Build _tile_by_coord dictionary from member visible_tiles (covers both
test-mode "tiles" key and live-server "visible_tiles" key), then replace
the linear scan with a single dict lookup. Net-zero complexity: adds one
dict-set per tile in an existing iteration, removes the separate scan loop.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
sprint-start: add step 9 (post-work lifecycle) — commit, push, review,
fix-comments loop, approve, then shutdown. Team stays alive until PR is
accepted. Step 8e updated to reference the new lifecycle.
pr-push: add step 9 (team awareness) — don't shut down agents after
pushing, suggest /pr-review next.
pr-review: add step 8 (post-review team actions) — cross-references
sprint-start step 9c for dispatching review comments to agents on
CHANGES_REQUESTED and shutdown on APPROVED.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Tyre #1: Extract run_dialogue_pipeline() shared helper — eliminates ~60
lines of duplication between process_talk_interaction and
process_dialogue_response (L1-L4 pipeline).
Hoshe #1: process_dialogue_response now updates ActiveDialogue with
current tick on follow-up selection — prevents stale started_tick.
Tyre #4: process_dialogue_response now updates InteractionMemory on
follow-up — multi-turn conversations are visible in history.
Hoshe #6 / Tyre #6: handle_dialogue_response adds server-side range
check (CLOSE_RANGE), matching Talk/Confront pattern (D-010 info
boundary).
Hoshe #2: Weighted selection fallback replaced with unreachable!() —
score_line always returns >= 1, so the fallback was dead code.
Hoshe #3: assert!(false, ...) → panic!() in serialization.rs (clippy).
Hoshe #4: SetFacing and TeleportToHub added to roundtrip test.
Hoshe #5: setup_dialogue_response_world inlined (trivial pass-through).
Tyre #2: Doc comment on DialogueCooldownTracker explains per-player-global
design choice (line IDs are NPC-scoped per D-035, no collision risk).
Tyre #3: CONFRONTATION_LINES comment updated with TODO for D-028/D-035
migration.
Tyre #5: DialogueResponse fixture added for cross-language GDScript
testing (input_dialogue_response.msgpack).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Fix client sprite naming in README (files are _64.png, not .png)
- Clarify --path working directory (renderer/ from repo root)
- Remove /sprite-gen reference (skill not in branch yet)
- Add D-043 citation on outline color in render_export.gd
- Add uid to wall_bar_green.tscn for reproducible imports
- Note wall_bar_green has no rendered sprites yet
- Note npc_generic capsule symmetry is by design (D-044)
- Add outline + D-033 tint compatibility guidance
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Runtime 64px sprites rendered through the new pipeline: generic NPC
(4 directions) and structural wall (4 directions). These unblock
client ticket #540 for sprite camera angle integration.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Update Camera3D to exact D-019 angle (-72.5° from horizontal).
Replace single DirectionalLight with three-point studio rig
(key 1.0, fill 0.4, rim 0.3) for clean silhouettes without
baked shadows. Add generic NPC capsule model (24×32px footprint
per D-044). Document full pipeline spec in renderer/README.md.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Move dialogue system registrations from BridgePlugin to NpcPlugin (#538):
game logic that depends on NPC-layer resources now registers where it
belongs. BridgePlugin retains only wire protocol concerns.
Implement DialogueResponse verb handler (#539): new process_dialogue_response
system runs the full D-028 four-layer pipeline to select follow-up lines
when the player picks a dialogue option. Clears ActiveDialogue when no
candidates remain. Fix latent schedule ambiguity — emit_observation_events
now has explicit .before(advance_tick) constraint.
Verify trust-gated gossip pipeline (#171): confirmed process_talk_interaction
correctly passes KnowledgeConfidence through relationship_to_trust() per
D-075. Added integration tests for Secret-tier access (Friendly+KnowsDetails)
and Surface-only fallback (Friendly+Suspects).
Wire DialogueCooldownTracker into selection (#338): added regression test
confirming no line_id repeats within the 600-tick cooldown window across
10 consecutive Talk interactions.
