e5dc9b529930b27081938148364520b1d0113733
95
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 slaguru666andClaude Opus 5
|
e5dc9b5299 |
Hide does not grow on membrane (R-96)
Option H, measured before it was built. Losing a wing already mattered
once it happened — a grounded supporter was 40.5% to be killed against
9.3% for one still flying, a swing of thirty points. It almost never
happened: the wings carried the creature's own six points of heraldic
hide, so folding one meant beating armour six eight times over, and it
occurred in 9% of fights. The rule was reachable, not weak, and every
option that made grounding HURT more moved the win rate by under five
points because you cannot spend a state you never reach.
So the lever is the wing, not the consequence. A location may now be
`bare`: worn armour still covers it — barding a wing works exactly as
before — but the animal's own hide stops at the edge of the membrane.
The supporter's wings are bare, and it gets the toughness back where a
six-hundred-year-old record should carry it: hide 6 -> 7, CON 20 -> 36,
so 22 hit points become 30. STR and SIZ are untouched, so the damage
modifier stays +2d6.
Measured over 20,000 fights against the same four agents:
grounded on borough off borough
before 9% 12.5% win 85.6% win / 14.4% wiped
after 67% 22.1% win 83.5% win / 16.5% wiped
The wing shot becomes the play. Fighting it on borough ground stays a
losing proposition, so ARGENT AND GULES is still the real answer and
grounding improves a bad fight rather than replacing the solution — and
the fight off borough ground keeps its teeth, marginally more lethal
than before rather than less. Bare wings ALONE was the trap: it made
grounding routine and turned the off-ground fight into a 96% walkover
with a 3.8% wipe rate, buying the text at the cost of the encounter.
check-anatomy gains the property this whole thread was about: a plan
whose behaviour turns on losing a wing must not armour that wing with
the creature's own hide, or the rule is decoration. Negative-tested by
putting the hide back.
Also re-records the lethality baseline, which was ALREADY stale: HEAD's
own code reproduced 27 of its 47 entries differently, because the
supporter joined NPCS in
|
||
|
slaguru666andClaude Opus 5
|
77465baa30 |
A flyer with a hole in a wing is a quadruped (R-95)
The supporter's tactics have said this since it was written: "It is a flyer, and a flyer with a hole in a wing is a quadruped: disable either wing and it is on the ground for the rest of the scene, which is the fight the agents can actually win." Playing a full combat against it showed the sentence was decoration. `flightLost` was computed, printed on the NPC sheet as GROUNDED, and read by nothing. Both wings disabled cost 20% off PHYSICAL rolls, and Brawl is a MELEE category, so woundPenaltyFrom returned 0 and a grounded supporter attacked at exactly the 75% and exactly the 1d6+1+2d6 it had in the air. In the played fight it was grounded from round five and went on to kill one agent, knock out a second and leave a third dying. Losing a wing now changes the animal. A plan declares what it becomes — winged says `grounded: "quadruped"` — and hitLocationRoll drops the creature onto that plan the moment a wing is disabled or destroyed. The wings leave its hit chart and the six ranged faces they covered go back to the body: vital hits rise from 3 faces in 20 to 5. Every other wound moves across on the R-94 remap with its severity intact. The wing's own damage is deliberately dropped rather than carried, and cleared in its own update so the remap never sees it. A ruined wing must not arrive as a ruined foreleg — the creature is not lamed, it is earthbound, and the injury has already been paid out as the change of shape. check-anatomy gains a fifth property: a declared landing must exist, must have no wings of its own, must not itself be a flyer, must have room for every wound the flyer was carrying without two collapsing onto one location, and must not heal a disabled location on the way down. Three induced failures confirm it fires; the third — landing somewhere with too few locations — passed at first because I had not stated the capacity property, and the guard was sharpened until it caught it. Measured, and reported honestly: this does NOT deliver the fight the tactics text promises. Over 4,000 simulated fights the win rate moves 12.7% -> 12.1% on borough ground and 85.8% -> 86.3% off it, because grounding only happens in 9% of fights there and arrives late. What actually decides this encounter is ARGENT AND GULES: 11.4% win on borough ground against 85.4% off it. The wing rule is now real and mechanical, but the creature's advice still points the GM at the wrong lever. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
4a80cc1eca |
Body plans keep what a wound MEANT, not just its number (R-94b)
1.7.1 carried wounds across a change of body plan by moving the raw points. Playing it showed that is the wrong invariant. A location is disabled at its maximum and destroyed at twice it, those maxima come from `frac`, and the plans disagree — an arm is 0.35 of the body and a foreleg 0.42 — so moving the number silently changed the meaning. Measured across the 20 plan changes: 234 severity flips, 138 of them turning a location that was merely disabled into a destroyed one. A destroyed arm arrived as a working-ish foreleg; two disabled arms stacked onto one Vesh grasping limb and arrived destroyed. So severity is now the invariant. A hurt location arrives hurt, a disabled one disabled, a destroyed one destroyed; the points are rescaled to the destination's own maximum to keep that true, and two wounds sharing one location take the worse rather than the sum, because relabelling a creature's anatomy must not destroy a limb that nothing in play destroyed. Correspondence is now stated rather than inferred. Every location gains a plan-independent `slot` (lowerR, upperL, vital, head...), and the remap matches on slot first. The previous version inferred it from `kind` plus table order, which inverted left and right on every crossing: a wounded right arm came back as the opposite hind leg. Where a destination has no matching slot the fallback prefers the same side and the same height, so a wing becomes a foreleg rather than a hind leg and a left wing becomes a LEFT arm. The vesh and the cadence are not bilateral — one grasping limb, five interchangeable units — so their slots carry no side, which is a fix to my own first pass: giving them sided slots was a category error the side guard caught. check-anatomy now proves all four promises over every ordered pair and severity band: nothing is stranded, severity is preserved, left never becomes right and fore never becomes hind between bilateral plans, and no update deletes a wound it is also setting. 432 carried wounds, 0 flips, down from 234. Five induced failures confirm each assertion fires; two of my first attempts at those did NOT fire and were sharpened until they did. Played through in a live world: a church grim filed as a humanoid with a destroyed right arm, a disabled left arm and a hurt chest becomes a dog with a destroyed right foreleg, a disabled left foreleg and hurt forequarters, and converts back to exactly the wounds it started with. Its penalties correctly stop being manipulation and start being movement — a man with two ruined arms drops his weapon, a dog with two ruined forelegs cannot run. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
a90c4f3e71 |
Carry wounds across a change of body plan (R-94)
locationDamage is keyed by location id and prepareDerivedData reads only
the keys belonging to the current plan, so changing speciesProfile on a
wounded actor stranded every wound it had. Found on a live Barghest that
had been shot in the leg and chest as a humanoid and then switched to
quadruped: 5 points under `legR` and 7 under `chest`, exactly its 12
missing hit points, sitting under keys no quadruped location answers to.
It read on the sheet as a badly hurt dog with eight pristine locations.
Stored and never read — the same defect this system keeps producing.
rules.mjs gains remapLocationDamage, which carries wounds across by KIND,
because kind is what the rules act on. Exact-kind matches are claimed
first in table order, so the obvious pairings land before any fallback
competes for them; that ordering is what makes the arms become the
forelegs going one way and the forelegs become the arms coming back.
Overflow past a location's destroyed ceiling is reported, not dropped.
It also gains locationDamageReplacement, which exists because the first
version of this fix had the bug it was fixing. Foundry merges object
updates so the outgoing keys must be deleted by name (R-93), but a
deletion and an assignment of the SAME key do not both apply — the
deletion wins. quadruped and winged share every location but the wings,
so switching between them emitted `-=hindLegR` alongside `hindLegR: 5`
and threw the wound away. Caught in live testing, not by reading.
