Apply user priority: save/recovery/privacy/custody assurance after all rules. Add quick types/lint, concise failure-preserving reporter, verbose mode, actual empty-execution guard, signal handling and focused verify name selection; compact canonical context/handoffs without mass moves. Tooling15, six negative CLI, named/empty source-bound smokes,206 doc links; same15-case278B vs2455B,2.35s vs2.36s no compute speedclaim. Native Advanced Ix/CHOAM47-card Stronghold profile: actual firstMentat/HMS/sixeffects,18native/four legalpolicies,five genuine2..6games3129actions/no rejection, human390px LAUDR2WNv14 bank2/personal1 plan and real battle. Earlier162/6513 checks before assurance amendment; finalquick25/build/guide pass. No new preservation/privacy audit or public/deployed/full-mode claim.
Fresh classic banker-income profile preserves actual bank legs, physical once-phase use and original-faction atomic Mentat collection. Provisional earned custody stays explicit; public/old/combined modes unchanged. Final check: types/lint and 6487 offline cases; focused 140; guide 10; build; native18/four legal budgets; two original six-seat CLI/human390px rooms and82 saved seats25rooms. Final two reviews no findings. Local HTTP49/55 six known upstream503 failures, not green. No reset or deployed claim.
Automatically return the full combined hand when it is the only legal
special-Karama return. Preserve real choices and restore the suspended
response or auction payment through the same settlement. Record the actual
blind draw as owner-only historical card faces, with a bound integrity
receipt; legacy forced saves describe held-card history truthfully.
Add readable private history and enlarged inspectors. Both manual and
forced returns publish only a faction/action completion tag. Wake idle
legacy exchanges through production recovery without writes for real
choices. Keep provisional auction funding rules and release gates.
Validation: types, lint, 5150 offline tests, build and 49 HTTP tests
passed. Independent review resolved history substitution and root review
completed idle recovery wiring. New tests cover 1-4 forced draws, real
choices, overlays, exact saved history, immutable rejection, all four legal
AI paths, production competing actions/recovery, private views and custody.
One genuine Advanced four-player sample completed 158 actions with
4 JSON restores and no rejected action; no difficulty tuning.
Browser QA QUTHVPTT used genuine Advanced setup and conserved Bidding
positions. The actual power returned four cards without a prompt; private
history/enlargement and exact refresh worked. Phone panel fit without
horizontal overflow with 44px inspector controls. Screenshot capture timed
out; complete visual acceptance and transient notice capture remain open.
Preserved every prior row across all ten tables, including 1586
rooms and 3354 seats. New QA/integration records bring totals to
1626 rooms and 3442 seats. No reset, restart or automation.
Let players compare normal, elite or total counters in reserves, the Tanks
or one live board territory and sector. Reuse compound facts and private
answers. Exclude concealed No-Field values and combat-strength modifiers;
guard Basic factions without a tracked starred-counter split.
Reparse saved fact questions before actions and projection, including
nested leaves and dynamic locations. Preserve interrupted decisions,
physical card disposal and historical answers after later legal movement.
Connect controls, contextual guidance and the capability checklist while
leaving full mode, combination and publication gates unchanged.
Validation: types, lint, 5122 offline tests and production build passed.
Independent review and 21 focused pure, engine, control and authenticated
concurrent recovery tests passed. Six complete-game samples finished with
1551 legal actions, 39 JSON restores and no rejected actions.
These are continuation checks, not AI strength calibration.
Browser room 86WE6LXV exercised genuine setup, compound physical counts,
AI response, mobile controls, human private-answer refresh and publication.
Two existing Truthtrance Cards retired once each and forces stayed intact.
All 1,584 prior rooms, 3,352 seats and auxiliary records remain unchanged;
one QA room and human seat were added. No server restart or database reset.
Existing screenshot capture failure leaves full visual acceptance open.
Connect ordinary Leader Skill paths to Basic CHOAM with base opponents,
the real 35-card Treachery deck and all fourteen skills. Retain wider
development initializer behavior and existing mode/publication gates.
Pause a declared sale until its Ghola revival and private skill choice
finish. Bind the actual revived leader, offer event and completion to the
saved interruption. Preserve nested Truthtrance disposal and exact sale
responses; canceling a resumed sale leaves the free revival completed.
Keep force-only and retained-skill revivals free of replacement offers.
Add genuine setup, native card-sale, mandatory disposal, private custody,
corruption and concurrent authenticated recovery regressions plus a
two-through-six-player complete-game sample profile. Record the unresolved
Sandmaster/native-HMS eligibility question without inventing a bonus.
Validation: types, lint, 5101 offline tests and production build passed.
Five genuine 2-6 player CHOAM Skills games completed with zero rejected
actions. Independent review and 50 focused compatibility tests passed.
Browser Y2UJGMCS exercised real setup, Ghola draw, exact private refresh,
mobile assignment and resumed sale paying two spice once. Screenshot
capture timed out; full visual acceptance remains open.
All 1,583 prior rooms, 3,351 seats and auxiliary records remain intact;
one QA room and human seat were added. No restart, reset or automation.
Full AI implementation and tuning remain deferred until all non-AI
features are complete, then target roughly 75% for each adjacent tier.
Admit genuine Basic Ixian setup with the full fourteen-skill deck, real
starting-card choice, optional Tleilaxu/base opponents and first-Storm HMS.
Keep public starts, Advanced and other module combinations gated.
Share per-origin cyborg movement range across engine, controls and legal bots.
Preserve independent Planetologist range after native Karama cancellation.
Connect Sandmaster force routes through room-local HMS pointer adjacency.
Retain canceled movement's no-op contract while validating actual execution.
Resolve Suk rescue before substitution with explicit ambiguous cyborg rescue
origins and exact remaining Tank losses. Finish Rihani before native
substitution, bind saved physical losses and declared exchange, and resume
mandatory/optional winner disposal on allow, decline or cancellation.
Admit exact green Ix Thumper, Harvester and Amal as Planetologist substitutes.
Validation: types, lint, 5,082 offline tests and production build passed.
Five genuine 2-6 player games completed with zero rejected actions and 54
JSON restores. Authenticated in-memory concurrent starting-card recovery,
private setup, physical card/force custody and independent review passed.
Browser FLGHBSA4 completed real setup, first Storm/HMS placement and a
conserved staged Suk battle. Mobile rescue, refresh and one-for-one
substitution passed. Screenshot capture timed out; full visual acceptance
remains open. All 1,582 prior rooms, 3,350 human seats and auxiliary records
were preserved; one new QA room/seat remains. No restart, reset or automation.
AI work remains limited to legal participation until every non-AI feature
is complete, then targets roughly 75% higher-tier wins for each adjacent pair.
