Commit Graph
34 Commits
Author SHA1 Message Date
jpmschweitzerandClaude Opus 4.6 1254e41dc5 docs(decisions): amend stale D-records, close resolved Q-records
Amend 7 high-priority D-records with supersession notes:
- D-012: map size superseded by D-094 (256x256)
- D-044: outline color scoped to perception mode per D-154
- D-061: width corrected from 640px to 1200px per D-076
- D-086: smuggler scope removed per D-117/D-122
- D-090: marked [SUPERSEDED] by D-117 (tycoon replaces smuggler/detective)
- D-035: character enum marked obsolete post-D-117
- D-119: sequencing superseded by D-166 cascade

Close 4 resolved scope questions (Q-002, Q-004, Q-005, Q-007).
Fix questions.md index — was 61, actually 96 (33 architecture Qs missing).

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-04-05 15:16:27 +02:00
jpmschweitzerandClaude Sonnet 4.6 b21b28df61 docs(decisions): D-168 Iserlohn (GJ-532c) IP evaluation — retained
Real German city origin (Sauerland wire-drawing centre) documented.
Context diverges fundamentally from LoGH fortress. Altena recorded
as preferred alternative. Ticket #773.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-05 09:28:34 +02:00
jpmschweitzerandClaude Opus 4.6 7aa1c71d92 docs(decisions): file D-166 (development cascade) and D-167 (heritage root supersession)
D-166: 6-phase cascade as architectural constraint. Each phase
completes before the next begins. No v0.2 target.

D-167: 7 abstract heritage roots retired. Real-world corridor
cultural system (Commonwealth/Iberian/East Asian/Germanic) is
canonical. Supersedes D-104, D-105, D-101 axis A, D-107 (partial).

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-25 14:36:22 +01:00
jpmschweitzerandClaude Opus 4.6 d9c2811842 feat(content): replace overheard.yaml with generator-compatible RON format
Archives the 48 hand-authored overheard conversations that referenced
deleted named NPCs (Kael, Sera, Torek). New RON pool format is
parameterized by role pair and zone type, culture-neutral, with
knowledge_payload for investigative value. 15 sample conversations
across 5 zone types. Adds content pattern note to decisions/content.md.

Ticket: #664

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-17 10:57:49 +01:00
jpmschweitzerandClaude Opus 4.6 a69c75d4ae fix(docs): address PR #92 review comments
- Fix D-134 cross-references to point to architecture.md (not content.md)
- Fix stale #620 reference in D-147 body (now #679)
- Fix content.md footer count (44, not 45)
- Add D-145, D-146, D-147 to decisions/README.md index
- Add Q-WTF-041 entry to questions-content.md (was referenced but untracked)
- Fix sprint-28-candidates.md totals line
- Align spelling to British "customisation" throughout new decisions
- Add information boundary classification to D-147 (public trait, per D-010)
- Add "first message" instruction to proposer rules in sprint-start skill

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-17 08:26:58 +01:00
jpmschweitzerandClaude Opus 4.6 0d18c76dfa feat(docs): assign teams to v0.2 story tickets (#614-#626)
Resolve Q-WTF-039 (tile-scale heavy zoom, not portrait — D-146) and
Q-WTF-040 (aesthetic taste trait as shared root — D-147). Split #617
into server generator + client rendering, split #620 into server
culture resolution + client location picker. Sprint 28 candidate list
with 7 unblocked tickets.

Ticket: #676

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-17 08:10:29 +01:00
jpmschweitzerandClaude Opus 4.6 5402435045 fix(docs): review cleanup — broken anchors, typo, status updates, db backup
- Fix 3 broken D-036/D-128 anchor slugs in decisions/content.md
- Fix typo: Stalowsia → Stalownia in stalownia-kowalski.md
- Mark proposed-llm-voice.md as superseded (D-138)
- Mark smuggler-moral-arc.md and smuggler-voice-card.md as v0.1-only
- Snapshot settledreach.db after planning branch merge

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-16 01:24:27 +01:00
jpmschweitzerandClaude Opus 4.6 181baf9f11 docs(decisions): amend D-095 gate lore, resolve Q-039 topology
D-095: clarify inter-system gates are alien-determined (humans cannot
add/remove apertures). Intra-system span gates are Institute-derived
human tech licensed to the Gate Corporation. Dormant activations
(Struve) are alien infrastructure events, not human actions.

