docs: update CLAUDE.md + AGENTS.md to the deployed state (voice subprocess, SSR off, LFS, 276 tests)

AGENTS.md

@@ -9,58 +9,60 @@ A horror NPC chatbot for the **Build Small Hackathon** (HF / Gradio). Track:
 
 **Spaces (deploy mirror-first, both every time):**
 - Personal mirror: https://huggingface.co/spaces/Pabloler21/hollow (remote `space`)
-- **Submission** (must live in the org): https://huggingface.co/spaces/build-small-hackathon/hollow (remote `hackathon`) — ZeroGPU `zero-a10g`, public.
+- **Submission** (must live in the org): https://huggingface.co/spaces/build-small-hackathon/hollow (remote `hackathon`) — ZeroGPU, public.
 
 ## WHERE WE ARE (read first)
-- **Branch `feat/child-redesign`** (off `feat/voice-awareness`), **NOT deployed**. Both Spaces frozen at the pre-voice commit `4deeed5`. **269 tests green.**
-- **Done this redesign:** menu → cinematic **intro** (lore cards) → **game-view rebuilt as a letterboxed foggy scene** (Direction 3b). The child is a **cut-out silhouette** standing in fog (not an arch); corner-drawer collections; conversation as a **fading subtitle transcript** with a labeled two-voice split; **amber = memory / red = cruelty** color language; **whispered in-voice feedback cues**; full **lore** (Caor / the Gaunt / brother Edren); "What it remembers" weaves the memories it stole from you; **O2** scene (edges dissolve to shadow/fog); **F1** stone command-box input; deterministic dev seed; finale pacing/convulse/sigh fixes; an **end screen** (§27 — epitaph + credits over the fog, revealed by an `ended-now` marker).
-- **Next:** (1) live-verify the redesign in `.venv-tts` + Ollama with Pablo; (2) **rework the end screen into layout B** (epitaph as hero, credits demoted to a quiet bottom footer — current §27 layout is unbalanced); (3) replicate the GOOD finale's snappy pacing + **silent convulse beat** to the **loop & bad** finales (only good has it); (4) optional: blur the central frame edges MORE via the scene **mask + lower contrast** (the heavy frame-edge shadow approach FAILED — see gotchas); (5) merge `feat/child-redesign` → `feat/voice-awareness`, then deploy **mirror-first**; (6) deliverables (demo video, LinkedIn, Field Notes) are Pablo's.
+- **DEPLOYED + LIVE on BOTH Spaces** (no longer frozen at `4deeed5`). The full redesign works on the Space — menu → intro → game (**chat + voice + recall + 3 finales + end screen**). **276 tests green.**
+- **Current work branch `feat/game-chrome-polish`** — game-view chrome polish: controls icon-cluster ✅, drawer counts ✅, dialogue band IN PROGRESS. Pending (plan `docs/superpowers/plans/2026-06-15-dialogue-fixes.md`): **center the player line** (right-align collapses in Gradio's user bubble), **remove the dialogue scrim rectangle**, **shrink the input box**.
+- **Next:** finish those dialogue fixes (verify on the Space); Pablo's deliverables (demo video, LinkedIn, Field Notes) link from the README.
 
 ## Commands
 ```bash
-# Tests (no model/GPU needed — tests mock or avoid run_turn). 3.13 .venv is silent (fine).
+# Tests (no model/GPU needed). 3.13 .venv runs silent (no kokoro) — fine.
 ./.venv/Scripts/python -m pytest -q
-./.venv/Scripts/python -m pytest tests/test_memory.py                 # one file
 ./.venv/Scripts/python -m pytest tests/test_app.py::TestRenderTitle   # one class/test
 ./.venv/Scripts/python -c "import app; print('ok')"                   # import smoke check
 
-# Local voice playthrough — PowerShell (&& and inline env-prefix don't work); needs Ollama running.
-$env:HOLLOW_FAST_FINALE='good'; ./.venv-tts/Scripts/python app.py     # good|loop|bad (1/neutral→good/loop)
+# Local voice run — PowerShell (&& and inline env-prefix don't work); needs Ollama. VISUALS are validated on the Space, not here.