Admit genuine Basic Tleilaxu/base setup with the full fourteen-skill deck,
Ix card inventories and native three Face Dancers. Share public profile guards
across existing skill effects while retaining public-start and Advanced gates.
Record native own-leader skill offers at successful revival commitment so
automatic Tleilaxu response chains preserve the optional private draw. Connect
Rihani to unrevealed Face Dancer exchange with exact physical/status custody,
private controls and existing legal AI choices. Release completed Rihani's
live snapshot only after validated Face Dance admission; keep signed history
and current census checks. Zoal Smuggler uses opposing printed disc value at
reveal, excluding opposing skill bonuses and his separate revival value.
Add genuine setup, response/cancellation, private choice, custody, Face Dance
reset, Zoal, legal AI and authenticated concurrent recovery regressions. Add a
reproducible two-through-six-player sample profile and saved continuation.
Preserve unresolved skill effects, foreign-ghola boundaries and delayed AI tuning.
Validation: types, lint, 5,058 offline tests and production build passed.
Five full games completed with zero rejected actions and 74 JSON restores.
Browser QA CRY4DD4J used genuine setup and an explicitly conserved staged battle;
real Rihani draw/exchange and same-state refresh passed on mobile and desktop.
Screenshot capture timed out; full visual acceptance remains open. No HTTP
endpoint/auth/storage implementation changed; authenticated SQLite recovery was
verified. All 1,581 prior rooms, 3,349 seats and auxiliary records were preserved.
One new QA room/seat remains. No server restart, database reset or automation.
Admit the genuine private setup profile for Basic Moritani with base opponents,
exactly the Ecaz expansion and no other optional modules. Keep all14 physical
skills, private two-card offers, starting-card/Traitor order and Moritani's final
six-force placement. Public starts and complete-mode/publication gates remain.
Share public profile eligibility across Banker, Diplomat, Sandmaster movement
and worm riding, Smuggler shipping/collection, Bureaucrat payments, Planetologist
controls/bots and Suk/Rihani/Sandmaster battle settlement. Admission reads no
hidden cards or random deal. Preserve base/CHOAM boundaries and existing source
questions; no full AI implementation, strategy or calibration changes.
Genuine setup regressions cover all14 assignments, legal profiles and guards,
Terror Assassination death and exact skill return, optional own
revival draw/privacy, Sabotage, movement controls/bots, Smuggler shipment and
Bureaucrat paid bribes. Battle actions cover Rihani exchange, Sandmaster spice,
Banker funds, Suk casualties and Smuggler collection. Actual enemy alliances
prove losing-ally retention excludes mandatory Planetologist Special and
Diplomat-copy disposal. A real worm/Nexus ride collects Sandmaster spice once.
Authenticated in-memory SQLite restarts and concurrent skill choices preserve
one assignment, private offers, saved seats and final Moritani placement.
Conserved staged positions are explicitly distinct from played phase histories.
Add a reproducible moritani-skills profile to the existing faction-game runner,
including fixed2-6-player Basic rosters, seed offsets16-20 and saved resume.
Verify expected module presence, all14 physical skills and ordinary Traitor
custody alongside existing card/force/private-view checks. Default six expansion
samples and base samples retain their rosters/ordinals. No live DB is used by
these complete-game samples. CLI regressions check interrupted continuation
and Advanced refusal; corruption tests catch lost cards and entire module loss.
Independent review caught an old all-base worm guard and a sample checker that
missed whole-module deletion; both fixed with meaningful regressions and reviewed.
Initial edits had a stale helper reference and final lint found an unused import;
these were corrected. Types, lint, all 5034 offline tests and build pass on
stable final source. Five genuine complete-game samples passed with no rejected
actions:2 players189 actions/5 restores,3 players53/1,4 players862/23,
5 players116/3 and6 players1338/36. These are legality samples, not calibration
or exhaustive faction/card acceptance. No endpoint/auth/storage changes; existing
HTTP suite not repeated, while new authenticated storage and live browser actions
exercise the newly admitted profile through production boundaries.
Browser created a fresh Basic human Moritani/Easy Emperor lobby, then the existing
backed-up local initializer started that named ready room without replacing seats.
Human inspected both Killer Medic bands, assigned Lupino Ord, chose a private
Traitor and placed six in Polar Sink after all other setup. Refresh preserved
the two public skills, own hand/Traitor and6 deployed/14 reserves at pending
Storm. Mobile skill panel339px, client/scroll337; viewport restored afterward.
Screenshot capture timed out, so full visual/screen-reader acceptance stays open.
All 1580 opening games, 3348 saved seats and auxiliary rows
remain identical. One isolated QA game adds one human seat and entry receipt;
its row is unchanged after setup through refresh, with all seats unchanged.
SQLite integrity and foreign keys pass. No server restart, migration, reset or
recurring automation. Existing unfinished Leader Skill bands and material
rulings remain explicit in the new integration guide and internal checklist.
Private evidence: 20260921-moritani-skills.
Verified source tree: d5078661ecc2199b2bbfebd18034b16585d06410a697a78dc514bfa5b3b77a1c
Before genuine Advanced Ecaz Traitor dealing, randomly set aside one of the five
native Ecaz cards face up for the game. Store its owner and exact identity once;
exclude it from the initial deck, Tleilaxu Face Dancer setup and Nexus replacement
universes. Do not change leader custody or add choices, payment or Karama windows.
Absent markers preserve older already-dealt saves without taking held cards away.
Validate owner, timing and distinct physical custody before actions, normalization
and projection. Expose a detached public card identity, never private deck order.
Show the named card, portrait and enlarged Traitor inspection above the table.
A dedicated Loyalty context explains public set-aside custody and cannot-call
rules; browser review caught and fixed the generic held-Traitor guidance.
Add internal help, checklist, shared Advanced faction text and source documentation.
The official page8 requirement and page16 noKarama entry support the effect;
relative pre-Traitor prediction/skill sequencing is an explicit composition.
Gated Skills/Ecaz and Moritani assassination/Ecaz are not certified.
Twelve new cases plus related regressions cover candidate selection, immutable
invalid-state rejection,2..6 genuine deals, Basic/noEcaz, completed Tleilaxu setup,
Harkonnen/Tleilaxu Nexus draws, full physical census, detached public views, all4
legal AI setups, JSON continuation, legacy custody and public rendering.
Independent review found no material issue after genuine setup/legacy test fixes
and the focused inspector correction.50 related focused cases pass. A mistyped
filename filter failed closed before the corrected union ran successfully.
Types, lint, all 5012 offline tests, production build and 49 HTTP tests
pass on stable source. Earlier type/lint findings were corrected before this run.
Six genuine faction-game samples finish with no rejected actions and repeated
JSON restoration: CHOAM Basic1054/Advanced225 actions, Ecaz Basic91/Advanced245,
combined Basic476/Advanced704. Existing physical Treachery/force custody checks
pass; focused Loyalty tests cover expansion Traitor census. These bounded legal
samples are not strength calibration or complete expansion acceptance.