Closes#538, #539, #171, #338
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Plan 8 tickets across server (4), client (2), copy (1), visual (1).
Focus: two-way dialogue exchange, trust-gated gossip, sprite pipeline,
and housekeeping fixes. No carry-overs from sprint 15.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
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>
All reviewer agents now read source files from worktree paths using
the Read tool instead of git show. Removes Bash dependency for
non-code reviewers and ensures files are always the actual branch
checkout.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- dialogue_box.gd: panel is always visible as permanent insert UI
element per D-061 — content fades but frame stays on screen
- protocol.gd: bump PROTOCOL_VERSION to 13
- Makefile: add check-protocol target that verifies server/client
protocol versions match, runs automatically before build
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
First wireframe set covering all major UI surfaces:
- HUD: basic insert (v0.1), full insert with perception modes (v1.0)
- Dialogue: speaking, choosing, overheard states (v0.1), full with
relationship indicators (v1.0)
- Monologue: idle, reaction (v0.1), thought threading (v1.0)
- Popups: modal, tooltip, interaction list (v0.1), context menu,
notification stack (v1.0)
- Menus: main, pause, save/load (v0.1), main menu, options (v1.0)
README.md cross-references each wireframe to governing D-records.
JSON source + PNG exports tracked in git.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Replace individual API call workflow with JSON-as-truth architecture.
Local wireframe JSON files are the source of truth; Frame0 is treated
purely as a renderer. New frame0-sync.py handles push/pull/export with
stable local ID <-> ephemeral Frame0 ID mapping via .idmap.json files.
Key changes:
- frame0-sync.py: push/pull/export with topo-sorted shape creation
- frame0-export-batch.sh: batch export with staleness check
- frame0-cmd.sh: fixed command names, list-pages JSON bug, added commands
- Removed frame0-wireframe.sh (superseded by JSON workflow)
- SKILL.md: renderer-only guidance, simplified workflow
- api-reference.md: corrected commands, type/color token mapping
- component-library.md: all 6 patterns rewritten as JSON templates
- .gitignore: added *.idmap.json
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
16 d2 diagrams covering architecture (IPC bridge, system tiers,
testability, audio bus), data flow (perception, dialogue, eavesdrop
pipelines), entity (knowledge graph, NPC generation, dialogue tags),
state (simulation tiers, movement stances, knowledge confidence), and
UI (HUD layout, interaction flow, fog layers). PNG output for universal
viewer compatibility. Removes test SVG from font debugging.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Replaces Frame0 MCP server with portable bash+curl scripts that call
Frame0's local HTTP API directly. Includes low-level API wrapper,
high-level composition helpers with project styling, component library
patterns, and setup guide. Requires Frame0 desktop app running locally.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Fix protocol version comment (12 → 13) in ObserverSnapshot doc
- Widen Want intensity range from 3..=9 to 1..=10 to match spec and test
- Replace duplicated pool selection in trigger_recognition_monologue with
call to select_pool_line helper (~50 lines removed)
- Document NaiveSpatialIndex migration cost for grid/quadtree swap
- Remove misleading Default derive from TellCategory (Nervous is not a
sensible default for neutral NPCs)
- Add caller invariant doc on generate_npc (no TilePosition spawned)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Reflects the intentional override — 1200px chosen for readability
after playtest feedback. Updates decision text, derivation, and
amendment history.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The 1200px value was a deliberate readability decision, not a spec
violation. Reverts the incorrect 640px change from round 1. Updates
comment and regression test to match.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Remove stale smoothing re-enable comments from main.gd (Hoshe #1).
Add DIALOGUE_MAX_WIDTH=640 regression test (Hoshe #2).
Extract GROUND_FLOOR const in tile_renderer (Tyre #3).