Guards, both negative-tested before being wired in:
- check-anatomy now walks all 20 ordered plan pairs and asserts every
wound lands on a location the destination has, the arithmetic closes,
the round trip does not leak, and no update both sets and deletes the
same key.
- check-rules gains 13 spot-checks over both new rules.
Verified in a running Foundry: baseline -> quadruped -> winged ->
quadruped -> baseline conserves all 12 points with no orphans, returns
to exactly {legR: 5, chest: 7}, and whispers the GM a receipt of what
moved each time.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
|
||
|
slaguru666andClaude Opus 5
|
ddc4f99ff4 |
Hit locations reach ordinary combat, and creatures get a body
THE DEFECT. locationFor had ONE caller in the engine, hitLocationRoll, and hitLocationRoll was called from fall, burn, detonate and burstAttack. Falling, fire, explosions and automatic fire. Nothing else. rollWeaponDamage rolled dice, printed a number and returned it, so the chat Damage button, the character sheet's damage button and the item sheet's all stopped there. A grenade knew which arm it took off. A sword did not. Every individual piece was already correct - the tables, per-location armour and coverage, disable and destroy thresholds, Major Wound, dying, death, the figure on the sheet. They simply were not joined up, which is why this reads as a wiring change rather than a new subsystem. - rollWeaponDamage now carries through to hitLocationRoll: it rolls the location, applies that location's armour, applies the wound, and handles the rest through the machinery that already owned it. Nothing here re-implements any of that - locationModeFor picks the melee or ranged column from the weapon's class. The two columns have always differed and nothing was choosing between them - with no target it does exactly what it did before and SAYS SO on a card, rather than failing quietly - burstAttack opts out with locate:false - it rolls its own location per round, and letting both fire would have applied every round of a burst twice BODY PLANS. Two new tables: quadruped (four legs, fore/hindquarters, neck, a head that is hard to reach) and winged (that, plus wings). Forequarters are the vital. DISABLE EITHER WING AND IT CANNOT FLY, which is the fight a party can win. Barghest, Kelpie and Church grim were all species "baseline" - a black dog, a horse and a grave-hound, each hit-located with two arms and a chest. Fixed. Added "The supporter", a winged heraldic beast, so the new table is exercised by real content instead of existing unused. Body plan is a NEW FIELD, not a species. A bear is not a playable people with characteristic dice and talents, so creature specs take bodyPlan and the playable species list is untouched. THE NPC SHEET had no hit locations at all - 102 lines against the character sheet's 627 - while every NPC in the game had them derived and wired to consequences. It now carries the body-plan selector, the figure, the per-location table, the condition and a GROUNDED flag. The sheet class already extended the character sheet, so it already HAD the context; the template never used it. FOUND IN PASSING, by the new guard: kind "body" - the abdomen, the trunk, the hindquarters - did nothing whatever. Its printed effect has always promised "-30% to all Physical actions and bleeding 1 hit point per round until First Aid" and the code delivered none of it, on every body plan including the humanoid default. It now bleeds 1, or 2 destroyed, exactly as bodyX says. NEW GUARD: check-anatomy. Every body plan must cover 1-20 exactly once in BOTH modes, every location must be drawn and every drawn shape must be a location, and every kind must be one locationEffectsFor acts on, has effect text for, and has a destruction outcome. Verified it fails on a d20 gap, a shapeless limb and an unhandled kind before wiring it in. Ninth guard. VERIFIED IN A RUNNING FOUNDRY, not just unit-tested: melee bar reads the Melee column and the rifle reads Ranged on the same creature; forequarters graded as a vital hit; a maxed wing set flightLost and the sheet said GROUNDED; the real chat Damage button rolled 1D20 15, found the right foreleg, applied 12 through armour, disabled the leg and took the dog from 16 to 4; and with nothing targeted nothing was touched. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
748e061b56 |
CLEAN GROUND: an outline where the department authored its own monster
A v0.1 rough outline, freely adapted from A Hollow Tree (2023) — the walk north that never arrives, and a predator that imitates. Everything else comes out of the system: a 1962 reserve entry that was never cancelled, forty-one people looping eleven duplicated miles, and Borrowed Authority as a detector that can only tell filed from unfiled. Uses existing bestiary entries only. Marked not-runnable: no pregens, read-aloud, handouts, Act Four exposure model or playtest yet. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
339d5dd8dc |
test runs the guards through bun, not through npm
`test` was "npm run check", so reaching bun meant shelling out to npm first. It is "bun run check" now — one runner, and npm is out of the path. Not the eight-guard chain copied into `test`: that is the same shape of mistake the rule formulas made before rules.mjs existed, a list written twice and changed once. `check` stays the only place the guards are named. Verified: npm test exit 0, bun run test exit 0, bun run check exit 0. Worth knowing: bare `bun test` does NOT run this. It is bun's own test runner, it matches no files here, and it exits with an error. The ways in are `npm test`, `bun run test` and `bun run check` — which is what the README already tells you to use. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
54a759e057 |
Run the guards under bun, including the two new ones
Rebased onto the creature-forge work, which had added check-creatures and check-lethality to the chain and had `check` back on node. Taking either side whole would have been wrong: the node side loses bun, the bun side silently drops the two new guards. So the runner is bun and the guard list is upstream's, all eight of them. Only `check` and `readme` move. build, icons, portraits, bestiary, mj and simulate stay on node — they were never part of this change, and build writes LevelDB packs through classic-level. Verified before committing: each guard run individually under bun, then `npm run check` end to end, exit 0. check-rules 103 rules, 33 files check-kits OK check-lang 995 keys, no collisions check-templates 19 templates compile check-behaviour 67 behavioural tests check-scenarios 9 files, 62 mechanic tags check-creatures 101 actor specs across 8 sources check-lethality 46 creatures, none drifted more than 6 points Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
0c0996858f |
A GM screen taken from the rules that actually run
The figures a GM reaches for mid-scene, on one page: the order of operations, the circumstance ladder, how a roll bands, reaction, the combat round, hit locations, wounds and bleeding, both tracks, the way home, the stations, and the environment. Every number was checked by running the function rather than by reading it — 58 of 58 against rules.mjs, including the eight that moved out of ringbrp.mjs. That matters more than usual here, because `design/` still carries four rulings the code has overtaken: hit points as CON+MAS, a Major Wound at max(9, HP/3), "no hit locations, final", and marksmanship style fixed by home-world gravity. rules.mjs is the authority and the page says so at the top. It is also written to drop straight onto a Loom board: 34 cards, 23 links, nothing swallowed. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
322389bc8a |
A bestiary a GM can actually read, and the defect it found immediately