Browser QA created a fresh Ecaz/Easy Harkonnen ready lobby and initialized its
Advanced setup before dealing using backup and CAS. Bindikk Narvi was public
before choices and remained excluded through actual human Traitor/force setup,
Storm and refresh. The saved human Traitor is Beast Rabban; a pointer action just
after viewport reset differed from its intended Feyd target, so exact-target
pointer acceptance across viewport changes is not claimed. Ordinary source
bindings remain name/ID-specific. One hand card,12spice,14reserves and the pending
Storm dial restore; both public projections agree and saved seat records match.
The Loyalty inspector shows correct card-specific guidance; Escape restores focus.
At390px its panel has no horizontal overflow,16px text and44/48px controls; the
expanded scroll body fits342px. Internal help opens its complete checklist topic.
Normal viewport restored. Screenshot capture timed out; full visual, long-inspector
keyboard and broader module acceptance remain open.
All 1539 pre-existing rooms, 3259 human seat records and
auxiliary rows remain unchanged. SQLite integrity and foreign keys pass.
Only isolated QA/HTTP fixtures add rooms. No server restart, database reset,
migration, recurring automation, AI strategy change or mode/publication gate change.
Private evidence: 20260921-gameplay-review.
Verified source tree: 881bf1912665350086d80be664d479a1b2724146da6ccc9576790c07e0449ee6
Players can select a non-allied recipient and explicitly pay a whole-spice bribe
outside Mentat and open decision/response windows. The server derives private
payment limits from uncommitted spice and existing joint Truthtrance completion
checks on isolated copies. Target-specific Bureaucrat guards preserve supported
redirect/allow continuations. Reserved bids, sealed spending and incoming escrow
cannot be spent twice. No strategic AI negotiation or mode/publication changes.
The panel separates deferred incoming bribes from spendable supply, disables
stale/invalid selections and uses existing request/version fencing. The chronicle
now describes actual payment and next-Mentat spendability rather than a promise.
Internal guidance and the implementation checklist record this boundary. The
separate Emperor gift/Bureaucrat source ambiguity remains explicitly pending.
Independent review found no material issue. Focused tests cover out-of-turn
payment, JSON escrow/collection, private budgets, invalid targets/amounts,
phase/response gates, reserved bids, joint promises, supported redirection and
rendered controls. Existing Bureaucrat SQLite/all-four legal response tests pass.
The new HTTP test races duplicate submissions and verifies one debit, refreshed
private balances and stale/over-budget rejection.
Types, lint, all 4968 offline tests, production build and 49 HTTP tests
pass on stable source. Preliminary checks caught a typed fixture opener and
semantic-status/internal-link/template lint requirements and the new SSR test
needed the existing Vinext Link alias. All were corrected before the
final4 run. Browser QA used genuine fresh Basic setup with a human payer and
Easy AI recipient: 11 rejected by controls, 4 paid from10 leaving6, one public
payment entry and unchanged recipient spendable10 with4 in escrow after refresh.
Keyboard selection and amount access passed. At390px the panel fits with16px
text and44px-or-larger controls; linked rules and back navigation work. Screenshot
capture timed out; full visual/assistive and all-configuration acceptance remain
open. Browser transport loss was not injected; HTTP concurrency is separately
covered. This is playability evidence, not difficulty calibration.
All 1418 original rooms, 2991 seats and auxiliary rows are
unchanged. SQLite integrity and foreign keys pass. The healthy server was reused
without restart, migration, reset or recurring maintenance.
Private evidence: 20260921-bribe-controls.
Verified source tree: b9a8594c5dcd85f78974c9647f24f5f6eba60a38c9dcd2badcf4529d1e825ae1
After legal placement, a living native Sandmaster may take one board spice from the single destination pile before faction arrival reactions. A ride has no intermediate ground route and never consumes normal movement. Multiple source sectors and typed elite forces earn one collection total. Each separate ride has its own opportunity. Decline and legacy actions retain ordinary riding; empty/ambiguous piles and unsupported module combinations suppress collection without preventing the ride.
Share the public quote across server, destination checkbox and existing bot candidates. AI only gains minimal legal participation: no new destination policy or strength calibration. Keep base factions plus Leader Skills as the bounded scope; other expansion rosters/modules, Ecaz treachery and multiple positive piles remain guarded. Record the composition of printed worm-movement wording with the existing Sandmaster contract, including the existing per-distinct-movement inference. No combined-effect FAQ or complete module compliance is claimed.
The room's atomic versioned write settles the physical transfer once before a pending Bene Gesserit reaction. Saved continuation resumes only the arrival and remaining ride queue. Ten new engine, control and SQLite cases cover Basic/Advanced custody, opt-out, legacy fields, rejected-action immutability, trainer visibility/control, public mode guards, private views, simultaneous duplicates, restarted arrivals, distinct queued rides and a genuine special-Karama worm through Nexus. Correct the shared shipment fixture to retain each faction's actual elite total and support Bene Gesserit as the opponent. Independent source and implementation review found no material defects. Focused work found and fixed the generic decision handler clearing the decision before the new quote, plus test-only fixture/type issues.
Types, lint, 4894 offline tests, production build and 43 HTTP tests passed on unchanged source. Two genuine four-profile Leader Skills samples finished in 265 Basic and 211 Advanced actions, no rejected candidates, and 14 JSON restorations. Neither naturally encountered the new effect; targeted checks establish it. This is playability evidence, not AI calibration.
Isolated browser room RMMA6LEF used a normally created Fremen seat and a conserved staged Advanced Leader Skills position. Selected two forces from each of two sectors, including one elite each, toggled collection off and on, then rode to Red Chasm. The UI showed four arrived forces and balance six before the opponent reaction. Refresh preserved that pending decision, balance and explanatory once-only collection history. The opponent's production-engine arrival response resumed to CHOAM charity; another refresh retained six spice and the private Chani traitor. Seat rows were unchanged. This is a targeted staged browser journey; screenshot and complete human-game acceptance remain open.
All 1140 original rooms, 2385 seats and every original recovery, entry and handover row remain unchanged. HTTP/browser checks added 34 rooms and 71 seats. SQLite integrity and foreign keys pass. No restart, reset, migration, maintenance automation or usage check was needed. No mode or publication gate is opened.
Verified source tree: 7996a37c13c71992d9bb6baf7016558615b98db064eda300ce19604b429b0d73
New Leader Skills battles bind territory spice to the public plan reveal and automatically pay the controlling player once the selected Smuggler survives, including a loser or captive with no surviving forces. Cap collection at the original single pile and unmodified disc strength. Pending spice cannot fund a plan. Settlement before Sandmaster's after-win addition remains an explicitly documented timing inference; modified strength, multiple positive piles and other module combinations remain gated pending the existing rulings and integration.