Q-039: resolved — 301-system star map generated with alien-determined
topology (334 edges, hub/junction/loop/spur distribution).

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-14 03:41:39 +01:00
jpmschweitzerandClaude Opus 4.6 e8263e209e refactor(data): rename Krenn to Van Maanen's Star
System S-057 assigned to real star GJ 35 (Van Maanen's Star, DG white
dwarf at 13.9 ly). Renamed across all content, server code, docs,
decisions, wiki lore, and config files. 224 files updated.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-14 00:23:26 +01:00
jpmschweitzerandClaude Opus 4.6 5f266ae002 chore(meta): merge main into planning, resolve CHANGELOG conflict
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-13 23:02:07 +01:00
jpmschweitzerandClaude Opus 4.6 048e13b5bf docs(docs): amend D-095 aperture range to 1-8
Revised from 4-8 per Miri's worldbuilding input — single-aperture
dead-end systems are valid and narratively important for isolated
frontier outposts.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-13 22:57:12 +01:00
jpmschweitzerandClaude Opus 4.6 aa55abb684 docs(docs): remove model provenance policy from D-138
Gemma 2B is picked and embedded — the provenance policy text was
an unnecessary addition. The model choice is final.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-13 12:24:16 +01:00
jpmschweitzerandClaude Opus 4.6 65e033443f docs(decisions): file D-140 dialogue quality rules, D-142 zone-type template architecture (#647)
D-140: Paula's six rules for dialogue re-voicing quality — secret-tier
passthrough, epistemic weight, access tier feel, named entity
passthrough, relationship-specific lines, culture register override.
D-142: Zone-type template architecture — three-layer behavior system,
31 zone types for v1.0, POI three-tier system, scale math for 300+
systems with 300+ cultures. Amend D-123 (baked vs pre-voiced modes),
supersede D-124 (resolved by two-spike strategy).

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-13 11:46:03 +01:00
jpmschweitzerandClaude Opus 4.6 50800cbf81 feat(simulation): composable behavior engine — action+modifier+context primitives (#633)
Replaces flat culture×zone×role behavior strings with three-layer
composition: BehaviorAction (role-generic), BehaviorModifier (culture
coloring), BehaviorContext (situation gating). Assembly at NpcBlueprint
instantiation. Backward compatible — falls back to legacy behaviors
when primitives are empty. D-139 filed, Q-057 resolved.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-13 09:12:07 +01:00
jpmschweitzerandClaude Opus 4.6 a2554118a5 docs(decisions): amend D-138 with Spike 2 findings
Spike 2 amendments: stdio IPC (not HTTP, Gemma 2 T&C compliance),
tell differentiation results (3/5 at 2B capacity), double-prompt
technique, ContentType::Factual for LLM bypass, all negative
injectors moved from universal RULES to per-culture voice_persona.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-07 19:26:07 +01:00
jpmschweitzerandClaude Opus 4.6 33030fcc58 feat(engine): voice pipeline Phase 1 — composition engine and data model
Add the voice pipeline composition engine (D-138 Spike 2, Phase 1):

- voice/prompt_builder.rs: full prompt assembly from culture profile,
  tell state, and base text. Handles occasional injection gating,
  epistemic marker extraction, content-length-gated tell injection.
  16 unit tests.

- blueprint.rs: CultureProfile gains voice_persona, voice_examples,
  occasional_injections fields. NpcBlueprint gains tell_behaviors.
  OccasionalInjection struct with kind discriminator (oath/faith/
  hesitancy/etc), frequency, and tell-suppression gating.