Forty-six creatures and no way to look at them. They live in content.mjs, which is source code, and in the compendium, which shows one actor at a time behind two clicks. A GM prepping a case can browse neither, so in practice the bestiary was whatever the GM happened to remember writing. docs/BESTIARY.md is generated, never hand-written, for the reason the rules journal is generated from rules.mjs: a reference that can disagree with the thing it references is worse than no reference. It opens with a one-page table sorted by how much of a party each creature puts on the floor - the part that gets printed - then an entry each with characteristics, attack, the POWER line pulled out of the tactics, and the measured figure read straight from the lethality baseline. Every number is derived, so the document cannot drift, and check-lethality already refuses the build if the measurements move. IT FOUND SOMETHING IN ITS FIRST RUN. The generic heavy natural attack is keyed `grasp` and was NAMED "Vesh grasp limb". Twenty-two creatures carry it and five are Vesh, so the at-a-glance table cheerfully printed "Vesh grasp limb" against a black dog, a kelpie, a church grim and an apple tree. Renamed to "Grasping limb"; the key is untouched, so every kit reference and every built pack still resolves. That defect was invisible for as long as nobody could see the bestiary in one place, which is the argument for the document. The caveats are given more room than the numbers, deliberately. The measured column is SOLO against ONE frozen party, and the document says so three times: encounter lethality moves with how many there are and, much harder, with who the players brought - the same fight runs 0% to 100% across party choice. It points at --spread rather than pretending a single figure is an answer. Also records the real remaining gap: 21 creatures fight with nothing but their own bodies and still carry human anatomy, so the barghest can be shot in the left arm. The machinery to fix that is built and measured; what is missing is anatomy tables for the shapes they actually are - a quadruped, an immobile object, a swarm. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
4b718595fe |
check-lethality: guard eight, and the README stops advertising a subset
Seven guards checked that content is WELL FORMED. None checked what it DOES, so a change to a damage modifier, a hit-point formula or the armour value on a service vest could double a creature's lethality without touching one line of that creature - and nothing in the build would notice, because the creature did not change. Built as a regression test rather than the hand-declared bands the creature-forge plan described. Bands are the wrong shape here: declaring "keepers: dangerous" across forty-six creatures means inventing forty-six judgements, and after --spread it is clear the interesting question is not "is this dangerous" - that has no single answer - but "is this the same as it was". A recorded baseline answers exactly that, needs no authoring, and cannot be argued with. Every creature is measured solo against a FROZEN party of four, two armed postings and two trades, at a fixed seed, so the only thing that can move a number is a change to the rules or to the creature. Tolerances are 6 points of wipe rate and 0.35 agents, which is outside the noise floor of 200 runs - a guard that cries wolf gets deleted. Verified by tampering: told the baseline the nuckelavee was harmless and the guard caught it at +16.5 points and +1.67 agents down, exit 1. Runs in half a second, so it joins the pre-build checks rather than being something to remember to run. Also: the README's guard block was a hardcoded list of four while the suite was seven. It had silently stopped mentioning every guard added after it was written, including check-creatures and check-scenarios. Now all eight, and the prose says eight. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
414dbd7037 |
simulate --spread: one number was the wrong instrument
THROUGH TRAIN Act Three publishes a single figure for its standoff - "the party is wiped in 24% of runs" - and tells the GM not to soften it. Measured against every party the duty roster can field, that fight runs from 0% to 100%. It is not that the published number is wrong. It is that no single number can be right for this fight, because the answer was settled at character selection: four armed postings walk it, four trades are massacred, and the GM reading one figure is reading somebody else's session. --spread measures the encounter against every C(16,4) party - 1820 of them, exhaustive rather than sampled, about a minute - and reports the floor, the median and the ceiling with the parties that produce them. Exhaustive on purpose: a GM planning a session wants the actual worst case, not an estimate of it. It also prints each agent's effect on the wipe rate averaged over every party they appear in, which is the line that gets used at the table. For the Act Three standoff: okonkwo -43.2 points ashcroft +16.1 sandoval -17.7 nkemdirim +13.5 holloway -15.9 ferriby +12.4 Okonkwo is worth forty-three points of wipe rate on his own. That is a scene-shaping fact about the encounter that no amount of re-measuring the median would surface. The published tables are NOT rewritten here. What to do about a 100-point spread is a design decision - constrain the party, print a range, or rebalance the fight - and it belongs to whoever wrote the scenario. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
2d4f7f18be |
Twenty-four new creatures, and the anatomies finally get used
The bestiary goes 22 to 46: eight more FOLKLORE, eight more HORROR, eight more FORWARD. Ten of the forty-six are now Vesh or Cadence, up from two. That ratio is the point. Both anatomies have had complete hit-location tables, body geometry, species talents and location effects for the whole life of this project and nothing used either, so none of it ran outside the character generator. They are now written to be WORTH the anatomy rather than to wear it: - The relay is five Cadence bodies holding a crossing open, and the crossing closes as they are destroyed. No vital, no head, 10% off everything per body lost - it cannot be dropped and cannot be ignored, which is the species argument entire. - The courier is a Vesh that fights at half rating while its one grasp limb is full. Disable the limb and it drops the consignment and fights properly. The whole encounter turns on a trade the party has to work out. - The census cannot be usefully fought at all: no head means no killing shot, so the fight lasts until somebody opens the ridge, by which time it has counted everyone. - Knockers are the only folklore Cadence, and the committee is a Cadence because a committee IS several bodies and one decision. The rest is written to the house line that most of these are arrangements rather than fights, because a bestiary where every answer is a gun has one page. The washer at the ford will not fight and cannot be made to; the apple tree man wants a drink it has been owed since 1961; the waiting room's exit was never locked and nobody stands up. Measured, 250 runs against four duty-roster agents, seed 11. Nothing accidental turned up: the non-combat entries sit at 0.0-0.1 agents hurt, the fighters at 1.4-2.3, and the only creature over a 10% solo wipe is the stanchion at 14.4% - which is a fixed object with 12 armour whose own tactics say it is a demolition problem and not a fight. The harness can only force a fight, so read that number as "the party punched the pillar". Art regenerated: 24 new portraits, and the eight new non-humans verified to draw exactly as many body parts as their location tables have entries. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
94479641b3 |
simulate: locate the damage
The harness measured a game in which every blow went to a general pool. That was
tolerable while every creature was a baseline human and merely wrong once the bestiary
had a Cadence in it: a creature whose entire mechanical identity is being whittled down
body by body was being measured as a person with an odd portrait, and the number it
printed was about nobody.
locationFor and resolveLocationHit move from ringbrp.mjs to rules.mjs, and the
category mapping inside woundPenaltyFor becomes woundPenaltyFrom beside them, for the
reason applyDifficulty and resolveBands moved before them: node cannot load the engine,
and the harness may not own a second copy of a rule. All three arrive with the
spot-checks they never had - 103 rules now, 346 formulas verified.
What the loop does now: 1d20 against the DEFENDER's own species table, melee finding
limbs and shooting finding centre of mass; damage to that location and to the general
pool, so the two agree; disabled at the location maximum and destroyed at twice it,
cumulatively. Then the consequences actually apply - a destroyed head is unconscious
immediately (conditionFor has always accepted that flag and was never passed it, so a
headshot used to leave the target swinging), a ruined arm costs -30 to every attack
including shooting, a ruined leg costs -30 to dodge, and each Cadence body lost costs
the whole creature -10 to everything.
Verified as mechanics rather than as numbers that moved: no species can return a
location it does not have across all 40 rolls; vesh and cadence have no head at any
roll and cadence has no vital either; a vesh ridge is hit 3/20 where a human head is
1/20, which is the "long target rather than a small one" the tables were designed for.
A built Cadence has five bodies at 4 points each and no vital; a built Vesh has six
locations with the ridge the largest.