Clean up entity_renderer migration comment (Hoshe #3).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Regenerate all msgpack test fixtures with tell_state and follow_state
fields. Update golden proof_room snapshot. Adjust serialization and
bridge tests for new ObserverSnapshot fields.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Register all new systems in NpcPlugin and SimulationPlugin with correct
ordering constraints. Protocol bumped to v13: tell_state on VisibleEntity,
follow_state and PostConversationQueue on ObserverSnapshot, Follow verb
on VerbKind. Observer snapshot populates tell state from DerivedTellState
and follow state from FollowTarget. System ordering: tolerance after
mood, deviation after activity, tell after mood+deviation, follow after
visibility geometry, event monologue after conversations.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add observe_npc, hear_sound, observe_anomaly, witness_interaction, and
post_conversation triggers to monologue system. trigger_event_monologue
system fires after sound collection, NPC conversations, and walk-away.
Context tags (location, situation, character_state) populated for D-035
content pool matching. COOLDOWN_TICKS=300 anti-spam guard respected.
witness_interaction fires after overheard NPC-to-NPC conversation per
D-078. Voice sounds and ambient sounds correctly excluded from
hear_sound trigger.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Follow verb on interaction dispatcher. FollowTarget component tracks
target entity, proximity ticks, and LOS-lost ticks. update_follow_state
system: observation events fire at double frequency while following,
NPC suspicion increases via stress when player within 2 tiles for 60+
ticks (configurable). Follow ends on LOS lost timeout, suspicion
threshold crossed, or player issues different action. FollowStateWire
emitted in ObserverSnapshot for client HUD.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
RoutineDeviationEvent emitted when NPC breaks daily routine: wrong
location for day phase or wrong activity at correct location.
CurrentlyDeviating marker prevents duplicate events per episode.
Respects pathfinding-in-progress (no false positives). Feeds
observation event generator for observe_anomaly monologue triggers.
Primary detective mechanic per D-027 criterion 4.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
TellCategory enum (Nervous, Angry, Friendly, Guarded, RoutineDeviation)
with DerivedTellState component. derive_tell_state system runs after
update_mood and detect_routine_deviation. Tell state derived from NPC
axis values per D-024: Secret+low Tolerance→Nervous, low Contentment+
Hostile→Angry, high Contentment+Friendly→Friendly, high Secret→Guarded.
v0.1 renderer is monologue text, not visual animation.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
RoleDefinition struct + generate_npc() seeds all 10 D-024 axes via
SimRng for deterministic generation. Constraint validation: no duplicate
relationship targets, stress < threshold, one routine entry per phase,
no contradictory personality trait pairs. Spawns fully-populated NPC
entity with 2-3 personality traits for tell system (#90).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
ToleranceBreachEvent emitted when NPC stress exceeds per-seed threshold.
ToleranceBreached marker prevents duplicate events per episode, cleared
on recovery. check_tolerance_threshold system runs after update_mood,
integrates with mood FSM to push toward Hostile/Anxious. Background-tier
NPCs excluded. Unblocks #250 (triangle escalation).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Define SpatialIndex trait with entities_in_range, entities_at, and
update methods. NaiveSpatialIndex uses Vec backend with Manhattan
distance. sync_spatial_index system auto-updates from Changed<TilePosition>.
Registered as Bevy resource. Trait abstraction allows grid/quadtree
replacement later without touching callers.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Critical: DIALOGUE_MAX_WIDTH 1200 → 640 to match D-076 spec.
tile_renderer: clarify z = server floor level, not scene z_index.
Add z-filter unit test (tiles at z!=0 must be skipped).
Camera test: is_equal → distance check for float safety, convergence
test frames 40 → 120 for robustness at lower smoothing speeds.
Teleport: remove redundant first snap in _teleport_transition (the
camera block in _process handles it via _teleport_in_progress flag).
entity_renderer: document y-sort bottom-anchor migration path.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Replace Godot built-in Camera2D smoothing with manual lerp using
CAMERA_SMOOTHING_SPEED (8.0) in constants.gd — same exponential
smoothing pattern as entity_renderer.gd. Teleport snap preserved
via _teleport_in_progress flag. D-015 fixed-north camera lock intact.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