- culture-krenn.ron: v2 voice injector from Spike 1 — persona block,
  3 examples, oath injection at 0.25 frequency.

- D-138 amended: Phi-3 dropped entirely, exact Gemma 2B provenance
  documented. Model file renamed to gemma2.gguf.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-07 16:35:39 +01:00
jpmschweitzerandClaude Opus 4.6 2d03776365 docs(decisions): amend D-138 with Spike 1 findings
- Tell-variant caching: 6→length-gated (short=neutral only, medium=3,
  long=6). 2B model produces identical output across tell states on
  short lines — confirmed across two test rounds.
- Composition-engine occasional injections: oath vocabulary, faith
  expressions etc. controlled by prompt generator frequency, not model.
  Systemic pattern for any culture marker that should appear occasionally.
- NI-1/NI-5 culture-gated: religious language and Earth-origin markers
  are per-culture injector constraints, not universal bans. Cultural
  heritage from colonization history is intentional. Earth is not lost.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-07 15:45:45 +01:00
jpmschweitzerandClaude Opus 4.6 9f91fd077f docs(workshops): LLM voice pipeline workshop — D-138, D-123 amended, D-124 superseded
3-round workshop (7 participants + Qatux + SI) deciding content generation
architecture for NPC observable behaviors and dialogue.

Key decisions:
- D-138: LLM re-voicing pipeline (Gemma 2B Q4, llama-cpp-rs, bundled)
- Behaviors + dialogue both re-voiced; tells always passthrough
- Tells as read-only context inputs shaping surrounding content tone
- Cache-as-determinism, separate thread pools, layered hardware detection
- Two-spike validation: plumbing first, then integration
- D-123 amended (authoring tool + runtime enhancement)
- D-124 superseded (door walked through)
- Q-012 and Q-057 resolved

Artifacts: Krenn injectors v2, NI-1-5, culture template, 6 dialogue
constraints, tell-tone injectors, spike payloads, 12-risk register.
9 tickets created (#638-#647).

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-07 13:31:27 +01:00
jpmschweitzerandClaude Opus 4.6 eea3f3cf25 docs(decisions): record 24 workshop decisions and updated questions
D-records from Where's the Fun workshop across architecture, content,
and scope domains. Updated open questions for v0.2 pivot.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-06 20:56:46 +01:00
jpmschweitzerandClaude Opus 4.6 3a481b32a5 fix(copy): address PR #84 review — pacing, prerequisites, terminology
Review fixes for triangle activation monologue lines:

- CHANGELOG: correct smuggler line count (4 → 5), add D-035 ref
- Smuggler comment: align beat labels to 5-line structure
- Smuggler 040: "looking at" → "seeing" for body-first register
- Detective 043: rewrite to remove implicit manifest knowledge
  reference — line now works without fact prerequisite gate
- Detective 044: soften from near-certainty to enumerated
  possibilities with "insufficient data" qualifier
- Triangle comments: align to D-087 terminology (T1: Kael-
  Smuggler-Ring, T2: Sera-Detective-Commission)
- Schema description: note triangle_activated also missing from
  server Situation enum alongside greeting
- D-035 amendment: register triangle-signal, tell-observation
  tags and npc_in_los prerequisite as conventions

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-05 10:56:50 +01:00
jpmschweitzerandClaude Opus 4.6 2f2f8eb06d docs(docs): file D-096–D-109 and Q-045–Q-050 from generator workshop
14 confirmed decisions filed across architecture.md (10) and content.md (4).
6 open questions filed in questions.md (Q-046 resolved immediately by D-108).
README.md index updated.

Key decisions: DistrictLayoutMode (D-096), guarantee tier system (D-097),
MobileChunk spec (D-108), DamageOverlay prohibition (D-109), WorldTier
enum with corrected constraint ceilings, heritage grammar overlay (D-104).