Measured consequences, seed 11, 400 runs, four duty-roster agents:
keepers x6 31.0% -> 36.5% wiped
the_choir x11 0.3% -> 0.0% wiped, and now degrading as bodies go rather than
being a person who cannot be shot in the head
long_walker 42.0% wiped, which nothing had ever measured
Still no bleeding, panic, Coherence, cover, range bands or fire modes. Still a floor.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
|
||
|
slaguru666andClaude Opus 5
|
f88c31649c |
The bestiary's first non-human creatures
vesh and cadence have had complete hit-location tables, body geometry, species talents and location effects since they were written, and no creature had ever used either. So none of that had run outside the character generator, and the whole bestiary was hit-located as a person - including a black dog and a stack of paper. THE CHOIR IS A CADENCE. It always was: its tactics say "run them as ELEVEN separate bodies with one initiative between them", which is Distributed and Quorum described in prose because the species that models it had no users. Retagged, and given the two talents. No scenario prose refers to it, so nothing else moves. PRECEDENT is new, and is a Vesh - the first. Built to exercise what makes the species different rather than to be another sack of hit points: no head, so no fight against one ends on a lucky roll, and a sensory ridge for a vital that runs the length of its back and is correspondingly easy to hit. It is also not hostile, which is the point; the encounter is a conversation with something that answers questions about an event that has not happened yet. Two things left deliberately unfixed and written down rather than fudged: - The choir is eleven in the fiction and five in the location table. Reconciling it properly means giving cadence a per-creature body count, which stops the table being a constant and makes the paper doll follow it. That is the right fix and it is not a small one. A note on the spec says so. - Precedent gets unblinking but not out_of_step, though SPECIES.vesh grants both. An npc actor has no Coherence track, and out_of_step costs 1 to use. Also corrects a comment in creature-schema.mjs claiming Out of Step is vesh-locked. It is not - five postings and a roster agent offer it to humans, and locking it would take it off all six. SPECIES[x].talents is what a species is GRANTED; TALENTS[].species is what only that species may HAVE. Two different fields, and only the second locks. Simulated: Precedent 0.68 of 4 hurt over 3 rounds, no wipes. The choir alone is trivial; at the eleven its own tactics call for, 3.05 of 4 hurt, median 18 rounds, 0.3% wipes. Read both as floors - the harness does not model hit locations, so a Cadence's entire mechanical identity is exactly what it cannot yet measure. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
bbe7b92459 |
Stop tracking LevelDB's LOG; it churns on read, not on change
LOG is LevelDB's diagnostic text — "Delete type=3 #1" and compaction notes — and it is rewritten every time a pack is OPENED. Not rebuilt: opened. So all eighteen went dirty whenever anything merely read the compendia, Foundry and any verification script alike, and pack churn kept turning up in commits that had not touched any content. Verified disposable before ignoring rather than assumed: a copy of packs/npcs with LOG and LOG.old deleted opens cleanly, reads back all 21 actors, and LevelDB writes itself a fresh LOG on open. CURRENT and MANIFEST-* stay committed, deliberately. They are how a fresh clone finds the data, and ignoring them would ship compendia nobody can open. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
77e9aadc79 |
Rebuild packs; every actor now carries a portrait
76 of 76, up from 16. The npcs pack goes from every creature on the generic silhouette to all 21 with art, and each token carries the same face as its sheet. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
187a317a0a |
mj-queue: the illustrated upgrade path, without a self-bot
The Mac bridge posts /imagine to Discord with a user token. It is not ported here and should not be: Midjourney's terms since 27 May 2026 forbid automated access, there is no official API, and the bridge's own memory note records Discord silently dropping its interactions from 22 July — which is what enforcement looks like from inside. A Windows clone would be a second self-bot on the same account. So this does everything except talk to Discord, which turns out to be most of the work. `next` derives a prompt from the creature's spec and puts it on the clipboard; you paste it yourself; `file` or `watch` renames the download to the canonical name and retires the generated placeholder. One at a time, because the memory note is explicit that concurrent jobs in one channel steal each other's grids. Prompts are derived, not written — anatomy from the location table, scale from SIZ, build from STR against SIZ, plating from naturalArmour — so they cannot drift from the statblock. Three things learned by reading the data rather than assuming it: - tactics is written for a GM, not an illustrator. "Advances by bounds, 15 m per round" put a rate of fire in an art prompt. role is the one-line essence and is already the sentence you would hand an artist. Sentences carrying mechanics are dropped; dates and ages are kept. - era is the wrong test for whether something is a person. It classified Mr Bewley of the Registry and everyone in 1881 Cordera as monsters, because their kit is old. - 18 creatures are species "baseline" and are not human-shaped — a black dog, a case file that eats case files, a machine that has decided it is the site. The prompt refuses to describe a body it cannot verify, and `list` names them. That is a statblock to write, not something to paper over here. `sheet` emits a prompt sheet that the existing gen-art.mjs parses unchanged, so if the bridge ever works again the whole bestiary drives from it. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
9ddea43720 |
make-portraits: every creature gets art it cannot disagree with
Sixty of the seventy-six actors had no portrait. The sixteen that did are the hand-made duty roster; the whole bestiary and every scenario NPC fell back to a generic silhouette. The point is not that this is free. A portrait composed from anatomy.mjs cannot contradict the statblock, because it is drawn from the numbers the hit-location roll uses: silhouette from BODY_ART, height from SIZ, build from STR against SIZ, plating from naturalArmour, a mark for the primary weapon's class. A Cadence would get five bodies because its table has five. Nothing has to be kept in step. Colour carries what a GM wants off a thumbnail — whether the thing is a person, folklore, recovered machinery, or the far side. That reads the shouted category the role lines already carry (FOLKLORE, HORROR, FORWARD), which nothing was using, and falls back to structure for the ones with no tag. build-packs resolves portraits by stem and extension rather than an exact .webp filename, so a creature gains art without an edit to content.mjs and a hand-made raster always beats the generated diagram. A spec that NAMES a portrait and has none is still a hard build failure. These read as diagrammatic field-guide plates and will not match the Midjourney agent portraits. That split is deliberate: the agents are the cast, the bestiary is case material. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
24b183642f |
all-specs: one place that knows where the actors are
check-creatures collects every spec in the repo, including the rule that a scenario cast is any exported array whose entries carry characteristics — so a new scenario is covered the day it is written. That rule is a definition, and two more tools are about to need it. Lifted into tools/all-specs.mjs and imported back. check-creatures reports the same 76 specs across the same 8 sources; this is a move, not a change. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
255b146753 |
Species anatomy moves where node can read it
RINGBRP.locationTables and RINGBRP.bodyArt lived inside ringbrp.mjs, which no tool can import — it touches game, Hooks and foundry at load. So nothing outside Foundry could ask how many hit locations a Cadence has, or what shape a Vesh is, which is exactly what art generated from a statblock needs to know. Moved byte for byte to anatomy.mjs and assigned back, the same move rules.mjs got when the simulator needed the combat rules. One definition, two readers. check-rules matched those tables with a regex against ringbrp.mjs and went blind the moment they moved — it reported "could not read locationTables" and the species check silently stopped running. It imports the value now. A guard that parses source text breaks when the source is refactored; one that imports it cannot. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
0633589061 |
Rebuild packs for the posting talent pools
Roster agents carry no talents of their own; they expand from their posting at build time, so changing the pools left the compendia a version behind the register. Rebuilt with Foundry stopped. Verified by reading the built pack back: 36 packed actors, none carrying Unblinking, Distributed or Quorum. No lost/ directories, so the packs took the write without needing repair. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
b0981008b3 |
No posting may grant a species