Share pre-seal admission across server, controls and all four bot profiles. Expose the public Ecaz card-variant flag so its guard also works in private views. Normalize the selected Kwisatz value once for UI validation and submission. Fix AI selection of unavailable native face-up skilled discs. Show reveal-time pending guidance and an automatic collection notice/chronicle without a new confirmation. Keep unversioned saved battles on their existing behavior.
Bind saved receipts to the original battle, controller, leader, plans, posture and pile; project no internal binding frame. Void collection on leader death, opposing/mutual Traitors and explosion without restoring destroyed spice. Apply captured-leader settlement before normal custody cleanup. Existing room versioning prevents concurrent final votes from paying twice.
Twelve new engine/SQLite tests cover Basic/Advanced winners and losers, caps, empty piles, zero remaining forces, weapon/Traitor/explosion outcomes, captive survival/death custody, Sandmaster ordering, corrupted receipts, immutable pre-seal guards, legacy continuation, legal paths for four profiles and authenticated concurrent/restarted settlement. Independent review prompted the Ecaz gate and stale-Kwisatz corrections and found no further material defect in the bounded delta. A focused regression also identified the missing public Ecaz configuration field, now projected.
Types, lint, 4884 offline tests, production build and 43 HTTP tests passed on the unchanged source fingerprint. Two genuine four-profile Leader Skills samples completed in 265 Basic and 211 Advanced actions with no rejected candidates and 14 JSON/private-view restorations; neither sampled game naturally exercised Smuggler collection, so focused and browser evidence establish that effect.
In isolated browser room YWV8AGKS, the Emperor selected Smuggler, sealed a zero dial, saw six pending spice without early credit, refreshed the pending decision and submitted no Traitor. After the opponent's production-engine final action, the browser showed the 6-8 loss, automatic six-spice collection, explanatory chronicle and restored balance 26 after another refresh. Opponent/setup staging was private and confined to the new QA room. Screenshot acceptance remains open. An initial private verification script compared SQLite null-prototype rows with parsed JSON too strictly; the transaction rolled back and the corrected content comparison passed.
After the reported outage the server was healthy and the previous private QA seat restored. All 1106 baseline rooms, 2314 seats and every original recovery, entry and handover row remain byte-for-byte unchanged; the HTTP/browser checks added 34 rooms and 71 seats. SQLite integrity and foreign keys pass. No restart, migration, database reset or automation was needed. No mode, module or publication gate is opened.
Verified source tree: dc55a36f7bfe108d6529b2e42add019d212648e6c17772b36e06fe73f9443265
New players can open Learn to play from the lobby or rules reference. Five lessons explain public/private information and turn phases, then offer three independent exercises for ordinary shipment budgets, a sealed battle plan and spice collection, followed by joining and recovering a real table. Shipping, combat and collection use existing production quotes rather than a separate simplified calculator. The fixed opposing plan is a teaching example, not an AI match.
Reuse the battle wheel and card inspectors, provide keyboard inputs, move focus to the current lesson heading and stack the layout on phones. Versioned browser-local choices preserve lesson position, shipment, defense/dial and reveal state. Whitelisted validation rejects corrupt saves and impossible committed shipments. SSR uses a stable hydration placeholder before reading browser storage; direct user handlers save choices and retain usable in-memory state on storage failure. No room, seat credential or multiplayer API is used.
Six new tests cover prices and affordability, every offered battle dial/defense, protected ties, weapon death, actual losses/bounty, collection limits and city capacity, saved-choice validation and internal rule links. Types, lint, 4872 offline tests and production build pass on the unchanged source fingerprint below. The first required check stopped on React setState-in-effect and semantic-status lint findings; these were fixed before the successful complete run. Independent review found no material defect in the examples or final hydration/save implementation.
Browser controls verified an unaffordable shipment, legal six-force shipment, card inspection, protected 8-8 aggressor victory, Snooper/projectile death and bounty, sealed refresh, city collection, lobby return/re-entry and newly changed choices after the hydration correction. At 390px the battle controls and lobby Learn link stayed within page bounds; the viewport override was reset. Screenshot capture timed out, so visual screenshot acceptance is still open. Storage-denial browser testing and the broader faction, movement, bidding, alliance, Advanced and expansion curriculum remain unfinished.
All 1106 current rooms, 2314 seats and all original recovery/entry/handover rows exactly match the fresh outage backup. SQLite integrity and foreign-key checks pass. The healthy server was reused; no restart, database migration/reset or maintenance automation was needed. HTTP integration and full-game samples were not rerun for this isolated teaching surface; server game behavior did not change.
A bounded independent primary-source review supports Smuggler's reveal-bound deferred survival receipt but found no exact ruling for modified leader strength. Record the one new user question and preserve the unanswered gate; lower collection remains missing. This does not enable any rules mode, expansion or publication gate.
Verified source tree: 99718ab49c4fec14820cbaf35f2e0f815e6424e910ee67c4daa1941afdeca2f0
During the existing public pre-plan opportunity, the holder can discard canonical unreserved Sapho to become aggressor before their readiness or any binding plan/Voice/inspection/Truthtrance commitment. Unchosen preparation prompts remain available. No new confirmation-only or holder-dependent response window is introduced. Later intervention and its allocation/privacy protocol remain unfinished.
A battle/turn/revision-bound history and independent event list preserve accepted priority while keeping attacker, defender and chooser stable. Legitimately recovered cards can change priority again at a fresh event. Reads, actions and normalization reject corrupt histories/custody. Shared tie priority respects Habbanya in either physical slot and reaches pre-commit Stone/Voice checks, public views, bot calculations and actual ordinary/Stone resolution. The final quote previously assumed original attacker priority and now consumes the validated aggressor. Harkonnen capture still follows actual winner/loser.
Owner controls, public aggressor/tie text, automatic notices and internal guidance are connected. Focused tests exercise actual tied results in Basic/Advanced, both Habbanya sides, mixed-force Stone finishability before opposing commitments, a real allied Prescience boundary, stale/late/cross-owner rejection, private rival noninterference, receipt corruption, saved recovered-card revision, actual policies at all four AI levels and authenticated SQLite races/seat restoration. Independent review found no material defect. A prior scheduling test now explicitly readies its human holder to save the card instead of allowing the new legal AI play.
Types, lint and 4866 offline tests, production build and 43 HTTP/session tests pass on the unchanged source fingerprint below. Two genuine CHOAM/Richese samples completed with 1279 accepted actions, 34 periodic JSON restores and no rejected candidates. One movement Sapho play occurred; neither sample used aggressor. Targeted tests and browser evidence, not those samples, establish the new effect. Full expansion and module acceptance remain incomplete.