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-27 10:40:15 +01:00
jpmschweitzerandClaude Opus 4.6 98d28b5a54 docs(discussions): archive workshop #153 station district layout
Three-round workshop producing D-093 (Sova Transit District spatial
layout), D-094 (chunk/block/district hierarchy), D-095 (horizon
stations and transport lore). Resolves Q-040, Q-041, Q-043, Q-044;
partially resolves Q-042. Establishes spatial hierarchy as
architectural precedent for Q-036 generator.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-25 19:08:29 +01:00
jpmschweitzerandClaude Opus 4.6 f874979ff9 docs(decisions): file 6 D-records and 10 Q-records from workshop backlog
D-087 through D-092: v0.1 triangle config, 3-state pause, self-contained
forks, PC voice registers, complicity as theme, anchor line requirement.
Q-030 through Q-039: long-term strategy questions from wiki-review R4
(seed schema, generator pipeline, Sacred/Profane, 300-world estimates).

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-25 12:18:53 +01:00
jpmschweitzerandClaude Opus 4.6 9cec3df68c feat(content): D-084 dual-namespace line ID scheme — resolve Q-028 (#544)
Role namespace (shared pool) stays unchanged; instance namespace
(role-slug + seeded counter) added as opt-in for per-instance
authored content on auto-generated NPCs. No schema migration needed.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-25 02:27:33 +01:00
jpmschweitzerandClaude Opus 4.6 b97f65e8ab docs(decisions): D-035 amendment — NPC-scoped line ID namespace
Sprint 15 amendment: line ID namespace changed from location-scoped
to NPC-scoped. Each NPC's IDs are independent, eliminating cross-file
coordination and collision risk at scale.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-23 15:36:29 +01:00
jpmschweitzerandClaude Opus 4.6 91eb11da16 refactor(content): rename mood vocabulary to match voice guide (D-035 Sprint 14 amendment)
Schema and Rust Mood enum renamed for author-friendly vocabulary:
fond→warm, comfortable→content, worried→anxious, concerned→frustrated.
Dropped: analytical (merged into focused), conflicted (modeled as
suspicious+warm collision). Added: hostile. Final 8 moods: anxious,
frustrated, content, suspicious, warm, hostile, relieved, focused.
Neutral = untagged. Resolves Gestalt's blocking issue on #121.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-20 19:00:07 +01:00
jpmschweitzerandClaude Sonnet 4.6 cfbd00f876 docs(decisions): record Gestalt endorsement of D-075 archetype dimension review
D-075 dissent was pending Gestalt and Nigel input on the no-archetype-tag
decision. Gestalt endorses: access tier tags in content handle archetype
distinction emergently; starting knowledge differentials produce per-character
gate timing without hardcoded archetype filters; maintenance cost of archetype
tags would undermine the "two keyholes on the same world" experience (D-027).
Nigel's input still pending.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-02-19 16:19:31 +01:00
jpmschweitzerandClaude Sonnet 4.6 d103e445e7 feat(simulation): contraband detection + dialogue confidence gate (D-075) (#425, #524)
Contraband detection (Task #425, D-065):
- New module: simulation/contraband.rs — NPC scan checks carried items
  against the KnowledgeGraph confidence gate. NPCs with Authority access
  can initiate a scan; scan outcome depends on item CarriedBy + KG entry.
- Adds ContrabandScanResult event type and ContrabanEntry component.
- Wired into simulation/mod.rs module list.

Dialogue confidence gate (Task #524, D-075 — OQ-18 resolution):
- relationship_to_trust() gains confidence parameter (KnowledgeConfidence).
- Trust tier mapping: (Friendly, KnowsDetails+)→Secret, (Friendly|Known,
  KnowsOf+)→Real, otherwise Surface. Access tier (Layer 1) unchanged.
- Caller process_talk_interaction passes observer KG confidence_of target.
- Resolves OQ-18: confidence co-gates TrustTier, not AccessTier.