+$env:HOLLOW_FAST_FINALE='good'; ./.venv-tts/Scripts/python app.py     # good|loop|bad ; HOLLOW_DEBUG=1 logs per-turn state
 
-# Deploy — mirror-first: push space → wait ~3-4 min (Kokoro/spacy build) → Ctrl+F5, verify → then hackathon.
-git push space main      # then, only after verifying the mirror:
-git push hackathon main
+# Deploy — mirror-first; history is LFS-rewritten so FORCE-push the branch HEAD to main.
+git push --force space HEAD:main        # wait ~3-4 min, Ctrl+F5, verify on the Space, THEN:
+git push --force hackathon HEAD:main
 ```
-No linter/formatter/typechecker. `pytest` + `import app` are the only automated verification. Visual/audio is **human-only** (Pablo, live).
+No linter/formatter/typechecker. `pytest` + `import app` are the only automated checks. **Visual/audio is human-only (Pablo), and verified on the live Space — local rendering differs.**
 
 ## Stack (decided — do not reopen)
-- **UI:** Gradio 6.x `gr.Blocks` + `gr.Chatbot`. Three full-viewport views toggled by visibility: `menu_view` (Fear & Hunger front door, modes Tester/Full + How-to-Play) → `intro_view` (client-driven typewriter lore cards) → `game_view` (the scene). Horror visuals are pure CSS; the `<head>` JS controller (`_HEAD_JS`) owns menu music, the chime→greeting ritual, the intro cards, mute, idle timing, lifting the tone/cue classes onto `#game-view`, and revealing the end screen.
-- **Model:** `Qwen/Qwen3-8B` via `transformers`+`@spaces.GPU(duration=90)` on Space; `qwen3:8b` via Ollama (`localhost:11434`) locally. Switch: `IS_SPACE = bool(os.environ.get("SPACE_ID"))`.
-- **Voice:** Kokoro-82M TTS (`voice.py`, `af_nicole`, speed 0.94, "28% fog" numpy DSP), **CPU**, import-guarded → **silent if unavailable** (3.13 `.venv` has no kokoro; voice lives in 3.12 `.venv-tts`, and on the Space at py3.11). `_clean_for_tts` strips dashes (Kokoro verbalizes them).
-- **State:** plain dict in `gr.State`. **Parsing:** Pydantic v2 + fallback. **Pkg:** `uv`. Space build needs `packages.txt` (espeak-ng) + `python_version: "3.11"` (Kokoro needs <3.13).
-- **Rejected (don't reopen):** Gemma 4 12B (VRAM); Qwen3.5/3.6 (multimodal breaks the loader); LangGraph, Qdrant, cloud APIs, fine-tuning, sponsor-model swap.
+- **UI:** Gradio **6.18.0** (PINNED: README `sdk_version: 6.18.0` + `requirements gradio==6.18.0`). `gr.Blocks` + `gr.Chatbot`. Three full-viewport views by visibility: `menu_view` → `intro_view` (typewriter lore cards) → `game_view` + a top-level `#end-overlay`. **SSR forced OFF** (`os.environ["GRADIO_SSR_MODE"]="False"` at the TOP of app.py — see gotcha). Horror visuals pure CSS; the `<head>` controller (`_HEAD_JS`, via `launch(head=...)`) owns menu music, the chime→greeting ritual, the intro cards, mute, idle, the tone/cue/end markers, AND the Google-Fonts `<link>`.
+- **Model:** `Qwen/Qwen3-8B` via `transformers`+`@spaces.GPU(duration=90)` on Space; `qwen3:8b` via Ollama locally. Switch: `IS_SPACE = bool(os.environ.get("SPACE_ID"))`. Model loads on CPU at import; `.to("cuda")` INSIDE the GPU call.
+- **Voice:** Kokoro-82M (`af_nicole`, speed 0.94, "28% fog" numpy DSP), **CPU**, in a SEPARATE subprocess (`voice_worker.py`, `CUDA_VISIBLE_DEVICES=""`) — MANDATORY on ZeroGPU (gotcha below). `voice.py` drives it over stdio + does the DSP. Import-guarded/silent if kokoro absent; `HOLLOW_VOICE_OFF=1` disables.