Eighteen postings offered vesh or cadence biology to any agent who took them: unblinking nine times, quorum seven, distributed three. That is the root cause behind the six characters check-creatures caught — they were not rogue, they had taken talents their own postings advertised, and every newly generated agent would have done the same. Unblinking is the clearest case. It is cat: species, species: vesh, and its own rule text reads "A Vesh has no face to read and does not need to read yours". Nine postings offered it, including Interview, Pathology and the doctor trade. Each of the nineteen slots is replaced with a talent drawn from what the posting actually does rather than from what was nearest alphabetically: Registry gets Requisition, the Anchor gets Echo Sense, Disposal gets Clean Scene, Signals gets Chain of Custody and Dead Reckoning. No pool gained a duplicate, and no talent is left unused — every legal talent is now offered by at least two postings. The guard lives in check-kits, which is the file that validates postings, and it imports TALENT_SPECIES from creature-schema.mjs so there is one definition of which talents are locked rather than two. check-creatures was looking at characters; nothing was looking at the pools they were picked from. Verified by reintroducing unblinking on the doctor trade and watching it fail, not merely by watching it pass. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
85c1062106 |
Pause overlay: tighten the caption, lift the field grid
Both found by looking at the thing render, which nothing had done before — the earlier work on this overlay was verified through the CSSOM because the browser pane would not composite frames, so the geometry and timing were confirmed and the LOOK was not. Michroma is already a wide face, and .4em on top of it set GAME PAUSED at 176px against a 190px field: 93%, running bracket to bracket, so the readout was competing with the lock for the same width. .22em measures 154px, 81%, leaving 18px clear at each bracket. text-indent moves with it so the word stays optically centred. The field grid was drawn at opacity .3, which composited over the void band is 1.44:1 against its own ground. On a dark table map that is not a faint grid, it is no grid at all. .45 measures 1.80:1 — present as texture, still far below the 3.1:1 --rb-line is chosen for, so the amber lock stays the only thing asking for attention. Both measured in a browser with Michroma confirmed loaded, not estimated. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
8d7508077c |
Rebuild packs, and correct the guard count in the README
The talent fixes changed source without the compendia following, so the packs said Braithwaite had Unblinking while roster.mjs said Clean Scene. Rebuilt with Foundry stopped. Verified in the built pack: no agent carries a vesh talent, and Ashcroft keeps Out of Step beside Borrowed Hour and Not Quite Here, which is now legal because it is anomalous rather than species. The README's guard box regenerated with the new totals, and its prose said "five guards" in three places when there are seven. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
4ded53f9b5 |
Fix: npm run build has never worked on Windows
Both spawners built their argument with a file URL's .pathname, which on Windows is "/C:/Users/...". Node resolves that against the drive root, so the build spawned "C:\C:\Users\...\check-rules.mjs" and aborted on a module it could not find. update-readme.mjs had the same bug and swallowed it in a bare catch, so instead of failing it silently wrote a README whose guard box was empty — "0 guard lines" where there should be four. A Windows build therefore could not happen, and if it had, it would have quietly deleted the guard output from the README. Same root as the packs/ CRLF bug: this repository has only ever been built on macOS and Linux, where .pathname happens to be a usable path. Also adds check-creatures to the pre-build guards, which it was missing, so the build and npm run check now refuse the same things. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
2c547efe72 |
simulate: a lethality harness that uses the real rules
THROUGH TRAIN publishes "MEASURED, 400 runs each" and a 24% wipe rate; the starter publishes its own table. Nothing in this repository could produce either number, so both were claims rather than results and neither could be re-checked after a rule moved. The harness knows no rules of its own. Everything comes out of rules.mjs, because a harness with its own copy of the combat loop measures a game nobody is playing and drifts silently the first time a rule changes. Two things had to move for that to be possible: applyDifficulty, resolveBands and gradeRoll -> rules.mjs. The whole of d100 resolution lived in ringbrp.mjs, which node cannot import. All three are pure, so they moved and the engine re-exports them; no caller changes. check-rules then refused them for having no spot-check, so they now have fifteen, covering the 1% floor, the 96-99 rule and 00-always-fumbles. They had none in their entire life inside the engine. expandFromRegister -> tools/expand-spec.mjs. A roster agent is written as a job and a rank and carries no skills or kit; everything it can do is derived at build time by a function locked inside build-packs.mjs, which cannot be imported because importing it runs the build. A simulated agent expanded by a second copy of that logic is not the agent that gets packed. Moved verbatim; both sides import one copy. Deterministic throughout: same creature, party and seed, same numbers, which is what lets a published table be verified rather than remembered. What it finds: for the Act Three standoff the seed barely matters (0.5-2.8% wipe across five seeds, eight rounds throughout) and party composition dominates everything. Four support agents against the six miners wipe 92.8% of the time; swap in the heavy and the warden and it is 0.3%. The published 24% is not a property of the encounter, and no single figure can be. Models Reaction re-rolled each round, banding, graded defences, the stacking defence penalty, damage by band, major wounds and the dying clock. Does NOT yet model hit locations, bleeding, panic, cover, range bands or fire modes, all of which make a losing fight worse. Read the output as a floor, never a ceiling. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
c3002e93c9 |
check-creatures: validate every actor spec, and fix the vesh talents it found
Nothing validated creature specs. check-kits validates postings and check-scenarios validates prose; buildActor throws on an unknown skill or item key and is silent about everything else, so species, marksmanship style, characteristic blocks, species-locked talents and misspelt field names could all be wrong and still build, pack and play. Every packed actor once shipped as species "hominid" and the whole cast was hit-located as something the rules do not contain. tools/creature-schema.mjs derives every allowed value from the catalogues rather than retyping them: skills from SKILL_CATALOGUE, kit from the item catalogues, species from SPECIES, stats from CHARACTERISTIC_DICE, and the two marksmanship styles out of the language file. It reports every problem with a spec rather than dying on the first, and names the nearest real key. It also reports fields buildActor never reads, because "armours:" is not an error, it is an actor that silently equips nothing. On its first run it found six baseline humans carrying vesh biology. "unblinking" is cat: species, species: vesh, and its own rule text reads "A Vesh has no face to read". Out of Step is reclassified as anomalous with a Coherence cost, which is what every talent in that category has and what the Anchor who is three years older than his birthday says always wanted to be; the other five are swapped to agency and field talents that fit them. Not fixed here: the root cause is upstream in postings.mjs, where twelve posting talent pools still offer unblinking, distributed or quorum to any agent. Picking replacements across those is a design decision, not a mechanical one. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
10f588b666 |
Plan: the creature forge
Five parts - schema and validator, lethality harness, forge CLI, procedural portraits, Foundry importer - built on three findings: the species machinery (vesh, cadence) is fully implemented and completely unused, the published measured lethality tables cannot be reproduced by anything in the repo, and RINGBRP.figures already holds the per-species geometry procedural portraits would need. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
c6924d6ea8 |
THROUGH TRAIN: art prompt sheet, and a runner for it
Prompts for all 43 assets - 16 duty-roster portraits, 14 NPC portraits from TT_CAST, and the 13 scene plates, none of which had a recorded prompt, so none of them could be regenerated. Art direction is frontier newspaper engraving, and the present-day Bureau material is drawn in the same 1881 hand as Cordera with modern people composed as tintype portraits, so the two halves read as one document. Three rules are written into the sheet because they protect meaning rather than style: nothing modern is drawn on the 1881 side, the train is absence rather than a glowing spectre, and nobody is lit as a villain. tools/gen-art.mjs parses the sheet rather than keeping a second copy of the prompts, expands the style tokens and drives mj-gen. Dry run by default, because --run fires paid generations. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
b31b1028a9 |
Rebuild the pause overlay as a containment hold
The uniselector rotated forever, which is the one thing a hunting stepping switch never does, and it was carrying two outright bugs: the motion trail pointed the way the arm was going rather than the way it came, because conic angles run clockwise from the origin and the tip was declared first; and the hub pip ran on a free 2.6s sine against a 7.7s arm, so the light drifted permanently out of phase with the mechanism and meant nothing. Replaced with an instrument that acquires, locks, reports and lets go: a static field, four corner brackets that close and snap, a scan line reading down the field at exactly cycle/3 so its third pass always lands on the lock, and a readout welded to the same cycle. Every layer reads --rb-pause-cycle, so nothing can drift. The field offset is derived from the column rather than measured and pasted in. Also ignores the impeccable hook cache and keeps its config, which records why broken-image is suppressed for a stylesheet. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
