Files
Dune/docs/CURRENT_STATUS.md
T
Quintine e13ad454e2 Add movement and Traitor practice to the introduction
Expand /learn from five lessons/three exercises to seven lessons/five exercises. Movement practice uses shared board geometry and occupancy rules to teach ordinary range, separate-city ornithopters, storm blocking, full strongholds, one-group movement and conserved force counts. Traitor practice uses the production combat quote for an optional post-reveal call, a declined losing plan, a sole successful call and mutual betrayal, with the existing full private Traitor inspector. The fixed opponent scenario is explicitly teaching data, not live private information or an AI match.

Migrate every v1 saved lesson to its same identity after insertion, preserving previous choices and outcomes under the existing browser key. Validate v2 movement commitments and reveal/call stages, whitelist saved fields and default only new choices. Browser testing exposed stale pre-update React state surviving hot refresh; a schema-keyed lesson remount runs the migration for development updates too. No live rules, AI strategy, room persistence or server scheduling changed.

Eleven focused cases include all 96 movement combinations against actual authoritative actions, exact force/spice preservation, sole/mutual/declined Traitor outcomes, all five old lesson mappings and malformed or impossible save rejection. Independent review found no material teaching, privacy or restore issue. Types, lint, all 4899 offline tests and production build pass on the unchanged source fingerprint. HTTP and complete-game runs were not repeated for this browser-local guidance change.

Browser checks verified range, storm and full-stronghold explanations, committed movement after refresh, private Traitor inspection, decline and successful-call outcomes, and mutual calls after a refresh at the revealed-choice stage. At a 390px viewport both new lessons fit within the 375px client width without horizontal overflow; keyboard movement selection committed four arrivals and one source remainder. Screenshot acceptance remains open. The viewport was reset and /learn retained as the deliverable.

All 1174 rooms, 2456 seats and every original recovery/entry/handover row remain unchanged. SQLite integrity and foreign keys pass. No new live room, migration, reset, restart, usage check or automation was needed. Broader curriculum, rules/mode acceptance and publication remain unfinished. The status dashboard identifies durable unattended bot recovery as the next independent backend capability; pending faction-rule questions were preserved.

Verified source tree: 6f25e31c2d8865dac823762d7da782f21f7538034c9d3c0e853872d796ced6f7
2026-09-20 15:44:14 +10:00