+- **Assets:** `assets/*.webp` via **Git LFS** (HF rejects plain-git binaries — Xet policy). **State:** plain dict in `gr.State`. **Parsing:** Pydantic v2 + fallback. **Pkg:** `uv`. Space build: `packages.txt` (espeak-ng); runtime py3.10.
+- **Rejected (don't reopen):** Gemma 4 12B (VRAM); Qwen3.5/3.6 (multimodal breaks loader); LangGraph, Qdrant, cloud APIs, fine-tuning, sponsor-model swap.
 
 ## Submission (Field Guide REQ-06)
 Space in the org + complete README IS the entry (no form). README frontmatter has track + badges + write-up. **Pending:** demo video, one LinkedIn post (both linked FROM README), Field Notes post. Models ≤32B. Badges: 🔌 Off-the-Grid, 🎨 Off-Brand, 📓 Field Notes.
 
 ## File structure
 ```
-app.py        # thread-cap (OMP/OPENBLAS/MKL=4) at top; 3 views (menu/intro/game) + end-overlay; chat() generator;
-              #   _start_turn (stashes msg in state["_pending_msg"], clears box); 3 _play_finale_*; _enter_game/_show_intro;
-              #   _reset / _to_menu; _render_title(state, show_end) (emits ended-now marker on final finale frame);
-              #   _voice_html / _voice_and_pause; _recovered(state); _render_bond; _on_idle; _HEAD_JS controller
-voice.py      # speak(text)->base64 wav|None — Kokoro whisper + locked DSP, CPU, import-guarded; _clean_for_tts; thread-cap
-engine.py     # run_turn + run_turn_stream generators; _EXTRACTION_SYSTEM; IS_SPACE switch; repetition_penalty. Gen+extract in ONE @spaces.GPU
-character.py  # build_system_prompt(...); _WORLD note; OWN_FRAGMENTS (5-line Caor arc); OPENING_LINE; INTRO_CARDS/INTRO_SEQUENCES; FRAGMENT_TONES
-memory.py     # PACING (tester/full); get_tier; apply_update (+sanitize/Hollow-words filter); should_recall; decide_ending (force_ending bypass); style_signal; pick_aware_memory — pure
-finale.py     # finale_steps (loop) / _bad (wound) / _good (redemption) — pure data; good has a SILENT frenzy beat + Edren recovery
-render.py     # render_entity(affinity, mode, seq, tone, cue) — emits .tone-now + .cue-now markers; cut-out silhouette + finale modes;
-              #   render_recovered(fragments_told, claimed) — its arc + the stolen memories; render_treasure(...)
-sting.py      # numpy audio synth (sting/chime/heartbeat/sigh/flatline/menu_loop/type_tick) — HF rejects committed .wav
-styles.css    # §19-22 finale/progress/typing; §23 menu; §24 how-to; §25 intro; §26 GAME SCENE (3b); §27 END SCREEN
-assets/       # hollow_{base,terror,almost,end,rage,peace}.webp + *_cut.webp (rembg alpha silhouettes); intro_*.webp; background.webp
-tests/        # 269 tests: test_{memory,render,finale,schemas,engine,character,app,sting,voice}.py
-docs/superpowers/{specs,plans}/   # approved specs + plans; AGENT-EXECUTION-BRIEF.md (how to execute a plan here)
+app.py        # thread-cap (OMP/OPENBLAS/MKL=4) + os.environ["GRADIO_SSR_MODE"]="False" at top; 3 views + #end-overlay;
+              #   chat() generator; _start_turn (stashes msg in state["_pending_msg"]); 3 _play_finale_*;
+              #   _enter_game/_show_intro; _reset/_to_menu; _render_title(state, show_end); _render_bond; _on_idle;
+              #   _HEAD_JS controller (font <link> + JS); HOLLOW_DEBUG per-turn telemetry; launch(head=, ssr_mode=False)
+voice.py      # speak(text)->base64 wav|None — drives voice_worker.py over stdio, then numpy DSP (_fog); _clean_for_tts