2332b449ba |
Refresh package-lock
The committed lockfile predated the handlebars devDependency and carried the wrong name and licence metadata. Regenerated by npm install; no dependency actually changes. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
f0edd54e83 |
Mark packs/ binary so Windows checkouts stop corrupting the compendia
core.autocrlf=true rewrites packs/*/CURRENT from "MANIFEST-000002\n" to "MANIFEST-000002\r\n" on checkout. LevelDB then tries to open a manifest whose filename ends in a carriage return, fails with "The filename, directory name, or volume label syntax is incorrect", and Foundry runs RepairDB across all eighteen packs on first launch. The packs are therefore broken from the moment they are cloned, and the only visible symptom is a pile of lost/ directories in git status. Linux and macOS checkouts are unaffected, which is why this went unnoticed. The committed blobs are already LF, so no renormalisation is needed: this only affects the smudge side. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
214e207cc6 |
THROUGH TRAIN: north bench plate without the park bench
Midjourney read "bench" as furniture in the first version. Regenerated with the word removed and "no furniture" stated: two coated figures from behind, empty ground, a dry gully, the town's water tank on the far horizon. Emptiness is the subject, which is what the scene is for - they arrived where the paperwork says the railway is. All thirteen plates are real art and there are no placeholders left. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
8a4d093a13 |
THROUGH TRAIN: the Bureau plate, and no placeholders left
tt_11_bureau regenerated after the first attempt timed out, converted to the house plate format. All thirteen plates are now real art. A version of tt_12_northbench without the word "bench" in the prompt is still rendering; the current one is good apart from the park bench Midjourney put in the desert. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
67b6f61c64 |
THROUGH TRAIN: two of the three new plates are real art now
tt_12_northbench and tt_13_standoff replace the placeholder copies. Converted to the house plate format, 1024x682 webp, to match the other ten. The Bureau plate timed out in Midjourney at 1800s waiting for a grid and is still a placeholder; it is regenerating. Known flaw in tt_12_northbench: Midjourney read "bench" as furniture and put a park bench in the Nevada scrub. Everything else about the plate is right - two coated figures, empty ground, no railway, the town smoking on the horizon - so it ships while a version without the word regenerates. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
973b4a53aa |
THROUGH TRAIN rewritten into five acts
Act One is now an induction at the Bureau, Act Two puts the players through the crossing and makes them find the train, Act Three is a town-side obstacle, Act Four is the boarding, Act Five is the four minutes. 4h00 in five acts, was 3h30 in three. The restructure earns one thing the old shape could not: THE CROSSING OPENS WHERE THE RECORD SAYS THE LINE IS, which is an empty bench of scrub on the wrong side of the wash. The agents arrive where the paperwork put the railway, not where the railway is, so the fraud is the ground under their feet and finding the train is real work. Registry's file has this wrong and Bewley believes the file. - Act Three: Rance Doyle and five unpaid miners take the agents for company detectives, and they are not wrong to. Resolves on what the players SAY - no roll makes it go away, no roll makes it a fight. Losing costs them the 1881 issue, not their lives. His brother Tom is aboard and dies at minute two of every pass, which is the wiring between the acts - new cast: Bewley, Prosser, Rance Doyle, Shard, Tom Doyle, Cordera man - new handouts: the company notice, and the six names in Ambrose's hand - 8 GM pages, 10 handouts, 15 actors, 13 plates MEASURED, 400 runs each: the Act Three standoff is 0.4/4 hurt from cover and 1.25/4 in the open, both ending early - but FOUGHT TO A FINISH it puts 1.75 of 4 down and wipes the party in 24% of runs. The early stop is what makes the act survivable, not the miners being weak. My first draft of that table was guessed and wrong in all three rows. FOUND AND FIXED: there is no Ride skill in this game and never has been. The boarding set-piece has asked for "[CUS: Ride, then Athletics]" since it was written, through a playtest, which is an instruction to roll a thing nobody has. Horses are now handled by a local and the agent rolls Athletics. NEW GUARD: check-scenarios resolves every [CUS: Skill] tag in every scenario against SKILL_CATALOGUE and fails the build on a skill that does not exist. Verified it fails on the real bug before wiring it in. Sixth guard. NOTE: the three new plates (Bureau, north bench, standoff) are placeholders in this commit - Midjourney is still rendering them. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
24b1aaa682 |
Development: skills improve from what beat you, not what worked
Reversed experience. A skill is marked when it FAILS - failure or fumble - and never when it works, and the improvement roll is the same reversal: roll OVER your rating and the skill goes up 1d6. Nobody learns anything from the lock that opened first time. The two halves pull against each other on purpose, so ratings converge instead of running away. Simulated over 2000 campaigns: a 10% and an 85% skill start 75 points apart and are 11 apart after 60 cases. - rules.mjs owns it: DEVELOPMENT, developmentMarksOn, developmentImproves, developmentCeiling, developmentGainFor - trainingCap has been on the skill schema and the item sheet since it was written and nothing has ever read it. Development reads it now: a skill stops at its own cap, and everything else stops at DEVELOPMENT.ceiling, because a d6 gain could otherwise push a 98% skill past 100 - GM opens the phase from a GM-only toggle in the token controls; it posts a card naming every agent and their marked count, and puts a Development button on each sheet. Players press their own - the toggle re-reads the setting on change, since getSceneControlButtons only fires at load and the button would otherwise lie if the phase were opened by the API or a second GM - rules journal and player primer rewritten from the tables; the old text described the pre-reversal rule on both halves - 7 spot-checks, 4 behaviour tests; verified end to end in a running Foundry Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
99e78f440a |
"Somebody you knew" costs one point, not two
At two, the closing beat of a hard case broke four agents in five at the same moment. A whole party freezing or bolting on the final scene reads as arithmetic rather than grief, which is the opposite of what the trigger is for. Found by playing THROUGH TRAIN end to end. Replayed the same beat from the same party state — Holloway 3, Okonkwo 2, Bhattacharya 4, Nkemdirim 3, Finch 2 — with the new cost: one agent breaks, the rest carry it at 3 and 4 with the penalties that implies. That is the shape it wanted. "It said your name" stays at two, and is now the only two-point trigger in the list, which is right — it should be the singular worst thing that can happen to somebody. The rules journal and the player primer both generate their trigger tables from PANIC_TRIGGERS, so both already print the new number. Version 1.4.2. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>v1.4.2 |
||
|
slaguru666andClaude Opus 5
|
c05d488b9d |
Play THROUGH TRAIN end to end; fix three scenes the dice could not reach
Ran the whole case with five agents — briefing, kit swap, Mercy Draw, the
crossing, Cordera, the water tower, the four minutes, the locomotive, the
brakes, the register. It completed, and it exposed one serious defect and two
smaller ones. All three are the same shape: the scenario put dice in front of
things the fiction says should simply happen.
1. THE ENDING WAS UNREACHABLE. Convincing Kell to stop was gated on Persuade
or Status. Only THREE of the ten duty-roster agents have either trained —
the Field Lead has Persuade 15 and Status 15. Three separate attempts on
him all failed and the scenario had no way to end; I narrated it as working
because otherwise there was no ending, which is exactly the improvisation a
scenario is supposed to prevent.
It now takes NO ROLL if the party can tell Kell one true thing about his own
train — the coal, the four minutes, the north bench. He has waited a
hundred and forty-five years for somebody who knows what is wrong to say
so. Only a party that knows nothing rolls, at average, and a failure costs
a pass rather than the ending.
2. THE PAPERS NEEDED PERSUADE. Showing one side the other's document is the
move the scenario is proudest of, and it sat behind Persuade at −30%:
15% became 1%. Both men WANT their paper read — Merritt is proud of his
letter, Ambrose's whole grievance is that nobody has looked at his writ.
No roll, between passes.
3. THE BOARDING SORTED ITSELF BADLY. Athletics runs 25–60% across the roster.
Putting all five on the water tower left the doctor and the Registry officer
missing twice each, carried across on the department's plank, passengers in
their own set-piece, two passes gone. The text now names which route suits
whom before they choose.
What worked, and is worth recording: the automatic clues absorbed two failed
rolls in Act One without the trail stalling; Standing Order 9 caught Okonkwo
checking a wristwatch in an 1881 street and cost a Coherence and a clock tick;
the fumble-is-a-fall rule fired on a 100 at the tower for five damage; the
exposure table held — one hit across the whole fight, on the agent who left
cover to work on Doyle; and the reset landed exactly as intended, with Doyle
standing up at the top of the car and Nkemdirim's hit staying where it was.
