed()}-day${session.view().day}.json`;
+  a.download = name;
   a.click();
   URL.revokeObjectURL(url);
 }
 
+/**
+ * WHICH GAME, HOW FAR IN, AND WHEN — Jesse, playtest 2026-09-16.
+ *
+ * The name was `station-master-day1-stage5.json` for every server game at that point in every
+ * Stage, so two saves off the same table collided in the downloads folder and neither said which
+ * table it came from. The join code is the one thing a player already says out loud to identify a
+ * game, so it leads: `whistle-6945.day1.stage5.2026.09.16.json`.
+ *
+ * Lowercased because a filename is not a thing you shout, and dotted because that is the shape
+ * Jesse asked for. A solitaire game has no join code and falls back to its seed, which is the
+ * equivalent identity for a game nobody else is sitting at.
+ */
+function saveFileName(prefix: string, f: { day: number; stage: number }): string {
+  const d = new Date();
+  const pad = (n: number): string => String(n).padStart(2, '0');
+  const date = `${d.getFullYear()}.${pad(d.getMonth() + 1)}.${pad(d.getDate())}`;
+  return `${prefix}.day${f.day}.stage${f.stage}.${date}.json`;
+}
+
+async function downloadSave(): Promise<void> {
+  const f = session.view();
+  /**
+   * A SERVER-BACKED GAME HAS NO LOCAL SAVE TO HAND OVER, so it asks the server for its own — the seat's
+   * token is the gate (`/api/save`), the same one the stream and every intent already use. The StartOS
+   * Manage Game action cannot do this: an action result is text only, with no file member in the SDK.
+   */
+  if (!isLocal(session)) {
+    if (remoteToken === null) return;
+    try {
+      const res = await fetch(`/api/save?token=${encodeURIComponent(remoteToken)}`);
+      if (!res.ok) return;
+      const body = (await res.json()) as { save: unknown };
+      const code = gameCode === '' ? 'station-master' : gameCode.toLowerCase();
+      writeFile(saveFileName(code, f), JSON.stringify(body.save, null, 1));
+    } catch {
+      // Offline, or the game has been ended under us: the button simply does nothing, which is the
+      // same thing every other server call on this page does when the server is not there.
+    }
+    return;
+  }
+  const solo = gameCode === '' ? `station-master-seed${session.seed()}` : gameCode.toLowerCase();
+  writeFile(saveFileName(solo, f), JSON.stringify(session.save(), null, 1));
+}
+
 function save(): void {
   if (!isLocal(session)) return;
   try {
@@ -2367,7 +2632,7 @@ document.head.appendChild(pageStyle);
 installTooltips();
 
 const saveBtn = document.getElementById('savefile');
-if (saveBtn) saveBtn.onclick = downloadSave;
+if (saveBtn) saveBtn.onclick = () => void downloadSave();
 
 /**
  * Forget the saved game and deal a fresh one.

src/web/panels.ts

@@ -777,9 +777,18 @@ h3{font-size:11px;text-transform:uppercase;letter-spacing:.07em;color:#8b94a3;ma
 /* And the piles that are NOT targets step back while a discard is being aimed, so the three that
    are stand out from the Salvage Yard beside them. */
 .cardrow.aiming .handcard:not(.target){opacity:.4}
-.stock .ld.addable,.stock .mt.addable{cursor:pointer;border-radius:3px;padding:0 4px;
-  outline:1px solid #5aa9e6;background:rgba(90,169,230,.16)}
-.stock .ld.addable:hover,.stock .mt.addable:hover{background:rgba(90,169,230,.34)}
+/* A CAR YOU MAY ADD IS AN ACTION, SO IT WEARS THE ACTION COLOUR — Jesse, playtest 2026-09-16: the
+   highlight was "just a small bold and basically the same color as everything else", which is the
+   whole difficulty with making the yard chip the button. #c8912f is the border colour that the
+   action buttons themselves use, so a clickable car looks like every other thing inviting a click,
+   rather than like a number that happens to be outlined.
+
+   THE LOADED/EMPTY TEXT COLOURS ARE LEFT ALONE. Green and blue-grey are what say which of the two
+   numbers is which; the amber answers "may I click this", which is a different question, and
+   painting over the first to answer the second would cost real information. */
+.stock .ld.addable,.stock .mt.addable{cursor:pointer;border-radius:3px;padding:0 4px;font-weight:700;
+  outline:2px solid #c8912f;background:rgba(200,145,47,.20);box-shadow:0 0 0 2px rgba(200,145,47,.16)}
+.stock .ld.addable:hover,.stock .mt.addable:hover{background:rgba(200,145,47,.38)}
 /* Twelve Stages across, so a Day is one glance. The current Stage is lit, Stages already gone are
    dimmed, and a slot the die has just filled flashes once. */
 .tt{display:flex;gap:3px;flex-wrap:wrap}

src/web/play.html

@@ -124,7 +124,7 @@ section{background:var(--panel);border:1px solid var(--line);border-radius:7px;
 #watching-what{flex:1;min-width:0;overflow:hidden;text-overflow:ellipsis;white-space:nowrap}
 .wbehind{font-variant-numeric:tabular-nums;font-weight:700;color:#c9cee0;
   background:#22263a;border:1px solid #343a52;border-radius:10px;padding:1px 8px;white-space:nowrap}
-#watching-skip{font-size:11px;padding:2px 10px;border-color:#e0b060;color:#e0b060;flex:none}
+#watching-skip,#watching-pause{font-size:11px;padding:2px 10px;border-color:#e0b060;color:#e0b060;flex:none}
 #watching-who{color:#c9cee0;font-weight:700}
 #presence:empty{display:none}
 /* division strip */
@@ -184,6 +184,14 @@ button.cardact.discard{color:#d6b48a}
 .scheduled{display:block;font-size:11.5px;margin:0 0 7px;padding:3px 9px;border-radius:5px;
   background:rgba(40,140,60,.22);border:1px solid #2f6b47;color:#bfe8cd;font-weight:600}
 .makeup-note{font-size:11px;margin:0 0 5px}
+/* WHAT THE TRAIN STILL WANTS. Deliberately NOT amber: amber means "you can click this" everywhere
+   else on this page, and this line is the REASON for clicking rather than a thing to click — the
+   cars in the Division Yard are. Brighter than the note beneath it, because it answers the question
+   the player actually has while looking at it. `done` goes green like `.scheduled`: a complete
+   consist is good news, not an instruction. */
+.makeup-needs{font-size:12px;margin:0 0 5px;color:#e6e9ee}
+.makeup-needs b{color:#f2e6cf}
+.makeup-needs.done{color:#bfe8cd}
 /* WHICH TRAIN AM I SWITCHING. A row of crews rather than a stacked list — they are alternatives,
    and the chosen one is the crew whose squares the board is drawing, so it wears the same violet
    "you are here" the rest of the page uses. */
@@ -878,7 +886,7 @@ ul.blocked li{padding:2px 0}
        the front door: `index.html`'s two doors are `play.html?lobby` and `play.html?solitaire`, so
        arriving from the splash REPLACES the query string and any pace with it. Jesse played a whole
        game believing he was at 7x when he was at 1x. -->
-  <span class="zoom" title="How long another player's or a bot's move is held on screen before the next one. Yours are never delayed. Off draws every move at once, as it did before v0.8.0.">
+  <span class="zoom" title="How long another player's or a bot's move is held on screen before the next one. Your own moves are never delayed — only theirs. Starts at 1×, which holds a switching move for one second; the slowest setting, 20×, holds it for twenty. Off draws every move at once.">
     <button id="paceslower" aria-label="Slower playback">−</button><span id="pacelabel">1×</span><button id="pacefaster" aria-label="Faster playback">+</button>
   </span>
   <button id="undo" title="Take the last action back. The save is the seed plus the moves made, so this replays the game without the last one — as far back as you like.">Undo</button>
@@ -919,6 +927,10 @@ ul.blocked li{padding:2px 0}
        the other end of the row — Jesse, 2026-09-09: "the skip button should be on the far left, in
        front of where it says [the count], so it's always close to where people are looking." -->
   <button id="watching-skip" class="ghost" type="button" title="Stop animating and jump the board to where the game actually is. Nothing is lost — every line is already in the History panel.">Skip</button>
+  <!-- PAUSE BESIDE SKIP, not instead of it: they are opposite answers to "that went past too fast".
+       Skip gives up the animation to reach the game; Pause holds the board on the step being shown
+       for as long as you want to look at it, and gives the step back the dwell it still had. -->
+  <button id="watching-pause" class="ghost" type="button" title="Hold the board on the move being shown. Nothing is lost and nothing is hurried — press again to carry on from the same step.">Pause</button>
   <span id="watching-behind" class="wbehind"></span>
   <span id="watching-what"></span>
 </div>
@@ -931,6 +943,10 @@ ul.blocked li{padding:2px 0}
       <h2><span id="districtwho">Your Office Area</span>
         <span class="dim" style="text-transform:none;letter-spacing:0">— hover any card for the full explanation</span>
         <span id="districttoggle" class="seg" role="group" aria-label="When to show your Office Area"><button id="dm-auto" class="ghost" type="button" title="Open during Local Operations and Cargo — the phases that change the district — and folded otherwise.">Auto-hide</button><button id="dm-open" class="ghost" type="button" title="Keep the Office Area open in every phase.">Always show</button><button id="dm-closed" class="ghost" type="button" title="Keep the Office Area folded in every phase. The summary line stays, so it reads as folded rather than missing.">Always hide</button></span>
+        <!-- LOOK AT ANOTHER PLAYER'S OFFICE AREA. Filled by `renderDistrict` with one button per
+             opponent, and collapsed at a table of one. The look is read-only and lasts a single
+             render on purpose — see `peekPlayer` in `main.ts`. -->
+        <span id="districtpeek" class="seg" role="group" aria-label="Look at another player's Office Area"></span>
       </h2>
       <div id="districtsummary" class="dim"></div>
       <!-- THE RULE THAT SHAPES EVERY DISTRICT, said once where the district is.