+voice_worker.py # SEPARATE process: Kokoro with CUDA_VISIBLE_DEVICES=""; protocol on a clean dup'd fd (lib noise -> stderr)
+engine.py     # run_turn + run_turn_stream generators; _EXTRACTION_SYSTEM; IS_SPACE switch. Gen+extract in ONE @spaces.GPU
+character.py  # build_system_prompt(...); _WORLD note; OWN_FRAGMENTS (5-line Caor arc); OPENING_LINE;
+              #   INTRO_CARDS (5; tester=[0,4,3] uses a condensed teaching card)/INTRO_SEQUENCES; FRAGMENT_TONES
+memory.py     # PACING (tester/full); get_tier; apply_update (+sanitize/Hollow-words filter); should_recall;
+              #   decide_ending (force_ending bypass); style_signal; pick_aware_memory — pure
+finale.py     # finale_steps (loop) / _bad (wound) / _good (redemption) — pure data; ALL THREE have a SILENT frenzy beat
+render.py     # render_entity(affinity, mode, seq, tone, cue) — .tone-now/.cue-now markers; render_recovered (count);
+              #   render_treasure (count) — drawer headers show "· N"
+sting.py      # numpy audio synth — HF rejects committed .wav
+styles.css    # §19-22 finale/progress/typing; §23 menu; §24 how-to; §25 intro; §26 GAME SCENE; §27 END SCREEN
+assets/       # Git LFS *.webp: hollow_{base,terror,almost,end,rage,peace}{,_cut}.webp; intro_*.webp; background.webp
+tests/        # 276 tests: test_{memory,render,finale,schemas,engine,character,app,sting,voice}.py
+docs/superpowers/{specs,plans}/   # AGENT-EXECUTION-BRIEF.md (how to execute a plan here) + dated plans
 ```
 
 ## State shape (gr.State) — read NEW fields with `.get(...)`
@@ -71,62 +73,60 @@ docs/superpowers/{specs,plans}/   # approved specs + plans; AGENT-EXECUTION-BRIE
   "wounds": list[str],  # cruel quotes (cap 10), shown redacted ▮
   "history": list[dict], "turn": int, "last_recall_turn": int|None,
   "fragments_told": int,        # OWN_FRAGMENTS surfaced (redemption arc)
-  "ended": bool, "ending": str|None,             # "good"|"loop"|"bad"
-  "force_ending": str|None,     # dev seed forces this ending (deterministic)
+  "ended": bool, "ending": str|None, "force_ending": str|None,   # "good"|"loop"|"bad"; dev seed forces
   "msg_lengths": list[int], "last_aware_memory": str|None,
-  "last_activity": float, "idle_count": int,
-  "greeted": bool, "mode": str, # "tester"|"full" pacing
-  "chosen_name": str|None, "named": bool,        # the child gropes toward a name
-  "_pending_msg": str }         # transient — _start_turn stashes the sent msg, chat() pops it
+  "barren_turns": int,          # consecutive turns with no capture (B4 plea); read with .get
+  "last_activity": float, "idle_count": int, "greeted": bool, "mode": str,  # "tester"|"full"
+  "chosen_name": str|None, "named": bool, "_pending_msg": str }   # _pending_msg: _start_turn stashes, chat() pops
 ```
-Dev seed: `HOLLOW_FAST_FINALE=good|loop|bad` (`1`/`neutral`→good/loop) — seeds state AND `force_ending` so the named ending fires deterministically on the next message.
+Env flags: `HOLLOW_FAST_FINALE=good|loop|bad` (`1`/`neutral`→good/loop; seeds state+`force_ending` so the ending fires on the next message); `HOLLOW_DEBUG=1` (per-turn `aff/tone/treasure/claimed/...` log); `HOLLOW_VOICE_OFF=1` (silence).
 
 ## Affinity tiers (base voice; tone modulates on top)
-0–25 Hollow (broken, barely speaks) · 26–50 Curious (repeats you) · 51–75 Too Human (claims memories) · 76–100 Almost (possessive, `almost` face, no sway).