In a new isolated browser room, genuine Advanced faction setup was followed by conserved battle staging. The human Guild used the actual aggressor button, restored the spent card and priority after refresh, readied, selected Master Bewt with zero dial/support, and declined a Traitor call. The staged Emperor opponent used real engine ready/plan/vote actions with Caid and zero dial/support. Guild won the 3-3 tie, retained four forces and 30 spice, and Emperor lost four forces. Final refresh and private JSON views restored the same outcome; the original seat row is unchanged. Screenshot capture timed out, so visual screenshot acceptance is not claimed.
All 1072 original games and original seat/recovery/entry/handover records are unchanged; 1106 rooms remain including isolated QA and HTTP fixtures. SQLite quick-check and foreign keys pass. The healthy server was reused without reset/restart or automation.
The Mirror audit now records why every genuine CHOAM/Richese setup includes unresolved copy candidates: an ordinary-copy-only gate would need hidden eligibility or altered inventory. Mirror stays gated pending the existing disposal ruling and copied-choice work. No mode/publication gate is opened.
Verified source tree: 2fb3160efd713a3f7068aedb474868ba6863bccc420848a56e69e1df1d84441b
The existing offline faction runner now selects genuine Basic and Advanced base games with two through six players. Original six expansion defaults and seed offsets remain stable; player-count filters require the explicit base profile. Saved AI profiles choose every setup and game action. This replaces repeated private harness copies while preserving source-bound private reports, failed traces and explicit snapshot resume.
Shared custody checks include every physical Treachery zone, nonnegative integer resources, twenty-force totals, the distinct Emperor/Fremen/Ixian elite totals and subsets, and canonical base Traitor identities after dealing. Rejected actions must leave their input unchanged; periodic JSON restoration compares all private views and excludes rival private hand/spice/Traitor/Face Dancer fields. Optional modules and separate private previews cannot be resumed as ordinary fixed scenarios. Resume still requires checking original setup provenance and intentionally restarts the selected random stream.
Independent review caught and corrected seven-cyborg accounting, save-derived Traitor expectations, missing elite records and private-preview classification. Meaningful regressions cover those failures plus strict filters, original names/seeds, failed subprocess reporting, source stability and resume. Types, lint and 4823 offline tests pass. No application, dependency, HTTP or database code changed, so build/HTTP suites were not repeated.
All ten base and six existing expansion samples completed on the final unchanged source: 6746 accepted actions, 175 periodic restores, 82 rejected candidates retained in traces. The expansion guarded-arrival fallbacks remain visible; these samples do not certify every faction, optional module, player-count rule or difficulty ordering. Exact expansion Traitor/Face Dancer custody still needs its own adapter. The earlier pilot/source-changed report remains private; the required check was repeated after the late review correction rather than accepting results from changing source.
All 1038 opening games and original seat/recovery records are unchanged; 1038 rooms remain. SQLite integrity and foreign keys pass. The existing QA table restored its saved seat and fulfilled shipment after browser refresh. The healthy server was reused without reset or restart.
The rules register now explicitly records the Moritani Secret Ally after-loss privacy blocker. A holder-only pause exposes its hidden Nexus; the existing alliance-retention helper does not resolve the shared pending response UX decision. No duplicate user question, new house rule or mode/publication gate change was introduced.
Verified source tree: 85d5a53fa72efb9704c1828526aaccba1554670307ce476f729eae9c5c4abff9
Shipment-only AND/OR questions now evaluate one eventual reserve shipment or skip as a whole. Yes to AND requires every leaf, Yes to OR requires one, and No negates the complete expression. Separate answered questions remain jointly binding. The completion search considers all destination branches and owned preparation paths. Release and fulfillment apply once to the aggregate promise, and legacy flat records retain their shape.
The two-condition browser builder, grouped question/history and respondent guidance, all four AI answer policies and saved continuation are connected. Nested expressions have explicit software size limits. Existing active-unused-shipment, Basic/no-Guild Advanced and module gates remain; mixed present/future predicates and unresolved rule questions are unchanged. Independent source, privacy and persistence review found no blocker.
Types, lint, 4815 offline tests, build and 43 HTTP tests pass with stable source. The first typecheck caught an omitted flat-only guidance consumer and a legacy test requiring type narrowing; grouped guidance and an actual promised-state regression fix them. SQLite Basic/Advanced tests verify private reauthentication, second-OR-branch fulfillment, zero-write rejected actions, legal No skip and duplicate CAS with one result.
Two fixed genuine Basic/Advanced mixed-profile games complete in 1534 accepted actions, 41 periodic private JSON restores and 0 rejected candidates. Each uses one scenario-directed compound question from an actual held Truthtrance at a real legal opportunity; AI supplies all responses and other actions. Exact card, force, typed-elite and post-setup traitor custody are checked. Initial private harness runs needed correct immutable setup configuration and full-game auction/Fedaykin inventory accounting; original failed traces remain private, and the final same-seed runs pass. The full sample tree is reconstructed exactly by replacing only the three later UI/test edits; all 219 game source files match the samples. No game logic changed after those runs.
Browser QA used genuine base initialization before conserved turn-two staging. The human asked twenty to Carthag OR four to Arrakeen; the AI answered Yes and shipped four to the second destination. Public grouped history survived refresh. All 6 accepted actions replay exactly, with one discarded Truthtrance, one fulfilled aggregate promise, private JSON views and original seat custody intact. This remains bounded prototype evidence, not complete Truthtrance or mode certification.
All 1004 opening games and original seat/recovery records are preserved (1038 rooms after checks), including all 1005 rooms present immediately before checks. SQLite integrity and foreign-key checks pass. The healthy server was reused without restart, reset or recurring automation.
Verified source tree: 1b24372fbe34e24db56e8f095c25744eeefb27581f2bc6cced2fb752f5bdcb94
Juice of Sapho can change the remaining Battle-phase chooser order at clean boundaries. A chooser may occupy the physical defender slot and select the actual other combatant. Physical storm order, attackers, defenders and tie advantage remain intact. Full-roster phase priority survives aftermath, newly eligible Face Dancers and saved continuation. Controls and all four AI profiles consume the same projected choices.
Independent review verified private options, public geometry, exact physical disposal, rejected-action immutability, phase expiry and boundary-bound saved chooser integrity. Actual Basic/Advanced Face Dance regressions verify later replacement priority. Authenticated SQLite tests cover duplicate concurrent card requests, private restore and battle completion. Chooser/aggressor separation and phase duration remain explicit implementation interpretations; the separate aggressor mode and unresolved timing stay gated. Reinforcements remains disabled pending the documented reserve-cost question.
Types, lint, 4792 offline tests, build and 43 HTTP tests pass with unchanged source. Six fixed genuine Basic/Advanced faction samples finish in 3672 accepted actions and 97 periodic private JSON restores. They retain 82 rejected candidates at existing guarded arrival combinations; subsequent legal candidates finish. This is bounded playability evidence, not complete compliance or AI calibration.