src/web/replays.ts

@@ -56,7 +56,12 @@ type Step = {
 function rebuild(save: Save): { steps: Step[]; stoppedEarly: boolean } {
   const s = createGame({ id: `replay-${save.seed}`, seed: save.seed, config: SOLO_CONFIG, playerNames: ['player'] });
   const steps: Step[] = [];
-  const ctx = { cardName: (id: string) => cardName(s, id), trainName: (id: string) => trainName(s, id) };
+  const ctx = {
+    cardName: (id: string) => cardName(s, id),
+    trainName: (id: string) => trainName(s, id),
+    // A player index is not a seat index, so the fallback names no number at all — see `game.ts`.
+    playerName: (p: number) => s.players[p]?.name ?? 'another player',
+  };
   const push = (events: ReturnType<typeof pump>): void => {
     const lines = events
       .filter((e) => e.type !== 'actorChanged')

src/web/step-queue.ts

@@ -19,7 +19,7 @@
 
 import type { PublicFrame } from '../sim/view.ts';
 import type { DisplayStep } from '../sim/display-step.ts';
-import { applyPublicDelta, changedPiles } from '../sim/public-delta.ts';
+import { applyPublicDelta, changedDivisionCards, changedPiles } from '../sim/public-delta.ts';
 import type { PileKey } from '../sim/public-delta.ts';
 import { dwellForStep } from '../sim/pacing.ts';
 
@@ -53,8 +53,62 @@ export type StepQueue = {
   lit(): readonly PileKey[];
   /** True while there is anything left to show. */
   busy(): boolean;
+  /**
+   * How many narrated lines belong to steps NOT yet shown.
+   *
+   * The log and the board are two different moments while the queue is behind: a push carries its
+   * narration and its steps together, so every line of a bot's turn is in the history panel before the
+   * board has drawn a single move of it (playtest, 2026-09-15: *"is it possible to stall history so it
+   * stays in sync with the number behind?"*). Those lines are the TAIL of the log — they arrived last —
+   * so the caller holds back exactly this many and reveals each as its step goes up.
+   */
+  pendingLines(): number;
+  /** Division nodes whose card changed in the step now on screen, for the map to flash. */
+  flashing(): readonly number[];
+  /**
+   * HOLD THE PLAYBACK ON THE STEP NOW SHOWING — Jesse, playtest 2026-09-16, asking for a pause
+   * beside Skip.
+   *
+   * Skip is the only control the row has had, and it is one-way and total: the way to look harder at
+   * a move that just went past was to not be too slow about it. Pause is the opposite lever — the
+   * board stops where it is and nothing is consumed, so a player can read the caption, look at the
+   * district and then carry on from exactly that step.
+   *
+   * TAKES `now` BECAUSE THE QUEUE OWNS NO CLOCK (see the note at the top of this file). The dwell
+   * still owing is preserved across the hold rather than being spent while nobody was watching:
+   * `resume` pushes the deadline out by however long the pause lasted, so a step paused with 200ms
+   * left resumes with 200ms left instead of vanishing on the next frame.
+   *
+   * Returns false when there is nothing to hold, or nothing being held.
+   */
+  pause(now: number): boolean;
+  resume(now: number): boolean;
+  paused(): boolean;
 };
 
+/**
+ * WHOSE MOVE THE SCREEN IS SHOWING (Gitea#25).
+ *
+ * The game and the board on screen are two different moments. The server plays every bot move the
+ * instant a human's turn ends (`driveBots`), so the LIVE game is nearly always waiting on the human —
+ * while this queue is still replaying the bots, step by step. The turn chart and the Division map's
+ * move marker read the live actor, so a table of one person and three bots said "waiting on" that
+ * person throughout, against a playback row naming the bot actually moving.
+ *
+ * While the queue is behind or still showing a step, the answer is that step's player — `null` for an
+ * automatic phase, which is "the Division is running itself". Otherwise it is the live actor, and
+ * `replaying` is false so a caller can keep live-only detail, such as a ruling the game is waiting on,
+ * off a screen that has not caught up with it yet.
+ */
+export function actorOnScreen(
+  queue: Pick<StepQueue, 'behind' | 'busy' | 'showing'>,
+  live: number | null,
+): { actor: number | null; replaying: boolean } {
+  if (queue.behind() === 0 && !queue.busy()) return { actor: live, replaying: false };
+  const shown = queue.showing();
+  return shown === null ? { actor: live, replaying: false } : { actor: shown.player, replaying: true };
+}
+
 /**
  * `pace` is read on every step rather than captured, so changing the setting takes effect at once.
  *
@@ -74,9 +128,12 @@ export function createStepQueue(
   let shown: PublicFrame | null = null;
   let last: DisplayStep | null = null;
   let litPiles: readonly PileKey[] = [];
+  let flashedCards: readonly number[] = [];
   let pending: DisplayStep[] = [];
   /** When the step now on screen is due to give way. Null when nothing is waiting. */
   let dueAt: number | null = null;
+  /** When the player pressed Pause, so `resume` can give the current step back the time it had. */
+  let pausedAt: number | null = null;
 