+0–25 Hollow (broken) · 26–50 Curious (repeats you) · 51–75 Too Human (claims memories) · 76–100 Almost (possessive, `almost` face, no sway).
 
 ## Endings (`decide_ending`, turn start, fully scripted, no GPU)
-`force_ending` (dev seed) wins first. Else, one tone-branched gate (no lower threshold races a warm run):
-- **tester:** `bad` at `tone≤−30 ∧ turn≥4 ∧ wounds≥2`; `good/loop` at `aff≥50 ∧ claimed≥2` branched by `tone≥20`.
+`force_ending` (dev seed) wins first. Else one tone-branched gate. Tester is tuned SHORT for judges:
+- **tester:** `bad` at `tone≤−30 ∧ turn≥3 ∧ wounds≥2`; `good/loop` at `aff≥35 ∧ claimed≥2` branched by `tone≥20` (recall_cooldown 1).
 - **full:** `bad` at `tone≤−30 ∧ turn≥6 ∧ wounds≥2`; `good/loop` at `aff≥68 ∧ claimed≥3` branched by `tone≥20`.
-- **bad** (Wound Loop): recites cruelties (redacted→un-redacted), frenzy + triple full-screen stab + scream, reveals rage face, "see you... soon."
-- **good** (Redemption): warmth surfaced its real past; confesses (Caor/the Gaunt), recovers brother **Edren**, RETURNS the treasure (un-claimed, lit), first smile (`peace`→`peace_dissolve`), resolves in peace — NO sequel/quest hook.
+- **bad** (Wound Loop): recites cruelties, SILENT convulse + frenzy/stab/scream, rage face, a **collective lore threat** (Caor/the Gaunt — "we are all still here… we will teach it back to you"), ends "see you... soon."
+- **good** (Redemption): warmth surfaced its past; confesses (Caor/Gaunt), recovers brother **Edren**, RETURNS the treasure (lit), first smile, peace — NO sequel hook.
 - **loop** (Visitor Loop): fed but never loved → stays a predator, consumes you, ends with YOU speaking its opening line.
-- The finale's final frame emits an `ended-now[data-ending]` marker → `_HEAD_JS applyEnd()` reveals the **end screen** (§27) with the per-ending epitaph.
+- The finale's FINAL frame emits an `ended-now[data-ending]` marker → `_HEAD_JS applyEnd()` reveals the **end screen** (§27 layout B: epitaph hero + footer credits + begin-again/leave-the-wood).
 - Prompt tone bands (separate from routing): warm ≥15 / wounded −22..−1 / hostile ≤−22 (injects wounds). Mimicry always on.
 
 ## Lore (canon)
-The child is a dead tithe of **Caor**, a salt-fen hamlet that, in famine, gave **one** child to **the Gaunt** (the old hunger breathing in the wood) to spare the rest. Its family paid the debt; reed-masked bearers bound it at the treeline and left; it starved unburied and became the Gaunt's hook — the thing that waits at the edge and begs memories to fill the hollow where its own used to be. It had a **brother (Edren)** who braided its hair. Recovered as the 5-line `OWN_FRAGMENTS` arc; `_WORLD` note in the system prompt keeps improvised dialogue consistent.
+The child is a dead tithe of **Caor**, a salt-fen hamlet that, in famine, gave **one** child to **the Gaunt** (the old hunger breathing in the wood) to spare the rest. Reed-masked bearers bound it at the treeline and left; it starved unburied and became the Gaunt's hook — the thing that waits at the edge and begs memories. It had a **brother (Edren)** who braided its hair. The **bad ending reveals a collective** — it was not the last the Gaunt was fed; the wood holds many of the given. Recovered as the 5-line `OWN_FRAGMENTS` arc; `_WORLD` note keeps improvised dialogue consistent.
 
 ## Game-view scene anatomy (§26) + end screen (§27)
-- `#game-view` is `position:fixed; inset:0` and defines the **frame geometry as CSS vars** (`--fw/--fh/--band-v/--band-h`). Every in-frame region (drawers, `#game-entity`, `#game-dialogue` subtitle, `#game-groundfog`) is `position:fixed` and anchored to those vars so it sits INSIDE the centered 16:9 frame. `#game-inputbar` is the stone command box at the viewport base, below the frame.