The browser used Sapho first from a later physical circle, chose Emperor, sealed and resolved a battle, then refreshed into its next choice. Emperor retained the tie win. All 9 accepted actions replay exactly using the recorded random battle event; physical custody, all private JSON views and the original seat row pass. A private harness initially needed random-event replay binding and SQLite-row prototype normalization; no production fix or game reset was required.
After the reported outage, the server responded and the staged QA game restored exactly. All 970 opening games and their original seat/recovery records remain preserved (1004 after checks), including all 971 rooms present immediately before checks. SQLite integrity and foreign keys pass. No restart, database reset or recurring automation was needed.
Verified source tree: b699f80600f8e297595a8e756edf367a607ddd99707474c03231e6479bd2f389
The independent Ecaz Treachery variant now supports Harass in either physical battle-card slot. A category-null Prescience answer keeps its identity private without erasing the selected card. Exact undialed normal/elite counters return before battle losses; opposing successful Traitors cancel withdrawal, ordinary leader deaths still resolve, and the card is discarded exactly once. Controls, all four AI profiles and saved continuation share the same physical quote. Existing Planetologist inspection and literal Truthtrance slot promises retain their meanings.
Unique Basic/Advanced physical allocations are supported. Ambiguous allocations, co-present Ecaz allied armies on either side, Richese card-family configurations and other optional modules remain gated before sealing from public/own context. The pending Stone Burner timing question is not assumed answered. Reveal-before-explosion order and mandatory disposal on an own-Traitor win are explicit source-timing/specificity inferences, consistent with existing Planetologist precedence. Reinforcements and full variant/public mode acceptance remain unfinished.
Independent rules/privacy/persistence review fixed asymmetric selector legality, misleading blocked-card guidance, invalid restored battle pairs, native-only inspection bypass and unsupported allied-army admission. Ixian fixtures preserve all seven physical cyborgs. The first broad run caught a semantic-status lint error in the blocked guidance; replacing the paragraph role with an output element fixed it. A subsequent broad run passed 4,775 of 4,776 tests and exposed an old UI assertion requiring category-null slots to be disabled. The updated regression verifies enabled category-null selectors exclude ordinary held cards, while exact leader/dial locks and submitted values remain bound; focused controls pass 14/14. No game source changed after the eight samples or browser replay. Types, lint, 4776 offline tests, build and 43 HTTP tests pass with unchanged source. Authenticated SQLite cases verify sealed private restoration and competing duplicate traitor actions with one physical resolution.
Eight fixed genuine six-seat games finish: basic/Easy: 1803 actions, 0 Harass plans, 0 withdrawals, 48 periodic restores; basic/Medium: 225 actions, 0 Harass plans, 0 withdrawals, 6 periodic restores; basic/Hard: 2184 actions, 1 Harass plans, 1 withdrawals, 59 periodic restores; basic/Brutal: 2073 actions, 1 Harass plans, 1 withdrawals, 56 periodic restores; advanced/Easy: 2057 actions, 1 Harass plans, 1 withdrawals, 55 periodic restores; advanced/Medium: 532 actions, 0 Harass plans, 0 withdrawals, 14 periodic restores; advanced/Hard: 538 actions, 0 Harass plans, 0 withdrawals, 14 periodic restores; advanced/Brutal: 544 actions, 0 Harass plans, 0 withdrawals, 14 periodic restores. All 218 game-source files match their source-stable runs; exact card/traitor/force custody and periodic/terminal private views are checked. Browser controls verify the full card, category None, Defense-first legality, actual sealing/reveal, four returned forces, ordinary leader death, disposal and refresh. A separate final-source replay verifies the conserved browser fixture and opposing continuation, including Face Dancer replacement. This is bounded playability evidence, not full compliance or AI calibration.
All 936 opening games and their original seat/recovery records remain unchanged (970 after checks); all 937 present immediately before checks are also preserved. Database integrity and foreign keys pass. The healthy server was reused after the power failure without restart, reset or recurring automation.
Verified source tree: be193613c455c5b76b49800fe3a7400504ea879af34c59621ade3d44f0fe019d
The explicit development initializer now includes the complete three-card Ecaz Treachery variant before genuine Basic/Advanced faction setup. Recruits has owned play/discard controls, full inspection, a public current-rate table, shared authoritative revival quotes, all four AI profiles and persistent exact free-use ledgers. Existing CHOAM/Tleilaxu choices settle before returning forces. Inactive games retain their prior allowance arithmetic.
Clean activation before normal returns or after exactly recorded free returns is connected. Prior-paid/unknown use, later Fremen grants, pending repricing, same-turn recovered-card replay and additional optional modules stay gated. Native unlimited faction advantages remain unlimited. The independent seven-force cap after faction cancellation is explicitly documented as an inference from independent effects, not a direct combined FAQ. Reinforcements and Harass & Withdraw remain unfinished; public starts and publication stay gated.
Independent rules/privacy/persistence review passed. Review fixed auction alias double-counting after a physical card sale, malformed canonical cards, legacy zero-ledger activation, a seat-control validation bypass and an ordinary-start bypass. Focused tests cover readiness and turn expiry as well as authenticated restart, exact free/paid accounting, stale/duplicate requests, competing writes and malformed-save rejection. Types, lint, 4743 offline tests, build and 43 HTTP tests passed with unchanged source.
Eight fixed-seed genuine six-seat samples finish: basic/Easy: 1961 accepted, 0 Recruits plays, 53 JSON restores; basic/Medium: 711 accepted, 0 Recruits plays, 19 JSON restores; basic/Hard: 239 accepted, 0 Recruits plays, 6 JSON restores; basic/Brutal: 1367 accepted, 1 Recruits plays, 36 JSON restores; advanced/Easy: 1402 accepted, 0 Recruits plays, 37 JSON restores; advanced/Medium: 530 accepted, 0 Recruits plays, 14 JSON restores; advanced/Hard: 545 accepted, 1 Recruits plays, 14 JSON restores; advanced/Brutal: 1240 accepted, 1 Recruits plays, 33 JSON restores. Reports retain rejected candidates and earlier harness attempts. All 216 game-source hashes remain identical to the samples; a later presentation-test expectation update changes no game code. Actual card use, custody and periodic/terminal private views are checked; this is bounded playability evidence, not full variant acceptance or AI calibration. A separate conserved browser Revival fixture verifies the full inspector, physical play, public rates, four free returns and active/resolved refresh.
All 902 opening games and original seat/recovery records remain unchanged (936 after checks). All 903 present immediately before broad checks are also preserved. Database integrity and foreign keys pass. The healthy server recovered after the power failure and was reused without restart, reset or recurring automation.