   /** How long this step holds the screen — zero for the viewer's own moves; see above. */
   const dwell = (step: DisplayStep): number =>
@@ -92,6 +149,9 @@ export function createStepQueue(
      * that gives your own steps no dwell — the display is for watching everybody else.
      */
     litPiles = step.player !== null && step.player === viewer() ? [] : changedPiles(before, shown);
+    // A Realignment changes the Division under everyone, so it is flashed for the player who did it
+    // too — unlike a pile, which only tells the drawer what they already know.
+    flashedCards = changedDivisionCards(before, shown);
   };
 
   return {
@@ -99,8 +159,12 @@ export function createStepQueue(
       shown = frame;
       pending = [];
       dueAt = null;
+      // A reconnect, an undo or a fresh deal replaces the board outright; a hold on the playback that
+      // no longer exists would leave the row stuck reading "paused" with nothing behind it.
+      pausedAt = null;
       // Nothing was watched arriving at this board, so nothing on it is lit.
       litPiles = [];
+      flashedCards = [];
       // `last` deliberately survives: a reconnect should not blank the caption line, and the
       // sentence describing the most recent action is still true.
     },
@@ -110,6 +174,9 @@ export function createStepQueue(
     },
 
     advance(now) {
+      // Held. Nothing is shown and, crucially, nothing is CONSUMED — `dueAt` is left where it was
+      // and `resume` moves it, so the hold costs the current step none of its dwell.
+      if (pausedAt !== null) return false;
       if (pending.length === 0) {
         // The LAST step of a burst still owes its dwell. Clearing `dueAt` here reported the queue
         // idle the instant that step was shown, which snapped the district panel home before anyone
@@ -143,6 +210,9 @@ export function createStepQueue(
     },
 
     skip() {
+      // Skipping while held is a decision to stop watching, so it also lifts the hold — otherwise the
+      // board would jump to the game and then sit there paused, with a Resume button that does nothing.
+      pausedAt = null;
       if (pending.length === 0) return false;
       for (const step of pending) show(step);
       pending = [];
@@ -150,6 +220,24 @@ export function createStepQueue(
       return true;
     },
 
+    pause(now) {
+      if (pausedAt !== null) return false;
+      // Nothing on screen owes any time and nothing is queued: there is no playback to hold.
+      if (pending.length === 0 && dueAt === null) return false;
+      pausedAt = now;
+      return true;
+    },
+
+    resume(now) {
+      if (pausedAt === null) return false;
+      // Give the step on screen back exactly the dwell it was holding when the player pressed Pause.
+      if (dueAt !== null) dueAt += now - pausedAt;
+      pausedAt = null;
+      return true;
+    },
+
+    paused: () => pausedAt !== null,
+
     current: () => shown,
     behind: () => pending.filter((s) => dwell(s) > 0).length,
     showing: () => last,
@@ -165,6 +253,8 @@ export function createStepQueue(
      * `dueAt` is non-null exactly while the step on screen has time left, so the two together mean
      * "there is more to come, or what is up has not had its moment yet".
      */
+    pendingLines: () => pending.reduce((n, s) => n + s.lines.length, 0),
+    flashing: () => flashedCards,
     busy: () => pending.length > 0 || dueAt !== null,
   };
 }

test/advance.test.ts

@@ -1141,7 +1141,12 @@ describe('the history says WHY a train moved, and says it truthfully', () => {
 
     // A train ahead of it in the same Subdivision, running the SAME way — §8.1's fourth condition,
     // which is the Superintendent's call rather than an absolute bar.
-    const ahead = s.division.nodes.findIndex((n) => n.kind === 'mainline');
+    //
+    // AHEAD MEANS EAST OF THE OFFICE for this eastbound train. This used to take the FIRST Mainline card
+    // in the Division, which is west of the Office — behind the train — and still expected a ruling,
+    // which is exactly the fault Gitea#26 reported. The card is now one the train would actually follow.
+    const office = s.division.nodes.findIndex((n) => n.kind === 'office' && n.seat === 0);
+    const ahead = s.division.nodes.findIndex((n, i) => i > office && n.kind === 'mainline');
     const node = s.division.nodes[ahead];
     assert.equal(node?.kind, 'mainline');
     s.trays.set('ahead', {
@@ -1511,3 +1516,69 @@ describe('an Extra starts where the player puts it (Gitea#4)', () => {
     assert.equal(check(s, 0, at), 'NO_EXTRA_PENDING');
   });
 });
+
+// ---------------------------------------------------------------------------
+
+describe('§8.1 counts only trains AHEAD of the one departing (Gitea#26)', () => {
+  /**
+   * REPORTED from playtesting v0.8.0.9: two westbound Extras, X15 at an Office and X18 still crossing
+   * the card to its EAST. The Superintendent was asked to rule on X15 against X18 — a train behind it —
+   * and holding X15 kept the Whistle Post's only A/D track full, so X18 arrived into it and was
+   * destroyed. Reproduced by replaying the exported save; the positions below are that situation in a
+   * one-seat Division, where every Office is a Whistle Post and the Subdivision spans them all.
+   */
+  const setup = (occupant: { direction: 'east' | 'west'; side: 'east' | 'west'; number: number }) => {
+    const s = game(7, { days: 5 });
+    const area = areaOf(s, 0);
+    const office = s.division.nodes.findIndex((n) => n.kind === 'office' && n.seat === 0);
+    s.trays.set('departing', {
+      id: 'departing', trainNumber: 15, trainIsExtra: true, engineAt: 0, consist: [],
+      direction: 'west', facing: 'w',
+      position: { at: 'grid', seat: 0, coord: area.officeCoord }, movesUsed: 0,
+    });
+    area.adOccupancy.push('departing');
+    const card = s.division.nodes.findIndex((n, i) =>
+      n.kind === 'mainline' && (occupant.side === 'east' ? i > office : i < office));
+    const node = s.division.nodes[card];
+    assert.equal(node?.kind, 'mainline');
+    s.trays.set('other', {
+      id: 'other', trainNumber: occupant.number, trainIsExtra: true, engineAt: 0, consist: [],
+      direction: occupant.direction, facing: occupant.direction === 'east' ? 'e' : 'w',
+      position: { at: 'mainline', index: card }, movesUsed: 0,
+    });
+    if (node?.kind === 'mainline') {
+      node.transits.push({ tray: 'other', stagesRemaining: 2, stagesTotal: 2, direction: occupant.direction });
+    }
+    s.clock.phase = 'mainline';
+    return s;
+  };
+
+  it('does not put a same-direction train BEHIND the departing one to the Superintendent', () => {
+    const s = setup({ direction: 'west', side: 'east', number: 18 });
+    const r = advance(s);
+    assert.ok(!r.events.some((e) => e.type === 'clearanceRequested'), 'a train behind was put to the Superintendent');
+    assert.ok(
+      r.events.some((e) => e.type === 'trainHighballed' && e.trainNumber === 15),
+      'the departing train did not highball with nothing ahead of it',
+    );
+    assert.ok(!r.events.some((e) => e.type === 'trainsDestroyed'), 'a train was destroyed');
+  });
+
+  it('does not bar a departure over an opposite-direction train BEHIND it, which is moving away', () => {
+    const s = setup({ direction: 'east', side: 'east', number: 18 });
+    const r = advance(s);
+    assert.ok(
+      r.events.some((e) => e.type === 'trainHighballed' && e.trainNumber === 15),
+      'a train moving away behind it held the departure',
+    );
+  });
+
+  it('still puts a same-direction train AHEAD to the Superintendent', () => {
+    const s = setup({ direction: 'west', side: 'west', number: 18 });
+    const r = advance(s);
+    assert.ok(
+      r.events.some((e) => e.type === 'clearanceRequested' && e.trainId === 'departing'),
+      'a train the departing one would follow was not put to the Superintendent',
+    );
+  });
+});

test/apply.test.ts

@@ -159,6 +159,60 @@ describe('Local Operations: drawing (§6.2)', () => {
     assert.notEqual(s.decks.departments[1]![0], target, 'refilled with the same card');
   });
 