Version 1.4.1.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
v1.4.1
|
||
|
slaguru666andClaude Opus 5
|
391c6a3679 |
Rebuild packs cleanly
LevelDB LOG timestamps only — no content change. Committed so the tree matches a build made with Foundry stopped, which is the state the published archive should carry. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
slaguru666andClaude Opus 5
|
5bc46879c7 |
THROUGH TRAIN: put the gunfight inside the loop
The loop is no longer just a train going round. It is FOUR MINUTES with a
gunfight in it, and those four minutes have run about two million times.
Two sides, and neither is the villain. Cass Merritt and three hired men have a
signed letter of engagement from the company telling them to surrender the
payroll to nobody "upon any pretext whatsoever, including the presentation of
any paper". Josiah Ambrose and four Cordera miners have a sealed writ of
attachment from a territorial judge naming that same payroll against eleven
weeks of unpaid wages. Both documents are genuine. Neither man has seen the
other's. The company sold the same money twice and is not aboard. Both are
handouts now, and they only mean anything held in the same hand.
The mission is restated everywhere it appears — briefing page, case brief, GM
essentials: CLOSE IT. Closing the crossing does not kill anybody, it lets them
go. The loop ends, 3 November 1881 resumes from 11:44, and everyone aboard walks
forward into the life they would have had — including Ambrose and a miner named
Doyle, who did not survive the real four minutes. The agents are not choosing
who lives; they are choosing to stop holding history still, having met the
people it lands on.
The standing order is louder in three places: NOTHING MODERN GOES THROUGH.
Two things the playtest found, both now written into the text:
- THE FIGHT WAS A TPK GENERATOR. Modelled with all nine shooters firing at the
party it wiped them 171 times in 200. The model was wrong — the two sides
are shooting at EACH OTHER — so the scenario now carries a measured exposure
table: head down draws nothing, acting from cover costs half a party member
and no wipes over 300 runs, moving in the open costs one, and going for the
payroll box wipes more than half of all parties. The GM is told to say that
last row out loud BEFORE anybody commits to it.
- THE LOOP RESETS AND THE AGENTS DO NOT. Everyone who belongs to 1881 is back
up unhurt on the next pass. The agents keep their wounds. Nothing said so,
and a player who assumes otherwise takes risks that kill them. It is now
stated in capitals in both documents, with the played pass that prompted it
— two of four down at 0 hit points, one untouched because he stayed on the
floor.
Nine cast members, eight handouts, ten plates.
Version 1.4.0.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
v1.4.0
|
||
|
slaguru666andClaude Opus 5
|
774e95dea8 |
THROUGH TRAIN — the crossing starter, and two more playtest fixes
A train has been running the same four miles of track since 3 November 1881. It
has never stopped. Sixty-one people are aboard and all of them are alive.
The supernatural is science fiction underneath, which is the point. A railway
company laid track to a filed survey instead of to the ground, then described the
line as being where it was not — ten thousand times, in timetables and waybills
and telegraph orders, for six months, at a constant speed. Reality keeps a record
of what belongs where, and that is the most patient argument there is. The record
gave way with No. 9 inside it, and the RUNNING is what holds the crossing open: a
standing wave sustained by its own motion the way a bow sustains a note.
So the players must board a train that cannot stop, and then find that sixty-one
people at six passes is four hours they do not have. The ending is what they
choose to do about arithmetic they cannot beat, in front of a conductor holding
the manifest. No roll resolves it.
It is also the case that teaches Standing Order 9: NOTHING MODERN GOES THROUGH.
Using a modern object across costs a Coherence and moves the clock; carrying it
costs nothing; anchors are exempt. The catalogue already had the 1881 issue —
Webley, Martini-Henry, oil lantern, brass compass, pocket watch, buff coat,
laudanum — so the constraint needed no new content and gives the table something
to feel in the first twenty minutes.
Ships as docs/scenarios/THROUGH_TRAIN.md and a Foundry Adventure: 7 GM pages, 6
handouts including the survey that started it and a manifest deliberately held
back to Act Three, the case file at Containment 6 / Resources 5 / Clearance 4
with its four legs loaded, four of the people aboard, and 7 plates to show.
Two fixes from playing it:
- Bhattacharya's RESEARCH was 1%. Placing the new skills last night I swapped
Occult into Registry's core and took Research out — a records office that
cannot look anything up. Research is back in core; Antiquities gets its
Knowledge back the same way.
- The boarding was unrunnable. At halved skill the set-piece put four agents on
12–31% and all four missed, which stalls the one scene the scenario is built
around. It is an average roll now and only a fumble is a fall, with the
reason written into the text so nobody "fixes" it back. Re-run: 3 of 4 aboard
on the first pass and one waiting for the next, which is the shape it wanted.
Version 1.3.0.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
v1.3.0
|
||
|
slaguru666andClaude Opus 5
|
37b4043746 |
Play a full case, then fix what it found
Ran THE HOLLOW ROAD with five agents — approach, crossing, contact, the thing on
the road, extraction, debrief — with panic and Coherence both live. It ran clean
and it found one thing that mattered.
ANOMALY LORE 0. Marta Holloway, Field Lead, Senior Field Officer, could not roll
the department's signature skill. Nor could Okonkwo. Induction was gated to
TRADES on the reasoning that a career posting is itself the agency training —
but 17 of the 25 postings do not list anomaly lore, its printed base is 0 on
purpose, and nothing else supplies it. So the people who joined the department
straight out of school were the only ones who never received its basic training,
which is exactly backwards.
Induction is a FLOOR for everybody now, posting or trade, in the live generator
and the pack builder both. It never demotes: Bhattacharya keeps Registry's 63,
Holloway goes 0 → 43, Okonkwo 0 → 33. All twenty built agents verified non-zero.
Two GM tools the case showed were missing, both found by feeling the friction
rather than by reading anything:
- clearPanic(actors). Panic clears when the scene does — one line in the book
and, until now, five sheets edited by hand while the table waited. It leaves
Coherence alone, which is the whole point of having two tracks.
- partyState(). Everybody's hit points, panic and Coherence in one card. With
five agents the alternative is asking "what are you on?" five times a scene,
which is friction a table absorbs silently and then blames on the game.
And the player-facing gap: the primer never mentioned panic. Players got a
one-page aide-memoire for a track they now meet in the first scene. It has a
page of its own and a row on the cheat sheet, both generated from the same
tables as the rules.
One thing I reported as a bug and withdraw: Okonkwo's "Firearm (Pistol) 5%" was
me rolling the wrong weapon. He carries a marksman rifle at 53%. check-kits was
right.
Version 1.2.0.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
v1.2.0
|
||
|
slaguru666andClaude Opus 5
|
7ec7efd29b |
Replay the scene with the fumble rule; expose panicReliefFor
Ran the east corridor again with fresh dice and every relief attempt routed
through easePanic, fumbles included.
The rule did not fire this time. All three attempts failed cleanly — Psychology
58 against 40, and two POW×5 rolls at 66 and 76 — so nobody was punished for
trying. That is worth reporting as loudly as a hit would have been: two
play-throughs in and a fumbled relief has come up once, which is about right for
something meant to be a memorable disaster rather than a tax.
Braithwaite still broke at 5 on "it said your name" and fled. Marrow went down
in a single round to a special with the entry bar — 19 through armour — against
the 4.5-round average the thirty-run tuning gave, so that was dice, not balance.
Then the question the rule actually raises. Relief rolls are made at your own
panic penalty, so the worse you are the worse your odds AND the wider your fumble
range — and now a fumble adds a point. That is a feedback loop, so I measured it
rather than trusting it, 300 runs per row:
from 2, POW×5 60% 280 calm · 0 broke
from 3, POW×5 60% 227 calm · 2 broke
from 4, POW×5 60% 138 calm · 31 broke
from 4, POW×5 50% 96 calm · 35 broke
from 4, POW×5 85% 267 calm · 5 broke
Recovery dominates everywhere. The worst case — overwhelmed, weak POW, no help —
spirals into breaking about one time in eight, which is a bad night rather than a
trap. No change needed.