Verified source tree: 1f934f996268b35c8a20d0d617fa962aad878108496fa1a4a05323360757b0e3
Reciprocal allied Ecaz and Fremen with fighters in Sietch Tabr now satisfy that part of the Fremen final-turn victory, as explicitly clarified by the GF9 Ecaz/Moritani p15 FAQ. One public board quote drives automatic victory and the table guide; the chronicle explains the result. Other blockers, ordinary stronghold/BG prediction precedence, Guild fallback, advisor posture and concealed marker privacy remain intact. Habbanya symmetry and full mode/publication acceptance remain unfinished.
A genuine Advanced sample also exposed a pre-existing accepted Guild shipment which could never finish: allowance opened BG accompaniment and then hit the Ambassador arrival guard. Prior HEAD replay reproduced the exact 90-action failure. Fresh shipment preflight now rejects unsupported arrival combinations before accepting their pending windows. A valid legacy ordinary Guild allowance can return its unspent declaration to choose another destination; special parent/card receipts retain their separate gates. The change does not invent simultaneous reaction order or refund spent cards.
Independent rules/privacy/persistence review and focused engine/controls tests pass. Seventeen Fremen/Ecaz tests include all four profiles and authenticated final readiness/restart/concurrent writes. Shipment regressions cover pre-declaration immutability, supported arrival, malformed saved declarations and authenticated unspent return/replacement/reconnect/concurrency. All 4716 offline tests, types, lint, production build and 43 HTTP tests passed against unchanged source.
The original fixed seeds were repeated after repair and the exact failed snapshot resumed with documented new continuation seed 202609147: resume-advanced-202609147: 842 accepted, 109 guarded rejections, 22 periodic JSON restores; genuine-basic-202609145: 563 accepted, 14 guarded rejections, 15 periodic JSON restores; genuine-advanced-202609146: 806 accepted, 81 guarded rejections, 21 periodic JSON restores. All three finish and preserve physical custody/private views. These samples do not establish naturally reached allied Fremen special wins; a separate conserved browser final-turn fixture verifies the actual Ready-to-win transition, public guide/log, private hand and pending/finished refresh.
All 835 opening games and original seat/recovery records remain unchanged (902 after checks). All 869 rooms and records present after the reported outage also remain unchanged; database integrity and foreign keys pass. The healthy server and finished browser seat recovered without restart, reset or recurring automation. Full rules coverage and release gates remain open work.
Verified source tree: 1d9fcf30b54132f66be5f994f410dec6ec019683ee66b24fd4efb7a6aa7f7bbd
Connect a qualifying native Advanced Moritani battle loss to private reveal/continue controls, live or already-dead opposing-faction targets, printed-strength bank bounty, and automatic Mentat replacement. Keep the revealed physical card held until Mentat, then set it aside as a public per-faction use marker and draw one private card. All four AI profiles use the owner-only legal quote. New local preview profile supports genuine setup without opening public starts.
Keep unresolved normal traitor-call duration and uniform private no-choice UX explicit. Record normal calls and guard later use without claiming a settled permanent loss. Exclude exceptional captured/foreign leader custody, other expansion factions and combined optional modules from this bounded preview. No mode or publication gate opens.
Independent source/privacy/persistence review and 32 focused cases pass. Battle obligations independently bind cleanup, sealed normal-call logs wait for public revelation, physical replacement custody rejects missing cards and forged same-turn chains, and invalid actions remain immutable. Five authenticated SQLite cases cover pending/resolved restoration, Truthtrance interruption, competing submissions, stale/duplicate actions and corrupt saves. The local writer preserves seats and unrelated rooms.
Validation: types, lint, 4690 offline tests, production build and 43 HTTP tests passed with unchanged source. Two genuine Advanced four-profile games finish in 296 and 101 actions with zero rejected candidates, ten periodic JSON rounds and eight restored final seat views. All physical cards, forces and elites remain conserved. Two observed assassination opportunities are declined; these complete games do not establish natural reveal/replacement use.
The separate browser exercise explicitly stages conserved battle forces and one physical traitor swap after genuine setup. Actual battle actions open the choice. Master Bewt dies once and pays three spice; the card remains held through cleanup and is automatically replaced at Mentat with a public set-aside marker. Pending and resolved refresh, readable context-correct inspectors, physical custody and private projection pass. Guild independently collects two spice from Arrakeen. The saved QA game remains at human Terror placement.
After the reported outage, all 801 opening games and original seat/recovery records are preserved (835 after HTTP checks). All 802 rooms and original records immediately before verification also remain unchanged. Database integrity and foreign keys pass. The healthy server was reused without restart, reset or recurring automation.
Verified source tree: ad64f6775fb4c481002cc5eccf6d31e490085d11e10b7b19fb7df4947bcf0ccd
Add typed questions about the respondent's stored Bene Gesserit prediction faction/turn, active submitted storm dial and actually known native Advanced Fremen storm card. Use the same eligible private facts for authoritative validation and target-only answer projection. Preserve unknown/rephrase/save continuation, compound aggregate privacy and physical card custody. Future promises, freeform interpretation and strategic AI question selection remain incomplete; no mode or publication gate opens.
Connect four question controls with all numeric comparisons, canonical faction choices and shared input guards. All four AI profiles answer the owner-only verified scalar. Independent source/privacy review and integration checks fixed missing comparisons, parent selector mapping and invalid submission guards. Thirteen new cases and the 44-case focused union cover normal/unknown/dishonest answers, private views, JSON and authenticated SQLite restoration, interrupted storm-peek continuation, stale/duplicate/competing requests and corrupt saved predicates without writes.
Validation: types, lint, 4662 offline tests, production build and 43 HTTP tests passed with unchanged source. Two genuine-setup four-profile games completed without rejected actions: Basic 192 actions/five periodic JSON rounds and Advanced 518/fourteen. Exact Treachery/force/elite custody and all eight final restored seat views pass. Three explicitly scripted legal questions use naturally held cards, covering prediction faction, prediction turn and submitted storm dial. Forecast questioning is not exercised by these complete games; the samples do not measure autonomous AI asking strategy or strength.
The separate Advanced browser exercise uses real prediction and forecast progression, staging only physical Truthtrance acquisition by relocation. Invalid dial input is blocked; an unknown answer can be rephrased; both compound prediction clauses and the forecast comparison work. Refresh after submission and final resolution restores history unknown/Yes/Yes and the remaining Shield. Both physical Truthtrances are discarded once, unrelated resources remain unchanged and private projection passes. SQLite tests establish pending-answer restoration; the browser observation does not claim a paused pending answer after reconnect.
After the reported outage, all 767 opening games and original seat/recovery records were preserved (801 after HTTP checks). All 768 rooms and original records immediately before verification also remained unchanged. Database integrity and foreign keys passed. Reused the healthy server without restart, reset or recurring automation.