+  it('a PLAYED timetabled train never comes back, but a discarded one does — Gitea#23', () => {
+    /**
+     * Jesse's ruling, 2026-09-10: *"Once you've played a regularly scheduled train and it's in the
+     * salvage deck, that train is already on the timetable. It does not make sense to put that back
+     * into a reshuffled home deck to get played again. By contrast, a regularly scheduled train
+     * that's in a discard pile could potentially get reused later, and so should have that
+     * capability. Extras run one time and then they're done — if they are in the Salvage deck, they
+     * should get shuffled back in so that they could get run again."*
+     *
+     * So the test is WHERE the card is, not only what it is: the same card is spent in the Salvage
+     * Yard and still runnable in a Department. That is what this pins, because it is the kind of rule
+     * a later tidy-up would happily "simplify" into filtering by card kind everywhere.
+     */
+    const s = game();
+    const kindOfCard = (id: string): string => s.cards.get(id)?.kind.kind ?? '?';
+    const pool = [...s.decks.homeOffice];
+    const trains = pool.filter((id) => kindOfCard(id) === 'timetabledTrain');
+    const extras = pool.filter((id) => kindOfCard(id) === 'extraTrain');
+    const others = pool.filter((id) => !['timetabledTrain', 'extraTrain'].includes(kindOfCard(id)));
+    assert.ok(trains.length >= 2 && extras.length >= 1 && others.length >= 5, 'the deal lacks the cards this needs');
+
+    const spentTrain = trains[0]!;      // played: in the Salvage Yard, its slot taken
+    const discardedTrain = trains[1]!;  // never played: sitting in a Department
+    const playedExtra = extras[0]!;     // a single run, free to run again
+
+    s.decks.salvageYard = [spentTrain, playedExtra, ...others.slice(0, 3)];
+    s.decks.departments = [[discardedTrain], [others[3]!], [others[4]!]];
+    s.decks.homeOffice = [others[5]!];
+
+    applyIntent(s, 0, { type: 'localOps.choose', option: 'draw' });
+    const r = applyIntent(s, 0, { type: 'draw.fromHomeOffice' });
+    assert.ok(r.ok);
+    assert.ok(r.events.some((e) => e.type === 'deckReshuffled'), 'no reshuffle was emitted');
+
+    const recovered = new Set([...s.decks.homeOffice, ...s.decks.departments.flat()]);
+    const hands = new Set([...s.decks.hands.values()].flat());
+
+    // THE RULING, both halves.
+    assert.ok(!recovered.has(spentTrain), 'a played timetabled train was shuffled back in');
+    assert.ok(!hands.has(spentTrain), 'a played timetabled train was dealt back into a hand');
+    assert.ok(
+      s.decks.salvageYard.includes(spentTrain),
+      'a played timetabled train should stay in the Salvage Yard, not vanish',
+    );
+    assert.ok(
+      recovered.has(discardedTrain) || hands.has(discardedTrain),
+      'a DISCARDED timetabled train must come back — it was never played, so its slot is open',
+    );
+    assert.ok(
+      recovered.has(playedExtra) || hands.has(playedExtra),
+      'a played Extra must come back — an Extra is one run, not a standing slot',
+    );
+  });
+
   it('reshuffles the Salvage Yard and Departments back in when the deck runs out', () => {
     // §6.2 — "If drawing a card has depleted the Home Office deck, immediately collect all cards
     // from the Salvage Yard and three Department decks, reshuffle, and reestablish the Home Office
@@ -182,7 +236,11 @@ describe('Local Operations: drawing (§6.2)', () => {
     assert.ok(r.ok);
     assert.ok(r.events.some((e) => e.type === 'deckReshuffled'), 'no reshuffle was emitted');
 
-    assert.equal(s.decks.salvageYard.length, 0, 'the Salvage Yard must be swept');
+    // Swept EXCEPT the trains whose slots are already on the timetable — see the ruling test below.
+    assert.ok(
+      s.decks.salvageYard.every((id) => s.cards.get(id)?.kind.kind === 'timetabledTrain'),
+      'the Salvage Yard must be swept apart from spent timetabled trains',
+    );
     assert.ok(s.decks.homeOffice.length > 0, 'the deck must be re-established');
     assert.ok(
       s.decks.departments.every((p) => p.length === 1),

test/public-delta.test.ts

@@ -254,3 +254,41 @@ describe('which piles a step moved', () => {
     assert.deepEqual(changedPiles(null, publicSnapshot(s)), []);
   });
 });
+
+describe('a Mainline card that changed under the players (Gitea#28)', () => {
+  it('names the node a Realignment converted, and nothing else', async () => {
+    const { changedDivisionCards } = await import('../src/sim/public-delta.ts');
+    const { REALIGNMENTS } = await import('../src/engine/content.ts');
+
+    const s = newState(4242);
+    const before = publicSnapshot(s);
+
+    // Realignment converts a card to another kind (`content.ts`'s table). Applied to the state directly:
+    // what is being tested is the DETECTOR, not the play that reaches it — which needs the card in hand,
+    // the draw option taken and no train on the card.
+    const at = s.division.nodes.findIndex((n) => n.kind === 'mainline' && REALIGNMENTS.some((r) => r.from === n.card));
+    assert.ok(at >= 0, 'no Mainline card in this Division can be realigned at all');
+    const node = s.division.nodes[at];
+    assert.equal(node?.kind, 'mainline');
+    if (node?.kind === 'mainline') {
+      node.card = REALIGNMENTS.find((r) => r.from === node.card)!.to;
+    }
+    const after = publicSnapshot(s);
+
+    assert.deepEqual(changedDivisionCards(before, after), [at], 'the realigned card was not the one reported');
+    assert.deepEqual(changedDivisionCards(after, after), [], 'an unchanged Division reported a change');
+    assert.deepEqual(changedDivisionCards(null, after), [], 'a first board has nothing to compare against');
+  });
+
+  it('says nothing when only the trains on a card moved', async () => {
+    const { changedDivisionCards } = await import('../src/sim/public-delta.ts');
+    const s = newState(1917398);
+    const before = publicSnapshot(s);
+    const at = s.division.nodes.findIndex((n) => n.kind === 'mainline');
+    const node = s.division.nodes[at];
+    if (node?.kind === 'mainline') {
+      node.transits.push({ tray: 'tray1', stagesRemaining: 1, stagesTotal: 1, direction: 'east' });
+    }
+    assert.deepEqual(changedDivisionCards(before, publicSnapshot(s)), [], 'a train arriving flashed the card');
+  });
+});

test/replay.test.ts

@@ -146,6 +146,33 @@ describe('narration', () => {
     assert.match(loss.text, /COLLISION/i);
   });
 
+  /**
+   * WHOSE OFFICE, AND WHOSE TRAIN — Jesse, playtest 2026-09-16.
+   *
+   * The line went to every seat reading "ARRIVED at the Whistle Post … You can work it in Cargo now".
+   * At a table of four that is one true sentence and one false one: every seat has an Office, so the
+   * tier alone does not say which district the train is standing in, and three of the four readers
+   * cannot touch it. Constructed here rather than fished out of a game so both halves are pinned
+   * exactly, and the resolver-less path is checked too — the replay viewers pass no names for a
+   * table they do not have.
+   */
+  it('names WHOSE Office a train reached, and never tells the table they can work it', () => {
+    const named = narrate(
+      { type: 'trainArrived', trainNumber: 3, consist: [], office: 'Whistle Post', owner: 1, expedited: false },
+      { playerName: () => 'Tom' },
+    );
+    assert.match(named.text, /Tom's Whistle Post/, `the arrival did not name the Office's owner: ${named.text}`);
+    assert.match(named.text, /Tom can work it/, `the arrival did not say whose train it is to work: ${named.text}`);
+    assert.doesNotMatch(named.text, /\bYou can work it\b/i, `the arrival still addresses every reader: ${named.text}`);
+
+    // No resolver — still English, and still no raw index leaking into a sentence.
+    const anon = narrate({
+      type: 'trainArrived', trainNumber: 3, consist: [], office: 'Whistle Post', owner: 1, expedited: false,
+    });
+    assert.match(anon.text, /at the Whistle Post/, anon.text);
+    assert.doesNotMatch(anon.text, /undefined|\bplayer \d+\b/i, anon.text);
+  });
+
   it('points at the board cell where something happened', () => {
     const n = narrate({ type: 'loadCompleted', player: 0, at: { row: 1, col: 2 }, carType: 'hopper' });
     assert.deepEqual(n.where, { row: 1, col: 2 });

