104 lines
6.6 KiB
Markdown
104 lines
6.6 KiB
Markdown
# Leader Skill battle effects
|
|
|
|
13 September 2026 checkpoint. **Prototyped, with Partial rules coverage.**
|
|
This dated follow-up connected both Rihani bands and the lower battle bands of
|
|
Mentat, Bureaucrat and Sandmaster. Its earlier absence claims for normal Mentat,
|
|
payments, movement collection, Banker, Diplomat and Smuggler are superseded:
|
|
the [26 October inventory](LEADER_SKILLS_RUNTIME.md#independent-band-inventory--26-october-2026)
|
|
records connected first versions of all28 independent bands. Full interactions,
|
|
mode acceptance and publication remain gated.
|
|
|
|
## Source contract
|
|
|
|
The [GF9 CHOAM & Richese rulebook, page 9](https://www.gf9games.com/dune/wp-content/uploads/2021/11/CHOAM-Rulebook-low-res.pdf)
|
|
provides the common skill procedure. The printed cards and publisher designer
|
|
walkthrough supply the individual effects; see the [component inventory](LEADER_SKILLS_RULES.md).
|
|
|
|
- [Mentat, 17:37](https://www.youtube.com/watch?v=XT_azRVLq_0&t=1057s): the
|
|
surviving selected skilled disc gains two battle strength. Its printed value,
|
|
bounty and Zoal source value stay unchanged.
|
|
- [Bureaucrat, 15:01](https://www.youtube.com/watch?v=XT_azRVLq_0&t=901s): the
|
|
surviving selected skilled disc subtracts one from its opponent's total per
|
|
distinct stronghold that opponent occupies. Contested and storm-covered
|
|
occupation count; advisors and virtual victory control do not. There is no
|
|
implicit zero clamp. The shared board helper recognizes HMS and Jacurutu,
|
|
though combined-module play remains gated.
|
|
- [Sandmaster, 15:48](https://www.youtube.com/watch?v=XT_azRVLq_0&t=948s): a
|
|
surviving skilled winner adds three board spice to an existing pile in the
|
|
battle territory. This grants no direct income and does not require surviving
|
|
forces. No pile means no addition; a single positive pile resolves automatically.
|
|
With several positive piles, the winning owner must select exactly one existing
|
|
pile after survival is established and before physical battle settlement. Each
|
|
option shows its current quantity and the exact +3 result; this mandatory effect
|
|
has no decline or extra acknowledgement. **Provisional allocation policy:**
|
|
owner selection is a prototype choice, not a publisher sector-placement ruling.
|
|
It cannot create a new sector pile. Smuggler collection precedes placement, so
|
|
exhausted piles are not offered.
|
|
- [Rihani, 12:49](https://www.youtube.com/watch?v=XT_azRVLq_0&t=769s): a face-up
|
|
native trainer grants the normal private inspection after a victory using
|
|
another leader. A surviving selected native skilled winner receives both
|
|
bands. A surviving captured skilled winner receives only the lower band.
|
|
|
|
Rihani first privately inspects two random physical Traitor Deck cards and
|
|
reshuffles them. It then offers the separate optional two-card draw. Choosing
|
|
to draw commits the player to keep one newly drawn card, publicly reveal and
|
|
return one unused old card, and secretly return the unkept new card. Those two
|
|
returns are shuffled into the actual deck. There is no post-draw decline or
|
|
return-both-new alternative. The two samples can contain the same identity.
|
|
|
|
**Ordering inference:** normal-before-lower follows the printed order and
|
|
designer narration; no explicit priority sentence was found. **Unused** means
|
|
not actually called, rather than merely known or publicly disclosed. Returning
|
|
an old card announces its name without adding it to the actual-use register.
|
|
A Traitor Card used to win this battle is already excluded from the exchange.
|
|
No eligible old card means no optional draw. There is no partial draw fallback
|
|
when fewer than two physical deck cards are available.
|
|
|
|
The [Ixian/Tleilaxu rules](https://www.gf9games.com/dunegame/wp-content/uploads/2020/09/IxianAndTleilaxuRulebook.pdf)
|
|
place skill resolution before faction capture and Face Dance. Tleilaxu Rihani
|
|
must instead exchange an unrevealed Face Dancer. The [Basic Tleilaxu follow-up](TLEILAXU_LEADER_SKILLS.md)
|
|
connects that exchange and the skill return when Face Dance kills a leader.
|
|
Foreign gholas remain gated.
|
|
The connected entry supports base factions in Basic/Advanced with Leader Skills
|
|
alone. These boundaries are implementation limits, not prohibitions in the game.
|
|
|
|
## Connected behavior and checks
|
|
|
|
Shared battle quotes calculate score changes; bots use the same public
|
|
stronghold count. Sandmaster resolves automatically for zero or one remaining
|
|
positive pile; multiple piles use an owned, event-bound placement decision with
|
|
one direct button per existing sector. Only supplied public pile quantities appear
|
|
in the controls. Rihani offers owned draw/decline controls, then separate inspectable
|
|
keep-new and reveal-old selections. All four AI profiles use only their permitted
|
|
view. Only the owner receives inspection history and drawn identities. The public
|
|
log names the returned old card, never the kept or unkept new identity.
|
|
|
|
Physical snapshots, an event-bound obligation and saved stages preserve Rihani
|
|
through casualties, winner cleanup and captive return. Independent review found
|
|
that completed receipts were not bound to custody during cleanup; this now
|
|
rejects changed deck order after either decline or exchange. JSON and SQLite
|
|
restart tests cover both choices, concurrent writes, stale replays, normal versus
|
|
captured eligibility, actual-use exclusion and native Suk rescue preceding a
|
|
captured Rihani exchange. Focused checks also cover score-boundary winners,
|
|
dead/bluffed skills, Sandmaster without surviving forces and UI privacy.
|
|
|
|
Pure Sandmaster boundary tests cover exact +3 offers, independent sector selection,
|
|
empty and singleton behavior, malformed locations and quantities, overflow,
|
|
unoffered keys and input immutability. Isolated Chromium at 390px exercised
|
|
the production placement control after Smuggler allocation, selected sector
|
|
15 and observed its actual +3 change before the original battle continued.
|
|
|
|
Fresh browser room `HNGSR8GB` used genuine module setup, then a separately
|
|
staged conserved battle. The human inspected two private cards, drew two
|
|
separate cards, refreshed the pending exchange and selected a keep and public
|
|
return. Completion and refresh restored the kept Traitor and winner card cleanup.
|
|
This is targeted browser evidence, not a naturally reached complete game.
|
|
Two genuine four-profile Basic/Advanced samples completed in 92/188 actions,
|
|
with 3/6 JSON continuations and no rejected candidates; neither reached a Rihani
|
|
effect. They establish sample continuation, not comprehensive skill acceptance.
|
|
|
|
The running server responded after the reported outage. A fresh online backup
|
|
captured all 393 opening rooms. No restart or reset was needed. Required broad
|
|
checks and final saved-game preservation are recorded in the checkpoint commit
|
|
and private source-bound report.
|