-- Layers: `#game-bg` (blurred forest) → `#game-scene` (sharp forest, radial-mask feathered = O2) → `#game-vig`/`#game-frame-edge` → `#game-entity` silhouette → `#game-groundfog`.
-- **End screen (§27):** `#end-overlay` is a top-level `gr.Column` (`position:fixed; inset:0`), always mounted, `opacity:0/pointer-events:none` until JS adds `.shown`. Epitaph text set per ending by `applyEnd()`; two real `gr.Button`s wired to `_reset` (begin again) and `_to_menu` (leave the wood).
-- **Voice/Presence:** every reply + finale line is spoken (Kokoro). Reply text streams; each finished sentence → a **client-side audio queue** in `_HEAD_JS` (sentences never cut). **Mute is client-side** (the controller owns `{muted}` in localStorage; the server always synthesizes). Awareness: B3 `style_signal`, B2 `pick_aware_memory` (every 3rd non-recall turn), B1 idle (client-timed 60 s → hidden `#idle-trigger` → `_on_idle`, escalating, −3 aff/−2 tone), B4 `_memory_plea_level` (barren turns plead harder). Feedback cues: capture→Treasure chip pulse + "— it keeps this —"; recall→amber line + claimed mark; recover→"— it remembers —".
+- `#game-view` is `position:fixed; inset:0` and defines the **frame geometry as CSS vars** (`--fw/--fh/--band-v/--band-h`). In-frame regions (drawers, `#game-entity`, `#game-dialogue`, `#game-groundfog`) are `position:fixed` anchored to those vars so they sit INSIDE the centered 16:9 frame. `#game-inputbar` is the stone command box at the viewport base.
+- Layers: `#game-bg` (blurred forest) → `#game-scene` (sharp, radial-mask feathered) → `#game-vig`/`#game-frame-edge` → `#game-entity` silhouette → `#game-groundfog`.
+- **End screen (§27):** `#end-overlay` is a top-level `gr.Column` (`position:fixed; inset:0`), always mounted, hidden until JS adds `.shown`. Epitaph set per ending by `applyEnd()`; two real `gr.Button`s → `_reset` (begin again) / `_to_menu` (leave). On dismiss the JS **removes the `.ended-now` marker** so the poll can't re-reveal (was the double-click bug).
+- **Voice/Presence:** every reply + finale line is spoken (Kokoro subprocess). Reply text streams; each finished sentence → a **client-side audio queue** in `_HEAD_JS`. **Mute is client-side**. Awareness: B3 `style_signal`, B2 `pick_aware_memory` (every 3rd non-recall turn), B1 idle (60 s → `#idle-trigger` → `_on_idle`, −3 aff/−2 tone), B4 `_memory_plea_level`. Cues: capture→chip pulse + "— it keeps this —"; recall→amber line; recover→"— it remembers —".
 
 ## Gotchas (DON'T re-learn the hard way)
-- **`position:fixed` is the layout primitive here** (menu/intro/game/end). A `transform`/`filter` on ANY ancestor traps fixed children → they fall into Gradio flow. `#game-stage` has `transform:none`; `#game-entity` centers via `left:calc(50% - var(--fw)*0.15)` (NO transform) so the full-screen `.entity-ghost` finale stab isn't trapped in the narrow column.
-- **Gradio wraps each component (`gr.HTML`, `gr.Row`, `gr.Button`) in a full-width `.block`.** So `align-items:center` on a parent flex centers the *block*, not your inner content, and a `gr.Row` of buttons defaults to spreading them. To truly center an overlay's content, neutralize the wrappers (`width`/`max-width`/`background`/`border`) and override the row/buttons explicitly.
-- **Don't dissolve scene edges with a heavy full-viewport inset shadow** — it darkens the viewport border and leaves a bright-center RECTANGLE (the "cut"). Dissolve via the `#game-scene` radial **mask + low brightness contrast** vs the blur. (210px/90px frame-edge is the accepted ceiling; 250/120 broke it.)