test/route-cache.test.ts

@@ -0,0 +1,88 @@
+/**
+ * The engine's speed-ups must not change a single game (2026-09-15).
+ *
+ * `applyIntent` became `prepareIntent` (check and execute, sharing one walk of the position's routes)
+ * followed by `commitEvents` (reduce and tally), so the switching planner can decide every candidate
+ * against one position and apply each to a copy. Two properties hold that together:
+ *
+ * 1. `prepareIntent` never writes the state it reads — including through the route cache it opens.
+ * 2. Preparing on one state and committing to an EQUAL copy lands on exactly what `applyIntent` does.
+ *
+ * Checked at every decision of seeded bot games rather than on hand-built positions, so the intents
+ * exercised are the ones real play submits.
+ */
+
+import { describe, it } from 'node:test';
+import assert from 'node:assert/strict';
+
+import { pump } from '../src/engine/advance.ts';
+import { applyIntent, commitEvents, prepareIntent } from '../src/engine/apply.ts';
+import {
+  DEFAULT_MAX_COLLISIONS_PER_DAY,
+  DEFAULT_MAX_COLLISIONS_TOTAL,
+  collectiveRevenueFloor,
+  lengthProfile,
+} from '../src/engine/content.ts';
+import { legalActions } from '../src/engine/legal.ts';
+import { createGame } from '../src/engine/setup.ts';
+import type { GameConfig, GameState } from '../src/engine/state.ts';
+import { developerBot, playGame } from '../src/sim/bot.ts';
+
+const config = (): GameConfig => {
+  const days = lengthProfile('short').days;
+  return {
+    mode: 'solitaire',
+    days,
+    minCombinedRevenue: collectiveRevenueFloor(1, days),
+    maxCollisionsPerDay: DEFAULT_MAX_COLLISIONS_PER_DAY,
+    maxCollisionsTotal: DEFAULT_MAX_COLLISIONS_TOTAL,
+    pvpCardsAllowed: false,
+    optionalRules: { reducedVisibility: false, employeeRotation: false, emergencyToolbox: false },
+  };
+};
+
+const serialise = (s: GameState): string =>
+  JSON.stringify(s, (_k, v) => (v instanceof Map ? [...v] : v instanceof Set ? [...v] : v));
+
+describe('applyIntent split into prepareIntent and commitEvents', () => {
+  it('prepares without writing, and committing to a copy matches applying in place', () => {
+    let decisions = 0;
+    let rejectedSeen = 0;
+    const s = createGame({ id: 'split-8919', seed: 8919, config: config(), playerNames: ['bot'] });
+    const policy = {
+      name: 'split-probe',
+      choose(st: GameState, player: number, options: ReturnType<typeof legalActions>) {
+        const chosen = developerBot.choose(st, player, options);
+        if (decisions < 400) {
+          decisions++;
+          const before = serialise(st);
+          const prepared = prepareIntent(st, player, chosen);
+          assert.equal(serialise(st), before, `decision ${decisions}: prepareIntent wrote into the state it read`);
+          assert.ok(prepared.ok, `decision ${decisions}: a legal choice was refused by prepareIntent`);
+
+          const viaCommit = structuredClone(st);
+          const viaApply = structuredClone(st);
+          commitEvents(viaCommit, prepared.events);
+          const applied = applyIntent(viaApply, player, chosen);
+          assert.ok(applied.ok);
+          assert.deepEqual(applied.events, prepared.events, `decision ${decisions}: the two paths produced different events`);
+          assert.equal(serialise(viaCommit), serialise(viaApply), `decision ${decisions}: committing to a copy diverged from applying`);
+
+          // A refused intent must come back refused from both paths, with nothing written.
+          const refused = { type: 'switch.end' } as const;
+          const r = prepareIntent(st, player, refused);
+          if (!r.ok) {
+            rejectedSeen++;
+            assert.equal(serialise(st), before);
+            assert.equal(applyIntent(structuredClone(st), player, refused).ok, false);
+          }
+        }
+        return chosen;
+      },
+    };
+    const r = playGame(s, policy, pump);
+    assert.ok(r.finished, 'the probed game did not finish');
+    assert.ok(decisions > 0, 'no decision was probed');
+    assert.ok(rejectedSeen > 0, 'no refused intent was exercised');
+  });
+});

test/server/claims.test.ts

@@ -0,0 +1,68 @@
+/**
+ * Seat recovery codes — Gitea#33.
+ *
+ * The properties worth pinning are the ones that make a code safe to put in a link: it is spendable
+ * exactly once, it stops working on its own, and a bad code is indistinguishable from a spent one.
+ * `now` is a parameter rather than a clock, so expiry is tested without faking timers.
+ */
+
+import { describe, it } from 'node:test';
+import assert from 'node:assert/strict';
+
+import { CLAIM_TTL_MS, createClaimStore } from '../../src/server/claims.ts';
+
+describe('seat recovery codes', () => {
+  it('mints a code that names the seat it was minted for', () => {
+    const claims = createClaimStore();
+    const { code, expiresAt } = claims.mint('tok-abc', 'game-1', 1000);
+    assert.equal(expiresAt, 1000 + CLAIM_TTL_MS);
+    assert.deepEqual(claims.redeem(code, 1000), { token: 'tok-abc', gameId: 'game-1' });
+  });
+
+  it('spends a code exactly once — a link in a chat log is worth nothing afterwards', () => {
+    const claims = createClaimStore();
+    const { code } = claims.mint('tok-abc', 'game-1', 0);
+    assert.ok(claims.redeem(code, 1));
+    assert.equal(claims.redeem(code, 2), null, 'the same code was accepted twice');
+  });
+
+  it('stops working once its time is up, without anything having to sweep it', () => {
+    const claims = createClaimStore();
+    const { code } = claims.mint('tok-abc', 'game-1', 0);
+    assert.equal(claims.redeem(code, CLAIM_TTL_MS - 1)?.token, 'tok-abc', 'expired early');
+    const again = claims.mint('tok-abc', 'game-1', 0).code;
+    assert.equal(claims.redeem(again, CLAIM_TTL_MS), null, 'a code outlived its expiry');
+  });
+
+  it('answers the same way for unknown, spent and expired codes', () => {
+    const claims = createClaimStore();
+    const { code } = claims.mint('tok-abc', 'game-1', 0);
+    claims.redeem(code, 1);
+    const expired = claims.mint('tok-abc', 'game-1', 0).code;
+
+    assert.equal(claims.redeem('never-existed', 1), null);
+    assert.equal(claims.redeem(code, 1), null);
+    assert.equal(claims.redeem(expired, CLAIM_TTL_MS + 1), null);
+  });
+
+  it('gives every mint its own code', () => {
+    const claims = createClaimStore();
+    const codes = new Set([0, 1, 2, 3, 4].map(() => claims.mint('tok-abc', 'game-1', 0).code));
+    assert.equal(codes.size, 5, 'two mints produced the same code');
+  });
+
+  it('forgets expired codes rather than accumulating them', () => {
+    const claims = createClaimStore();
+    claims.mint('tok-a', 'game-1', 0);
+    claims.mint('tok-b', 'game-1', 0);
+    assert.equal(claims.outstanding(0), 2);
+    assert.equal(claims.outstanding(CLAIM_TTL_MS), 0, 'expired codes were still being held');
+  });
+
+  it('keeps a short-lived code short-lived when asked for one', () => {
+    const claims = createClaimStore();
+    const { code, expiresAt } = claims.mint('tok-abc', 'game-1', 500, 60_000);
+    assert.equal(expiresAt, 60_500);
+    assert.equal(claims.redeem(code, 60_500), null);
+  });
+});

test/step-queue.test.ts

@@ -15,7 +15,7 @@ import { currentActor, newMultiplayerGame, submit } from '../src/web/game.ts';
 import { publicSnapshot } from '../src/sim/view.ts';
 import { takeSteps } from '../src/sim/display-step.ts';
 import type { DisplayStep } from '../src/sim/display-step.ts';
-import { createStepQueue } from '../src/web/step-queue.ts';
+import { actorOnScreen, createStepQueue } from '../src/web/step-queue.ts';
 import { DWELL } from '../src/sim/pacing.ts';
 
 const config: GameConfig = {
@@ -109,6 +109,61 @@ describe('the step queue', () => {
     assert.equal(q2.behind(), 4, 'and must be replaced once its dwell is up');
   });
 