Also: panicReliefFor and panicAfter were missing from the API, which I only
noticed because measuring the loop needed them and the alternative was
re-implementing the rule in the test — the exact duplication check-rules exists
to forbid. Both exposed.
Version 1.1.3.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
v1.1.3
|
||
|
slaguru666andClaude Opus 5
|
5aaae186e0 |
A fumbled relief attempt puts a point on
Trying to steady somebody and making it worse is the most familiar thing on that list, and until now it was the only outcome that cost nothing. Done as a NEGATIVE relief rather than a second code path: panicReliefFor returns -1 on a fumble, and easePanic already subtracts what it is given, so subtracting a negative raises the track. One number, one direction, no branch that could drift from the other. The clamp still holds — fumbling at 5 stays at 5. The chat card had to be told, because otherwise it prints the word "Psychology" over a number that has just gone UP and the table reads that as a bug rather than as the rule. It now leads with a red FUMBLED tag. Guarded three ways: a spot-check in check-rules for the -1 itself, and two behaviour tests — one that the sign survives easePanic's subtraction and actually raises the track, one that it cannot push past the top. A test on the constant alone would have passed even if easePanic had healed on a fumble. Verified live: 2 → fumble → 3 → success → 2 → special → 0, and a fumble at 5 staying at 5, with the cards reading correctly. Version 1.1.2. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>v1.1.2 |
||
|
slaguru666andClaude Opus 5
|
20306d83d9 |
Play a scene with panic; tighten one reference row
Ran LAST ADMISSION's east corridor with the six civilians and real dice. The
track works end to end and, more importantly, the penalty REACHES things:
Ferriby at panic 3 swung at Old Marrow with Brawl 25% printed and 15% effective,
while Sharrock at panic 1 rolled his full 58%. That is the whole point of the
track and it is the part that is usually decoration.
The arc played well. Lights out took four of them to 1. The body took Braithwaite
and Ferriby to 2. Enderby counting took three of them again. Braithwaite fumbled
his Psychology roll trying to talk Colin down, then it said his name — 5, broke,
fled, came back at 3. Renshaw steadied herself on POW×5 and dropped to 1.
One thing I got wrong and one thing worth knowing:
- I flagged a 99 grading as a fumble as a bug. It is not. A 58% skill fumbles
on 98-00, and both the primer and the rules page say so in those words. The
"96-99 never succeed / 00 always fumbles" pair are floor rules, not the whole
fumble range. Only the quick-reference row was ambiguous — it now says
"never succeeds ... and is a fumble if it is inside your fumble range".
- The civilians cannot use two of the three new skills. Occult, Religion and
Psychology sit at 5% on the "(before)" sheets, so in the prologue the only
relief that works is POW×5 at 50-85%. That is correct — a site foreman is
not an occultist — but it means the skills are an AGENT'S tools, and the
prologue is deliberately the game without them. The agents have them where
they should: Bhattacharya Occult 63, Lindqvist 58, Renshaw Psychology 40,
Braithwaite Religion 35.
Version 1.1.1.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
v1.1.1
|
||
|
slaguru666andClaude Opus 5
|
cbad87ba98 |
Add Occult, Religion and Psychology, and the Panic track
From the first game's list. The three skills are not decoration — they are the
levers on the new track, which is why they arrived together.
PANIC, 0 to 5, up. It is the other half of Coherence and the rules page says so
in the first line, because a table that confuses them will make the wrong call
at the worst moment:
Coherence what knowing has cost you weeks 10-0, down
Panic what is happening to you NOW minutes 0-5, up
A trigger gives a point with NO ROLL. Asking for a die to find out whether you
are frightened slows the scene and tells the player how to feel; the die is for
getting rid of it, which is the interesting direction.
0 steady · 1 rattled · 2 shaken -5% · 3 losing it -10%
4 overwhelmed -20% · 5 broken
The first point is free on purpose, so the opening shock of a scene costs the
table nothing. At 5 you break — freeze, flee or fixate, choose fast — and come
back at 3, not at 0, because breaking is not a rest.
Relief is where the skills come in. A success takes one point, a special takes
two. Psychology is the only thing that steadies somebody who is NOT you.
Religion steadies you alone and does not care whether you are right. Occult is
the odd one: naming a thing correctly, out loud, helps everybody who can hear —
the department's actual work, and the only relief that reaches a room. Plus
POW×5 to steady yourself, and your anchor object once per case, which is what
that briefing scene was for.
The three skills are placed in six registers rather than left floating: Occult
to the professor and to Registry, Psychology to the doctor and to Interview,
Religion to Antiquities and Pathology.
Guards, because a track nothing reads is this project's oldest failure: the five
new rules carry spot-checks in check-rules, and check-behaviour gains nine tests
— the free first point, monotonic penalties, only 5 breaking, clamping at both
ends, breaking not clearing the track, special taking two, every band/trigger/
relief having a label, and every relief naming a skill the catalogue actually
has.
Verified in a live game: penalty reaches the roll (58 → 58 → 53 → 48 → 38 for
panic 0-4, criticals and specials recomputed with it), the full arc runs 0 → 2 →
4 → eased to 2 → pushed to 5 → breaks → 3 with no errors, the sheet shows the
band, and the rules journal has a Panic page generated from the same tables.
Version 1.1.0.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
v1.1.0
|
||
|
slaguru666andClaude Opus 5
|
ab619f4b73 |
Remove the dice textures and the whole Dice So Nice path
Deleted: the seven PNGs in dice/, tools/make-dice-textures.mjs, the `npm run dice` script, the RB_DICE colourset and texture tables, diceSetFor, the diceSoNiceReady hook, and rbDressDice — which was tagging ten roll sites with coloursets that no longer existed. 99 lines of module, plus the assets. README and DEPLOY updated; they still advertised all of it. Recoverable with `git revert` if it is ever wanted back. This also explains the 01-that-was-12. The RESULT was never wrong: skillRoll rolls one 1d100 and grades on roll.total, and the chat card derives the tens and units it prints from that same total. Checked all 100 values of percentileFaces and 60 live rolls — the printed faces reconstruct the graded number every time, with no mismatches. What disagreed was the ANIMATION. The custom addDicePreset labels registered for d100 and d10 remapped which physical face means what, so Dice So Nice could land showing 01 while the roll behind it was 12. Nothing downstream ever read the animated faces, so no roll was ever adjudicated on the wrong number — but a table watching the dice had every reason to believe otherwise, which is its own kind of wrong. Disabling the skins in 1.0.6 already removed it; deleting them makes it permanent. Verified in a running Foundry after removal: skill roll, weapon damage, hit location, major wound and Coherence all resolve and post their cards, no errors, and game.ringbrp.diceSetFor is gone from the API. Version 1.0.7. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>v1.0.7 |
||
|
slaguru666andClaude Opus 5
|
9b6c61ffcb |
Turn the custom Dice So Nice skins off
Registering them with "preferred" made them the default for everybody in the world, which overrides whatever dice a player had already chosen — the system helping itself to a setting that belongs to the user. Off at the author's request. One named constant rather than a settings API: this codebase registers no game settings at all, and adding that surface, its lang keys and its UI to switch one feature off is more machinery than the job needs. Flip RB_DICE_ENABLED to true and everything below it still works. The textures in dice/ and the generator behind `npm run dice` are left alone. Nothing reads them while this is false, and deleting work in order to turn a feature off is a poor trade. Verified in a running Foundry with Dice So Nice active: "custodians" is no longer among the registered systems, and preferredSystem has gone back to "standard" rather than ours. Version 1.0.6. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>v1.0.6 |