-- **Verify CSS against the RUNNING app, never the static harness.** Headless screenshots are fragile in this shell; `.venv` for fast layout probes, `.venv-tts` + Pablo's eyes/ear for the real playthrough. Audio can only be confirmed by ear. **Do NOT launch/headless-verify as an agent — that step is human-only.**
-- **Gradio replaces the DOM each update:** CSS *animations* restart (the flash/screamer mechanism); *transitions* don't animate across updates. `prefers-reduced-motion`: disable sway/fog only, NEVER opacity (killed the terror flash once).
-- **Cut-out silhouettes:** `hollow_*_cut.webp` (rembg alpha). Child materializes via brightness/opacity + a bottom fade mask into the fog — NO `blur` (reads as broken image), NO `scale`. Finale/flash overlays (`.entity-flick` etc.) MUST use `object-fit:contain`+bottom like the base, or they balloon past the frame.
-- **Finale audio:** one-shot cues (sigh/flatline/sting) carry `class="cue-audio"` (NOT autoplay) → routed through the client voice queue so they don't talk over the playing line; the looping heartbeat stays autoplay. `_voice_and_pause` paces spoken steps to `audio_len + 0.1` (silent beats keep their scripted pause). Good finale splits the convulsion into a spoken line + a **silent `frenzy` beat** so the queue drains and the relief sigh lands exactly on the smile.
-- **Input clears on send via state stash:** `_start_turn` sets `state["_pending_msg"]` + clears the box; `chat()` pops it. Do NOT recover from chatbot history — Gradio normalizes message `content` to a list (`'list' has no attribute 'strip'`).
-- **The opening always renders the base face** (`_enter_game` → `render_entity(20)`), never the seeded Almost tier — it's the vulnerable first contact.
-- **Marker pattern for client state:** `render_entity`/`_render_title` emit hidden `.tone-now[data-tone]` / `.cue-now[data-cue][data-seq]` / `.ended-now[data-ending]` spans; `applyTone()`/`applyCue()`/`applyEnd()` (polled in `_HEAD_JS wire()`) lift them onto `#game-view` or reveal the overlay. Markers are siblings of `#game-bg`, not ancestors — that's why the JS lifts them. Reuse `openingRitual()` etc.; don't redefine existing JS functions. Don't add outputs to the finale generator tuples (tuple-desync hazard) — use the marker.
-- **Qwen emits `+2` (invalid JSON)** without positive anchors → bond froze. Sanitizer in `apply_update` + "no leading +" + dual examples. Keep both.
-- **Recall pollution:** extraction grabbed Hollow's own recalled line as a "new memory". `_is_hollows_words` (>60% overlap dropped) + `repetition_penalty 1.3`. Keep both.
-- **One `@spaces.GPU` per turn** (gen+extract). NO threading around `run_turn` (breaks ZeroGPU); a `TextIteratorStreamer` thread INSIDE the GPU call is the supported stream pattern (verify on the Space; fall back to non-stream if it breaks).
-- **transformers 5.x:** `apply_chat_template(...,return_tensors="pt")`→`BatchEncoding` (`hasattr(t,"input_ids")`); `dtype=` not `torch_dtype`; thinking OFF (`enable_thinking=False` / `"think":False`).
-- **HF rejects committed binaries (.wav):** synth at runtime (`sting.py`). New non-image binaries: same.
-- **Don't edit `.py`/`.css`/`.md` via PowerShell string ops** — mangles UTF-8 (em-dashes, `❯`, accents, emoji). Use the Edit tool.
-- **Cap CPU math threads before torch loads** (`OMP/OPENBLAS/MKL=4`) or OpenBLAS exhausts the CPU allocator → Kokoro silently disabled (NOT real RAM exhaustion).
-- **Backend safety net:** `run_turn` failures → in-character "fog swallowed your words" reply. Msgs >500 chars declined in character. **Ollama 500 = ROCm `cudaMalloc` corruption** (every reply becomes the fallback) → restart Ollama; the Space is immune.