+  /**
+   * PAUSE — Jesse, playtest 2026-09-16, asking for one beside Skip.
+   *
+   * The property that matters is not "it stops", which any flag gives you. It is that a hold COSTS
+   * THE STEP NOTHING: a move paused half-way through its dwell has to resume with half a dwell left,
+   * or pausing to look at something would punish you by throwing the rest of it away the moment you
+   * let go. That is the whole assertion below, measured against `DWELL.switching` rather than a
+   * hand-picked number so it follows the tuning table.
+   */
+  it('pause holds the board where it is, and resume gives the step back the dwell it had left', () => {
+    const { steps } = realSteps(1917398, 400);
+    const switching = steps.filter((s) => s.cause.startsWith('switch.') && s.cause !== 'switch.end');
+    assert.ok(switching.length >= 6, `only ${switching.length} switching steps found`);
+
+    const q = createStepQueue();
+    q.reset(baseline(1917398));
+    q.push(switching.slice(0, 6));
+    q.advance(0);
+    assert.equal(q.behind(), 5, 'the first is shown at once; five are still to watch');
+
+    // Held half-way through the first move's dwell, and left held for ten times that long.
+    const half = DWELL.switching / 2;
+    assert.equal(q.pause(half), true);
+    assert.equal(q.paused(), true);
+    const held = q.showing()?.seq;
+    q.advance(half + 10_000);
+    assert.equal(q.behind(), 5, 'a held queue consumed a step');
+    assert.equal(q.showing()?.seq, held, 'the board moved while it was supposed to be held');
+    assert.equal(q.busy(), true, 'a held queue must still read busy, or the render loop stops');
+
+    // Resumed, the move still owes exactly the half-dwell it had left — no more, and no less.
+    const at = half + 10_000;
+    assert.equal(q.resume(at), true);
+    assert.equal(q.paused(), false);
+    q.advance(at + half - 1);
+    assert.equal(q.behind(), 5, 'the step was robbed of time it was owed while held');
+    q.advance(at + half);
+    assert.equal(q.behind(), 4, 'and it never gave way once the rest of its dwell was up');
+  });
+
+  it('refuses to hold an idle queue, and Skip lifts a hold rather than leaving it stuck', () => {
+    const q = createStepQueue();
+    q.reset(baseline(1917398));
+    assert.equal(q.pause(0), false, 'an idle queue has no playback to hold');
+    assert.equal(q.paused(), false);
+
+    const { steps } = realSteps(1917398, 400);
+    q.push(steps);
+    q.advance(0);
+    assert.equal(q.pause(10), true);
+    assert.equal(q.skip(), true, 'Skip must still work while held');
+    assert.equal(q.paused(), false, 'Skip left the queue held, with a Resume that does nothing');
+    assert.equal(q.busy(), false);
+  });
+
   it('counts only what will be watched, so the countdown is steady', () => {
     // The counter's whole purpose: a backlog of mostly-bookkeeping must not read as a huge number
     // that collapses the instant it starts.
@@ -278,3 +333,77 @@ describe('the step queue', () => {
     assert.equal(q.showing(), null);
   });
 });
+
+describe('whose move the screen is showing (Gitea#25)', () => {
+  const step = (player: number | null) => ({ player }) as DisplayStep;
+  const queue = (behind: number, busy: boolean, showing: DisplayStep | null) => ({
+    behind: () => behind,
+    busy: () => busy,
+    showing: () => showing,
+  });
+
+  it('names the live actor once the board has caught up', () => {
+    assert.deepEqual(actorOnScreen(queue(0, false, step(2)), 0), { actor: 0, replaying: false });
+  });
+
+  it('names the player of the step on screen while the board is behind — not the live actor', () => {
+    // One human (seat 0) against bots: the live game already waits on seat 0 while bot 2's moves replay.
+    assert.deepEqual(actorOnScreen(queue(3, true, step(2)), 0), { actor: 2, replaying: true });
+  });
+
+  it('keeps naming the last step while it is still on screen, after the counter reaches zero', () => {
+    assert.deepEqual(actorOnScreen(queue(0, true, step(1)), 0), { actor: 1, replaying: true });
+  });
+
+  it('names nobody for an automatic phase being shown', () => {
+    assert.deepEqual(actorOnScreen(queue(2, true, step(null)), 0), { actor: null, replaying: true });
+  });
+
+  it('falls back to the live actor before any step has been shown', () => {
+    assert.deepEqual(actorOnScreen(queue(1, true, null), 3), { actor: 3, replaying: false });
+  });
+});
+
+describe('the log is held back with the board, and a changed card flashes (playtest, 2026-09-15)', () => {
+  it('owes exactly the lines of the steps not yet shown', () => {
+    const { steps } = realSteps(1917398, 400);
+    const q = createStepQueue();
+    q.reset(baseline(1917398));
+    assert.equal(q.pendingLines(), 0, 'an empty queue holds nothing back');
+
+    q.push(steps);
+    const owed = steps.reduce((n, s) => n + s.lines.length, 0);
+    assert.equal(q.pendingLines(), owed, 'every queued step still owes its lines');
+
+    // Drive the clock as a render loop would; the debt falls monotonically and ends at nothing.
+    let now = 0;
+    let last = owed;
+    for (let i = 0; i < 20_000 && q.busy(); i++) {
+      q.advance(now);
+      const left = q.pendingLines();
+      assert.ok(left <= last, 'the held-back count grew while the board caught up');
+      last = left;
+      now += 50;
+    }
+    assert.equal(q.pendingLines(), 0, 'the board caught up but lines were still withheld');
+  });
+
+  it('skipping reveals the whole log at once', () => {
+    const { steps } = realSteps(1917398, 400);
+    const q = createStepQueue();
+    q.reset(baseline(1917398));
+    q.push(steps);
+    q.skip();
+    assert.equal(q.pendingLines(), 0, 'Skip left lines withheld — the history would stay short');
+  });
+
+  it('flashes nothing on an ordinary step', () => {
+    // Realignment is rare in bot play, so this pins the quiet case: the map must not pulse at random.
+    const { steps } = realSteps(1917398, 400);
+    const q = createStepQueue();
+    q.reset(baseline(1917398));
+    q.push(steps.slice(0, 5));
+    q.advance(0);
+    assert.deepEqual(q.flashing(), [], 'a step that changed no Mainline card flashed one');
+  });
+});