126 lines
11 KiB
Markdown

# Current development status
Updated 20 September 2026. **The full twelve-faction goal is unfinished. Basic
is playable; public Advanced and unfinished expansion/module starts remain
gated.** Prototypes do not open completion or publication gates.
## Current checkpoint and work
Latest delivered gameplay checkpoint: `54c8030`, [Sandmaster worm rides](SANDMASTER_WORM.md).
Types, lint, **4,894 offline tests**, production build and **43 HTTP tests** passed.
Optional native Fremen destination collection connects multi-sector forces,
controls, minimal legal AI and saved Bene Gesserit arrival continuation. All 1,140
baseline rooms and 2,385 seats stayed unchanged; isolated checks added 34 rooms
and 71 seats. No restart was needed. The `8f7a62c` policy checkpoint removed the
usage cutoff and deferred AI refinement until feature completion, with three
adjacent 75% win-rate targets. Earlier evidence remains in the
[implementation history](IMPLEMENTATION_STATUS.md).
Current work: the [interactive introduction](INTERACTIVE_INTRODUCTION.md) adds
movement and Traitor practice through shared rules, legal controls and migrated
browser-local saves. Broader curriculum remains unfinished. Final verification
and delivery belong to the source-bound private report and Git message.
A bounded remaining-effect audit found that the reviewed faction/Nexus candidates
still cross recorded pending rulings. Continue independent features rather than
inventing answers. Durable unattended bot resumption is the next clear backend
capability: it needs a platform scheduler and saved due-work metadata; current
request/reconnect continuation does not establish it.
The connected [Harass & Withdraw prototype](HARASS_WITHDRAW_RUNTIME.md) supports either
battle-card slot, private category inspections, exact undialed normal/elite
returns, opponent-Traitor cancellation, mandatory disposal, controls and all
four AI profiles. New battles also have [revealed physical allocation](HARASS_ALLOCATION.md) for
ordinary/elite mixtures and multi-sector returns. Richese card combinations
and additional optional modules stay gated. Withdrawal before explosion and
card-specific discard precedence are explicitly labeled inferences. The Stone
Burner timing question is pending; no user answer is assumed. Reinforcements
remains unfinished. This does not open variant, mode or publication gates.
The preceding [early Sapho aggressor](SAPHO_AGGRESSOR.md) connects the
printed battle-priority effect during the existing shared pre-plan opportunity.
Physical attacker/defender/chooser stay stable; Habbanya takes precedence in either
slot. Shared priority feeds Stone admission, ordinary/Stone resolution, public
controls and all four AI profiles. Battle-bound history survives private saved
continuation. Later intervention remains unfinished; no new ruling is assumed.
Final checks, samples, browser evidence and delivery belong to the source-bound
private report and Git message.
The [Mirror admission review](MIRROR_WEAPON_ENGINE_AUDIT.md) found that every genuine
Richese setup contains the unresolved copied weapons. Ordinary-only activation
cannot safely depend on hidden custody or alter the printed deck. Mirror remains
gated pending the existing disposal ruling and copied-choice work. Continue with
independent missing faction/card functions, then integrated play and refinement.
The Smuggler follow-up review supports a saved reveal-time receipt with settlement
only on leader survival, but located no official ruling on modified leader
strength. The user has been asked about Kwisatz/trainer/Stunner composition; no
answer is assumed. See the [source follow-up](LEADER_SKILLS_RULES.md#20-september-smuggler-follow-up).
## Authoritative navigation
- [Rules checklist](../game/reference.ts), topic `implementation-checklist` and
each feature's entries: track implementation, controls, AI, documentation and
verification, plus Missing, Prototyped, Integrated, Verified and Polished stages.
- [Rule decision index](RULE_DECISIONS.md): settled contracts, source precedence
and pending material interpretations. Consult it before reopening research.
- [Development guide](DEVELOPMENT.md) and [verification tools](VERIFICATION_WORKFLOW.md):
architecture, focused checks, source-bound reports, reusable faction samples,
private backup and saved-seat restoration.
- [Implementation history](IMPLEMENTATION_STATUS.md): newest first. Historical
evidence and outage records stay there and in feature documents; read bounded
relevant sections and verify old absence claims against current code/tests.
- [Efficiency pilot](EFFICIENCY_PILOT.md): the first three milestones are pushed;
measurement limits remain explicit. Reuse tools and checks, without promising
an unmeasured savings percentage.
## Connected development capabilities
These links define bounded working behavior, not complete module certification.
| Area | Current connected work and evidence |
| --- | --- |
| Base/Advanced setup | [Genuine setup](ADVANCED_SETUP_TEST_SEAM.md), [Advanced study](AI_ADVANCED_SETUP_CALIBRATION_20260907.md), [Fremen cancellation](FREMEN_MOVEMENT_KARAMA_RULES.md), [Atreides full-plan audit](ATREIDES_FULL_PLAN_TIMING.md). Earlier Basic-then-flip fixtures do not establish current setup coverage. |
| Expansion factions | [Ixians/Tleilaxu](IX_PROTOTYPE.md) and [all seven faction selections](EXPANSION_FACTIONS_PROTOTYPE.md): genuine setup, Ecaz six-force placement, separate optional card variants, private controls and saved per-lot Ixian/Richese decline. Actual special-lot exchange remains pending. |