Verified source tree: 09daaac288cb91ea407f4573f853a59d307e12ed30d2eccd39e671a47336f40d
Add the absent-Emperor Secret Ally alternative returning exactly three eligible forces free in Basic/Advanced base-faction Nexus games. Preserve ordinary force/free-revival allowances and the shared Advanced Fedaykin cap. Keep the action optional and private, consume the physical Nexus once, and validate durable source receipts through later ordinary revival and Ghola. Purchase timing remains pending the earlier uniform winner-payment UX question; partial returns and combined-module interactions remain guarded. No mode or publication gate opens.
Connect private controls, explicit physical elite selection, all four legal AI profiles and a dedicated rules checklist entry. Extend the reusable local prototype writer with genuine twelve-card Nexus setup. Independent review covers rules timing, private projection, physical custody, JSON and production SQLite restore, lost-response replay and competing requests. Seventeen focused effect cases and four prototype-writer cases pass.
Validation: typecheck, lint, 4649 offline tests, production build and 43 HTTP tests passed with unchanged source. Genuine-start Basic and Advanced samples completed in 715 and 263 actions with 19 and 7 periodic JSON restores, respectively, and no Emperor uses. Two explicitly staged continuations completed in 18 and 17 actions with one normal-force Emperor return each. All four had zero rejections and preserved exact Treachery, force, elite and Nexus custody; all fourteen final seat views restored unchanged. They are bounded continuation evidence, not natural full-game acceptance of the effect or AI calibration.
In the conserved staged browser scenario, Advanced Fremen returned three forces including one Fedaykin using the Nexus, then three normal forces using the ordinary allowance. Reserves rose from three to nine, Tanks fell from seven to one, spice stayed three and the physical Nexus was spent once. The inspector was readable, the action notice displayed, and pending/resolved refresh restoration, physical custody and private projection passed.
After the reported power outage, all 733 opening games and original seat/recovery records were preserved (767 after HTTP checks); database integrity and foreign keys passed. All 734 rooms immediately before broad verification remained unchanged. Reused the healthy server without restart, reset or recurring automation.
Verified source tree: fd09a4bc71018cb7ef98c7e25c25dacc495ffed304dbd0cdfa223e539e8fa5e1
Add an optional owner-controlled two-spice diversion from eligible single-payer ordinary auctions, Guild shipment income, Richese auction proceeds and actual bribes in the bounded base/CHOAM Leader Skills profile. Debit the payer once, credit the recipient once, retain bribes until Mentat Pause and restore the original response/auction/advisor callback. Decline preserves later phase opportunities; successful physical-card usage persists ownership changes. Keep qualifying allied funding and unfinished payment families explicitly guarded or outside the connected batch. Record source authority and the out-of-battle Karama composition and physical-card usage inferences; no public mode or publication gate opens.
Connect readable owner controls and all four AI profiles using public payment/alliance information. Independent review found and fixed duplicate usage-event aliases and an overbroad guard on Bank-bound independent Karama shipments. Focused tests cover native/hidden/captured/dead eligibility, unchanged costs and card/force custody, normal and Richese auction callbacks, real Bene Gesserit advisor continuation, private views, JSON and production SQLite restore, stale/duplicate rejection and same-version competing writes.
Validation: typecheck, lint, 4631 offline tests, production build and 43 HTTP tests passed with unchanged source. Four controlled-shuffle genuine-setup samples completed: base Basic 718 actions/20 restores, base Advanced 551/15, CHOAM Basic 681/19 and CHOAM Advanced 174/5. All had zero rejected candidates and preserved full physical Treachery, force and skill custody. None reached a Bureaucrat payment decision; these are regression-continuation samples, not natural full-game redirection acceptance. The separate conserved staged browser exercise allowed one five-spice bribe and redirected two from a second; the payer spent ten, the payee retained eight deferred spice, and one skill use was recorded. Pending and resolved states survived refresh with unchanged physical custody and private projection.
After the reported outage, all 699 opening games and original seat/recovery records were preserved (733 after HTTP checks); database integrity and foreign keys passed. Reused the healthy server without restart, reset or recurring automation.
Verified source tree: a30615deed306e06ab4c153b9a7eabdc92d646ed79474e348a3500d525bcb8c8
Connect the optional named-weapon question before skill posture and faction powers. Every accepted question waits for the same target-owned private card-show action, preserving held/absent privacy; matching cards are compulsory, otherwise the owner selects a fallback. Keep cards in hand and retain historical observations only for the questioner and target, without binding later plans. Support actual controls, inspectors, all four AI profiles and saved JSON/SQLite continuation.
Normal activation remains disabled pending the user's uniform-response UX exception. The private flag is neither projected nor writable through actions. Legacy battles remain unchanged. Empty hands skip with an explicit unavailable receipt and no invented answer. Source and unresolved boundaries are documented, with a compact current-status dashboard preserving all goal, Git, subagent, usage and saved-game requirements.
Independent review passed privacy, custody, sequencing and preview gating. Twenty-two focused tests cover canonical card roles, exact-match/fallback symmetry, target-only controls, captured lower-only behavior, stale/duplicate requests and concurrent SQLite writes. Typecheck, lint, 4605 offline tests, production build and 43 HTTP tests passed with stable source. The first broad run retained one outdated Mentat card-copy assertion; update it to verify private disclosure, custody, optional question and subsequent concealment. Gameplay source was unchanged by that test-only correction.
Two genuine-setup sample games deliberately deal Mentat via a controlled initial skill shuffle. Basic: 100 actions, 3 JSON restores, turn-one Harkonnen win. Advanced: 507 actions, 14 restores, turn-five Harkonnen win. Each has one private observation, subsequent Atreides Prescience and no rejected candidate. These bounded force/skill custody samples do not establish general card-custody coverage, calibration or full module acceptance.
After the reported outage all 665 opening rooms and original seat/recovery/receipt records were preserved (699 after integration). A fresh 666-room backup preceded checks; database integrity and foreign keys pass. The existing localhost server was reused; its live lock prevented a duplicate start. Existing Ecaz and Ixian browser seats restored their same hands and pending choices. A separate conserved staged Mentat preview restored its private history and posture, then legally concealed its leader before subsequent faction powers. Human/browser paths are distinguished from genuine-setup simulations. No database reset, recurring automation, public activation or publication.
Verified source tree: 297ae06e0764d7cad525a802fe383a84971c3e9c5d6362c8cd0b79c5c3977caf
Final source tree: 177c98415098ea42174a860d8b1fac4f7c7dd6c3e67c93b18330d1d0fc4edf0e
The only post-verification edit removed one surplus newline at the end of the new developer document; its exact before/after content and fingerprints are retained privately. Staged diff checking caught this whitespace issue before commit. No application or test code changed.
Quintine