test/switch-planner.test.ts

@@ -0,0 +1,100 @@
+/**
+ * The switching planner (`sim/switch-planner.ts`) — the two properties it cannot be allowed to lose.
+ *
+ * 1. PLANNING TOUCHES NOTHING. The planner applies intents to a partial copy of the game
+ *    (`forkForSwitching`) that shares everything a switching intent is not supposed to write. If a
+ *    reducer ever starts writing somewhere new, the copy leaks into the real game, and this is where
+ *    that shows: the game is serialised before and after planning and must not have changed.
+ * 2. A PLAN IS WHAT THE ENGINE WILL DO. Every step replays through `applyIntent` on a FULL copy, and
+ *    lands on the fingerprint the planner promised for it. That is what makes the partial copy
+ *    trustworthy, and it is what the bot relies on to know it is still on plan.
+ *
+ * Taken from real seeded bot games rather than hand-built positions, because a district that
+ * satisfies the track geometry by hand tests the builder as much as the planner.
+ */
+
+import { describe, it } from 'node:test';
+import assert from 'node:assert/strict';
+
+import { pump } from '../src/engine/advance.ts';
+import { applyIntent } from '../src/engine/apply.ts';
+import { legalActions, legalSwitchingActions } from '../src/engine/legal.ts';
+import {
+  DEFAULT_MAX_COLLISIONS_PER_DAY,
+  DEFAULT_MAX_COLLISIONS_TOTAL,
+  MOVES_PER_LOCAL_OPS,
+  collectiveRevenueFloor,
+  lengthProfile,
+} from '../src/engine/content.ts';
+import { createGame } from '../src/engine/setup.ts';
+import type { GameConfig, GameState } from '../src/engine/state.ts';
+import { actingPlayer, turnOf } from '../src/engine/state.ts';
+import { developerBot, playGame } from '../src/sim/bot.ts';
+import { planSwitchingTurn, switchFingerprint } from '../src/sim/switch-planner.ts';
+
+const config = (): GameConfig => {
+  const days = lengthProfile('short').days;
+  return {
+    mode: 'solitaire',
+    days,
+    minCombinedRevenue: collectiveRevenueFloor(1, days),
+    maxCollisionsPerDay: DEFAULT_MAX_COLLISIONS_PER_DAY,
+    maxCollisionsTotal: DEFAULT_MAX_COLLISIONS_TOTAL,
+    pvpCardsAllowed: false,
+    optionalRules: { reducedVisibility: false, employeeRotation: false, emergencyToolbox: false },
+  };
+};
+
+const serialise = (s: GameState): string =>
+  JSON.stringify(s, (_k, v) => (v instanceof Map ? [...v] : v instanceof Set ? [...v] : v));
+
+describe('switching planner', () => {
+  it('asks for switching intents that are exactly the switching subset of legalActions, in order', () => {
+    const SWITCHING = new Set(['switch.move', 'switch.dropCars', 'switch.sortConsist', 'maneuver.flyingSwitch', 'switch.end']);
+    let compared = 0;
+    for (const seed of [1000, 8919]) {
+      const s = createGame({ id: `legal-${seed}`, seed, config: config(), playerNames: ['bot'] });
+      playGame(s, developerBot, pump, 50_000, undefined, (st) => {
+        const p = actingPlayer(st);
+        if (p === null || st.clock.phase !== 'localOps') return;
+        const all = legalActions(st, p).filter((i) => SWITCHING.has(i.type));
+        assert.deepEqual(legalSwitchingActions(st, p), all);
+        compared++;
+      });
+    }
+    assert.ok(compared > 0, 'no Local Operations decision was reached');
+  });
+
+  it('never changes the game it plans for, and every plan replays to the position it promised', () => {
+    let checked = 0;
+    let withSteps = 0;
+    for (const seed of [1000, 8919, 16838]) {
+      const s = createGame({ id: `plan-${seed}`, seed, config: config(), playerNames: ['bot'] });
+      const r = playGame(s, developerBot, pump, 50_000, undefined, (st) => {
+        const p = actingPlayer(st);
+        if (p === null || st.clock.phase !== 'localOps') return;
+        const turn = turnOf(st, p);
+        if (turn.option !== 'switch' || turn.movesRemaining !== MOVES_PER_LOCAL_OPS) return;
+
+        const before = serialise(st);
+        const plan = planSwitchingTurn(st, p, { budget: 400, beam: 16 });
+        assert.equal(serialise(st), before, `seed ${seed}: planning wrote into the real game`);
+        assert.ok(plan.score >= plan.rootScore, 'a plan is never worse than stopping where the crew stands');
+
+        const copy = structuredClone(st);
+        plan.steps.forEach((step, n) => {
+          assert.equal(switchFingerprint(copy, p), plan.keys[n], `seed ${seed}: step ${n} started off plan`);
+          const applied = applyIntent(copy, p, step);
+          assert.ok(applied.ok, `seed ${seed}: step ${n} (${step.type}) was refused by the engine`);
+        });
+        assert.equal(switchFingerprint(copy, p), plan.keys.at(-1), `seed ${seed}: the plan did not end where it said`);
+
+        checked++;
+        if (plan.steps.length > 0) withSteps++;
+      });
+      assert.ok(r.finished, `seed ${seed}: a game with the planner switched on did not finish`);
+    }
+    assert.ok(checked > 0, 'no switching turn was reached, so nothing was tested');
+    assert.ok(withSteps > 0, 'every plan was empty, so replay was never exercised');
+  });
+});

test/tally.test.ts

@@ -151,10 +151,14 @@ describe('the tally counts every event exactly once (Gitea#16)', () => {
 
   it('splits Revenue into what was earned and what was given back', () => {
     // Reconciliation is the real assertion and it holds for any game, earned or not: gained minus
-    // lost IS the score the engine kept. Seed 42 is named because it is one where Revenue actually
+    // lost IS the score the engine kept. Seed 44 is named because it is one where Revenue actually
     // moves in both directions — it earns 1 and gives back 5 to a collision — so the two halves are
     // being told apart rather than both sitting at zero.
-    for (const seed of [1, 7, 42]) {
+    //
+    // It was seed 42 until v0.8.0.10. That game's collision was the Superintendent holding a train over
+    // one BEHIND it (Gitea#26); with the ruling gone the collision is too, and seed 42 now earns 5 and
+    // loses nothing — a better game and a vacuous test. The seed moved, not the assertion.
+    for (const seed of [1, 7, 44]) {
       const { state } = playKeepingEvents(seed);
       const me = state.tally.byPlayer[0]!;
       assert.equal(
@@ -163,10 +167,10 @@ describe('the tally counts every event exactly once (Gitea#16)', () => {
         `seed ${seed}: gained minus lost does not reconcile with the score the engine kept`,
       );
     }
-    const { state } = playKeepingEvents(42);
+    const { state } = playKeepingEvents(44);
     const me = state.tally.byPlayer[0]!;
-    assert.ok(me.revenueGained > 0, 'seed 42 earned nothing — the gained half is not being counted');
-    assert.ok(me.revenueLost > 0, 'seed 42 lost nothing — the lost half is not being counted');
+    assert.ok(me.revenueGained > 0, 'seed 44 earned nothing — the gained half is not being counted');
+    assert.ok(me.revenueLost > 0, 'seed 44 lost nothing — the lost half is not being counted');
   });
 