Supporting changes:
- decisions/content.md: add D-075 (16 decisions, dated 2026-02-19)
- knowledge/types.rs: expose KnowledgeConfidence comparison helpers
- knowledge/registry.rs: minor API polish
- bridge/types.rs: ContrabandScanResult wire type
- bridge/text_renderer.rs: render contraband scan status
- perception/observer: include carried item count in snapshot
- npc/mod.rs: NPC scan range constant, authority flag

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-02-19 11:12:24 +01:00
jpmschweitzerandClaude Opus 4.6 4af2197c59 fix(ci): address PR review — filter comments, trim whitespace, warn on missing scripts
Address all review comments from Hoshe and Tyre on PR #27:
- Remove 2>/dev/null from pre-pr-fixtures (critical: swallowed errors)
- Remove dead _file_type function
- Check 4: error on districts with no locations declared
- Check 5: print advisory message when skipping
- Check 8: cross-file line ID uniqueness (not just per-file)
- Check 9: document D-034 asymmetric relationships in docstring
- Document regex fallback rationale in _scan_knowledge
- Add D-035 decision trace to schema descriptions
- Use concrete protocol version in DEVOPS.md example
- Amend D-035 with focused (9th mood) and greeting (14th situation)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-18 00:23:21 +01:00
jpmschweitzerandClaude Opus 4.6 ac86c8a72b docs(decisions): record audio discussion outcomes D-067 through D-074
Recognition chime timing (D-067), 5-bus audio architecture (D-068),
audio dip profiles (D-069), confrontation as cognitive vulnerability
(D-070), monologue chime placeholder strategy (D-071), universal
conversation murmur (D-072), zone crossfade (D-073), hybrid audio
generation (D-074). Amends D-038 scope, resolves Q-014.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-16 01:03:02 +01:00
jpmschweitzerandClaude Opus 4.6 c28d9b6501 feat(decisions): formalize D-053 through D-065 from Control & Interaction Workshop
13 new decisions covering movement stances, tile-based movement, cursor states,
entity interaction, fog rendering, dialogue UX, confrontation, and smuggler inventory.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-14 09:34:42 +01:00
jpmschweitzerandClaude Opus 4.6 010c4158f9 docs(copy): address PR #15 review feedback
Miri (critical):
- Fix 3 stale "Hael" references in decisions/content.md (D-034, D-036)

Paula (suggestions):
- Flag dual-fire 15-30s delay as tuning target in smuggler moral arc
- Clarify MIRROR NPCs can be wrong (lied to) without being dishonest
- Add implementation note for Path C deferred follow-up fork
- Note detective-NPC dialogue rotation variants for post-v0.1
- Clarify detective vs smuggler moral_weight criteria in entity attributes

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-12 19:07:24 +01:00
jpmschweitzerandClaude Opus 4.6 745ce01117 docs(decisions): add D-032 through D-039 from content workshop
D-032: separate monologue pools per character
D-033: entity color = relationship to player
D-034: THE FRIEND production-level NPC pattern
D-035: converged tag taxonomy for line pools
D-036: Sova Transit District / Krenn System setting
D-037: contraband specification (unlicensed lattice components)
D-038: audio in v0.1 scope (8 files via Stable Audio Open)
D-039: v0.1 wow moment scope (all 6 moments)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 20:27:37 +01:00
jpmschweitzerandClaude Opus 4.6 4f5ee0274b docs(decisions): split DECISIONS.md into domain-scoped files
Replace monolithic 474-line DECISIONS.md with 7 domain files under
decisions/: architecture, perception, content, scope, process,
questions, rejected. Each file owns its decisions with cross-reference
hyperlinks. Root DECISIONS.md becomes a redirect with domain index.

Reduces per-agent context load by ~70-80% (80-150 lines vs 474).
Taxonomy: architecture (D-008..D-031), perception (D-011..D-019),
content (D-023..D-029), scope (D-001..D-027), process (D-004..D-022),
questions (Q-001..Q-011), rejected (R-001..R-010).

Restores D-002 original text (was a tombstone referencing D-005).

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 09:28:57 +01:00