| Leader Skills | [Common lifecycle](LEADER_SKILLS_RUNTIME.md), [known skilled capture](LEADER_SKILLS_CAPTURE.md), [Mentat question preview](MENTAT_QUESTION.md), [Bureaucrat payments](BUREAUCRAT_PAYMENTS.md), five battle disciplines, [Planetologist](PLANETOLOGIST_RULES.md), [Suk Graduate](SUK_GRADUATE_RULES.md), [Rihani and other battle effects](LEADER_BATTLE_EFFECTS.md), [Smuggler shipment](SMUGGLER_SHIPMENT.md), [No-Field](SMUGGLER_NO_FIELD.md) and [battle collection](SMUGGLER_BATTLE.md), [Sandmaster routes](SANDMASTER_MOVEMENT.md) and [worm rides](SANDMASTER_WORM.md), [Banker spending](SPICE_BANKER_RUNTIME.md), [Diplomat defense](DIPLOMAT_DEFENSE.md). Remaining bands and combinations are explicit in those contracts. |
| Discoveries | [Prototype](DISCOVERY_PROTOTYPE.md): genuine setup, Great Maker, seven cards/eight tokens, inspection and stash rewards, nested sites, signed later free entry, carried Ornithopter, sole Cistern, bounded Jacurutu income, Testing Station and Shrine. Orgiz and contested/mixed cases remain pending. |
| Nexus/Homeworlds | [Decision index](RULE_DECISIONS.md) links each integrated faction family, native/borrowed effects, physical custody, private choices, payments and transport. [CHOAM Collection trade](NEXUS_CHOAM_SECRET_ALLY.md) and [Emperor extra revival](NEXUS_EMPEROR_SECRET_ALLY_RUNTIME.md) are connected; CHOAM inspection and Emperor purchase remain pending. A source audit or helper alone is not a completed effect. |
| Truthtrance | [Spice facts](TRUTHTRANCE_SPICE_FACTS.md), [card counts](TRUTHTRANCE_CARD_COUNT.md), [hand inventory](TRUTHTRANCE_HAND_INVENTORY.md), [recorded knowledge](TRUTHTRANCE_KNOWLEDGE.md), and [Basic/no-Guild Advanced reserve-shipment promises](TRUTHTRANCE_SHIPMENT_PROMISES.md) use authoritative private state. Earlier readiness audits do not supersede these follow-ups. |
| Other cards | [Sapho runtime](JUICE_OF_SAPHO_RUNTIME.md) includes first among remaining unstarted movement turns after Advanced Guild has finished. [Recruits](RECRUITS_RUNTIME.md) connects clean Revival play; [Harass & Withdraw](HARASS_WITHDRAW_RUNTIME.md) connects the bounded battle return in the same independent three-card preview. Richese card contracts and the checklist identify other connected effects and explicit gaps. |
| Multiplayer | Saved rooms/seats, [recovery](../README.md#saved-seats-and-reconnecting), uncertain-request retry, [own-seat AI](AUTOPILOT.md), and [voluntary seat handover](SEAT_HANDOVER.md) have connected controls and recovery evidence. |
## Remaining readiness
| Area | Remaining gate |
| --- | --- |
| Basic core | Ordinary Guild repricing/transport settlement, broader truthful commitments, unresolved special timing and integrated rules/human/AI acceptance. |
| Six-faction Advanced | Base gaps, provisional special-Karama outcomes, remaining timing audits and authentic multiplayer/browser acceptance. Setup matrices and complete samples do not open public starts. |
| Twelve factions and all optional modules | Every missing effect, remaining skill bands, Ecaz card variant, Kull Wahad, unfinished Richese cards, Terror effects, occupation entitlement, Nexus interactions, contested Discoveries and complete games across valid combinations. Detailed boundaries remain in the checklist/decision index. |
| AI | Maintain minimal legal participation and critical correctness fixes while game features are unfinished. Full implementation, strategy refinement and calibration wait until all non-AI features are complete; then target approximately 75% higher-tier wins for Medium/Easy, Hard/Medium and Brutal/Hard. See the [AI development plan](AI_DEVELOPMENT_PLAN.md). These targets remain unverified. |
| Multiplayer | Broader disconnected/abandoned-seat recovery, unattended continuation and network-failure acceptance, while preserving authoritative versioning, custody, privacy and saves. |
| Components and experience | Full component inventory/text verification, all readable inspectors, original artwork, animation/sound/accessibility/mobile polish, complete internal guidance and end-to-end human play. |
[AI calibration](AI_CALIBRATION.md), [multiplayer audit](MULTIPLAYER_AUDIT.md),
[component inventory](COMPONENT_INVENTORY.md) and [visual playtest](VISUAL_PLAYTEST.md)
record their measured scope. [README](../README.md) retains publication gates.
## Working sequence and preservation
Prototype every remaining function in dependency order, using existing controls,
legal AI and the saved-state model. Continue independent functions while material
rulings are pending. Then run integrated games, repair failures, deepen combination
and recovery coverage, and complete all non-AI features. Only then fully implement
and refine AI strategy and calibrate the adjacent difficulty pairs. Keep focused
checks for crashes, deadlocks, legality, custody and privacy; complete required
broad checkpoint checks.
Use selective subagents and independent review for complex rules/privacy/persistence.
Commit and push verified checkpoints under **“Push all from now on”**, verify push
success. The user removed the usage cutoff and routine account-usage checks;
efficient work, Git delivery and selective subagents remain required.
The user-authorized 13 September reset happened once; preserve all games created
since it. Keep `.wrangler/state`, credentials and private backup artifacts out of
Git. Reuse a healthy server; restart only for a relevant change or observed problem,
with a fresh backup, safe human-play boundary and verified restoration. No hourly
restart or recurring maintenance automation is installed; do not recreate one.