   it('records a Circus set-up as the one-off it is, not as a streak', () => {

test/watchable.test.ts

@@ -309,3 +309,71 @@ describe('steps reach a seated player — TODO #13', () => {
     assert.ok(seen > 0, 'no steps reached a push, so this proved nothing');
   });
 });
+
+describe('the log says who acted, once, and in what capacity (Gitea#30, #31)', () => {
+  it('never names a player twice — no "Player Jesse player 0 finished …"', async () => {
+    const { newMultiplayerGame, currentActor, submit } = await import('../src/web/game.ts');
+    const { legalActions } = await import('../src/engine/legal.ts');
+
+    const game = newMultiplayerGame(1917398, config, ['Alice', 'Bob', 'Carol']);
+    for (let i = 0; i < 400; i++) {
+      const actor = currentActor(game);
+      if (actor === null) break;
+      const options = legalActions(game.state, actor);
+      if (options.length === 0) break;
+      if (!submit(game, options.find((o) => o.type === 'localOps.choose') ?? options[0]!)) break;
+    }
+
+    assert.ok(game.log.length > 50, 'the game barely ran, so this proved little');
+    for (const line of game.log) {
+      // `record()` prefixes the acting player's NAME; a narration that also named them read
+      // "Player Alice player 0 finished Local Operations" (playtest, 2026-09-15).
+      assert.doesNotMatch(
+        line.text,
+        /\bplayer \d+\b/i,
+        `a line still carries a bare player index: ${line.text}`,
+      );
+    }
+  });
+
+  it('attributes a clearance ruling to the office, not to the seat\'s own turn', async () => {
+    const { newMultiplayerGame, drain, submit } = await import('../src/web/game.ts');
+    const { areaOf } = await import('../src/engine/apply.ts');
+
+    const game = newMultiplayerGame(7, config, ['Alice', 'Bob', 'Carol']);
+    const s = game.state;
+    const area = areaOf(s, 0);
+
+    // A westbound train at seat 0's Office, and another westbound AHEAD of it — west of the Office —
+    // which is §8.1's fourth condition and the Superintendent's to rule on (see Gitea#26).
+    s.trays.set('departing', {
+      id: 'departing', trainNumber: 15, trainIsExtra: true, engineAt: 0, consist: [],
+      direction: 'west', facing: 'w',
+      position: { at: 'grid', seat: 0, coord: area.officeCoord }, movesUsed: 0,
+    } as never);
+    area.adOccupancy.push('departing');
+    const office = s.division.nodes.findIndex((n) => n.kind === 'office' && n.seat === 0);
+    const card = s.division.nodes.findIndex((n, i) => i < office && n.kind === 'mainline');
+    const node = s.division.nodes[card];
+    assert.equal(node?.kind, 'mainline');
+    s.trays.set('ahead', {
+      id: 'ahead', trainNumber: 9, trainIsExtra: false, engineAt: 0, consist: [],
+      direction: 'west', facing: 'w', position: { at: 'mainline', index: card }, movesUsed: 0,
+    } as never);
+    if (node?.kind === 'mainline') {
+      node.transits.push({ tray: 'ahead', stagesRemaining: 2, stagesTotal: 2, direction: 'west' });
+    }
+
+    s.clock.phase = 'mainline';
+    drain(game);
+    assert.equal(s.clock.pendingDecision?.kind, 'clearance', 'no ruling was called for, so nothing was tested');
+
+    const before = game.log.length;
+    assert.ok(submit(game, { type: 'mainline.clearance', allow: false }, s.clock.superintendent));
+    const said = game.log.slice(before).map((l) => l.text);
+    assert.ok(
+      said.some((text) => text.startsWith('Superintendent Player ')),
+      `a ruling did not read as the office's: ${said.join(' | ')}`,
+    );
+  });
+});

test/web.test.ts

@@ -2641,8 +2641,16 @@ describe('the static build', () => {
      */
     const base = { day: 2, stage: 5, clock: '2:20', phase: 'Mainline', phaseKey: 'mainline' };
 
+    /**
+     * AND AN AUTOMATIC PHASE NAMES THE WORK, NOT AN ABSENCE (playtest, 2026-09-16). It used to read
+     * "waiting on nobody — the Division is running itself": an answer by negation, on a line whose
+     * whole job is to say where the game is. `base` is the Mainline Phase, so this is the case Jesse
+     * described — the Division moving trains with nobody to wait for.
+     */
     const idle = turnChartHtml({ ...base, actor: null, awaiting: null }, null, 'Bob');
-    assert.match(idle, /nobody — the Division is running itself/, 'an automatic phase should say so');
+    assert.match(idle, /the Division is moving trains/, 'an automatic phase should say what it is doing');
+    assert.doesNotMatch(idle, /waiting on/, 'nobody is being waited on, so the line must not claim it');
+    assert.doesNotMatch(idle, /nobody/, 'the line still answers by negation');
 
     for (const [asks, train] of [
       ['a clearance ruling', 'Train 4'],
@@ -2767,6 +2775,21 @@ describe('the static build', () => {
     assert.ok(fallback !== '', 'the no-git fallback moved and this test cannot see it any more');
     assert.doesNotMatch(fallback, /^'nogit'$|^"nogit"$/, 'the no-git fallback is a constant again');
     assert.match(fallback, /Date\.now\(\)/, 'the no-git fallback carries nothing that varies per build');
+    /**
+     * AND IT MUST NOT LEAD WITH THE VERSION — which is what fixing the constant first reached for.
+     *
+     * The stamp is `v${pkg.version} · ${git} · ${when}Z`, so a fallback of `${pkg.version}-${…}`
+     * spends the version twice, and does it on precisely the builds that take this path: every
+     * `.s9pk`, because the Dockerfile copies the tree in without `.git`. Reported from play —
+     * Jesse, 2026-09-16: *"the version number is in the header twice."* The uniqueness this test
+     * exists to defend comes from the timestamp, not from the version, so the two requirements do
+     * not compete.
+     */
+    assert.doesNotMatch(
+      fallback,
+      /pkg\.version/,
+      'the no-git fallback repeats the version the stamp already prints in front of it',
+    );
   });
 
   it('lets a build-tagged URL be cached and nothing else', () => {
@@ -3647,34 +3670,6 @@ describe('the Day rolling over says so (Gitea#10)', () => {
     );
   });
 
-  it('names the train on top of the Salvage Yard, instead of "a card"', () => {
-    /**
-     * The Salvage Yard is a FACE-UP pile and its tile reads the top card — but `trainScheduled`
-     * pushes a synthetic `train-<number>` id rather than the id of the card that was played
-     * (`apply.ts`), so there was nothing in `s.cards` to look up and the tile said "a card". A train
-     * is scheduled several times a Day, so that id is on top most of the time: the pile was face up
-     * and had nothing to say. Jesse, 2026-09-10: *"why is salvage deck not face up. I should see the
-     * card played onto salvage."*
-     */
-    const s = createEngineGame({
-      id: 'salv',
-      seed: 5,
-      config: {
-        mode: 'competitive', days: 5, minCombinedRevenue: 60,
-        maxCollisionsPerDay: 0, maxCollisionsTotal: 0, pvpCardsAllowed: false,
-        optionalRules: { reducedVisibility: false, employeeRotation: false, emergencyToolbox: false },
-      },
-      playerNames: ['Joe', 'Bot 1'],
-    });
-    s.decks.salvageYard.push('train-13');
-    const f = snapshot(s, [], null);
-    assert.equal(f.salvage.top, 'Train 13', 'a scheduled train on the pile must be named');
-
-    const html = pilesHtml(f);
-    assert.ok(html.includes('Train 13'), `the Salvage tile does not name the train:\n${html}`);
-    assert.equal(html.includes('>a card<'), false, 'the face-up pile still says "a card"');
-  });
-
   it('lights only the pile a watched move touched', () => {
     const s = createEngineGame({
       id: 'piles2',
@@ -4233,6 +4228,30 @@ describe('the lobby screen', () => {
   const chosen = (groups: Record<string, { value: string; checked: boolean }[]>, name: string): string | undefined =>
     groups[name]!.find((r) => r.checked)?.value;
 
+  /**
+   * A SEAT RECOVERY LINK — Gitea#33.
+   *
+   * The properties that make this safe to hand round are the ones worth pinning: the page trades the
+   * CODE for the token (so no token is ever in a URL), and it does not keep the code afterwards. The
+   * store's own single-use and expiry rules are proven in `test/server/claims.test.ts`; this is the
+   * client half, which is the part that could silently stop asking.
+   */
+  it('trades a ?claim= code for a seat, and does not leave the code in the address bar', async () => {
+    const { sent } = await open('?claim=code-123', {
+      '/api/claim': { token: 'tok-restored', gameId: 'game-9', player: 1, gameCode: 'WHISTLE-6945' },
+    });
+
+    const claim = sent.find((r) => r.url.includes('/api/claim'));
+    assert.ok(claim, 'the page never redeemed the code');
+    assert.deepEqual(claim.body, { code: 'code-123' }, 'the code was not sent as the request body');
+    // The token must never travel in a URL (`lobby-and-sessions.md` §1) — it comes back in the
+    // response, and the only thing that went out was the one-time code.
+    assert.ok(
+      !sent.some((r) => r.url.includes('tok-restored')),
+      'a session token appeared in a request URL',
+    );
+  });
+
   it('opens on the join door, with the create form behind it', async () => {
     // Somebody who was handed a code used to have to scroll past the entire create form to find the
     // box to type it into.