-- **Gradio 6 moved `js=`/`head=` to `launch()`.** Inject JS via `launch(head=...)`. `gr.HTML` initial values are client-rendered (in `/config`, not page HTML).
+- **ZeroGPU: NEVER touch CUDA in the main process.** Any CUDA init in main (even Kokoro/torch on CPU probing `cuda.is_available()`) poisons the `@spaces.GPU` worker fork → `RuntimeError: No CUDA GPUs are available` → EVERY gen falls to the "fog swallowed your words" fallback. Fix: voice (Kokoro) runs in a SEPARATE process (`voice_worker.py`, `CUDA_VISIBLE_DEVICES=""`). In-process patches were NOT enough. Confirm with `HOLLOW_VOICE_OFF=1` (gen recovers → voice was the poison).
+- **Subprocess stdio protocol:** Kokoro/torch/tqdm print to stdout and corrupt a line protocol — `voice_worker.py` dups the real stdout for the protocol and redirects fd 1→stderr (lib noise to stderr).
+- **SSR must be OFF on the Space.** HF defaults Gradio 6 to SSR (Node proxy → app runs in a SUBPROCESS on :7861), breaking the `_HEAD_JS`/CSS client DOM (blank intro, broken layout) AND ZeroGPU. `launch(ssr_mode=False)` is IGNORED when HF imports the app instead of `__main__`; force it with `os.environ["GRADIO_SSR_MODE"]="False"` at the top of app.py.
+- **Fonts: `@import` is rejected on the Space** (not the first rule once Gradio injects its CSS first). Load Google Fonts via a `<link>` in `<head>` (in `_HEAD_JS`), not `@import` in styles.css.
+- **HF rejects plain-git binaries (Xet policy):** webp assets are **Git LFS**. The migration rewrote history, so deploys are **`git push --force`** (`git lfs migrate import --include="*.webp"`; `git stash -u` first — migrate refuses on a dirty tree, even untracked).
+- **Local rendering ≠ the Space** (Gradio version, SSR, font load differ). The Space is the visual source of truth — **verify there, not locally**. Agents must NOT judge visuals locally; that's human-only (Pablo, on the mirror).
+- **Gradio chatbot gives user messages a narrow right bubble** → right-aligning/clamping the player line collapses it to a 1-char vertical column. CENTER it like the bot; distinguish "you" by label + muted color.
+- **Intro typewriter races the server view-flip:** poll for `#intro-text` (rendered after the `_show_intro` round-trip; a fixed `setTimeout` loses on Space latency).
+- **`position:fixed` is the layout primitive** (menu/intro/game/end). A `transform`/`filter` on ANY ancestor traps fixed children. **Gradio wraps each component in a full-width `.block`** — `align-items:center` centers the block, not the content; neutralize wrappers to center an overlay.
+- **Don't dissolve scene edges with a heavy inset shadow** (bright-center rectangle). Use the `#game-scene` radial mask + low brightness (210px/90px ceiling). A dialogue **scrim `::before` also reads as a dark rectangle** — avoid it; rely on text-shadow.
+- **Gradio replaces the DOM each update:** CSS *animations* restart; *transitions* don't animate across updates. `prefers-reduced-motion`: disable sway/fog only, NEVER opacity.
+- **Cut-out silhouettes:** materialize via brightness/opacity + bottom fade — NO `blur`, NO `scale`. Finale overlays use `object-fit:contain`+bottom.
+- **Marker pattern:** `render_entity`/`_render_title` emit hidden `.tone-now`/`.cue-now`/`.ended-now` spans; `applyTone/applyCue/applyEnd` (polled in `_HEAD_JS wire()`) lift them. Don't add outputs to the finale generator tuples (tuple-desync) — use the marker.
+- **Input clears on send via state stash:** `_start_turn` sets `state["_pending_msg"]`; `chat()` pops it. Do NOT recover from chatbot history (Gradio normalizes content to a list → `'list' has no attribute 'strip'`).
+- **The opening always renders the base face** (`render_entity(20)`), never the seeded tier.
+- **Qwen emits `+2` (invalid JSON)** without positive anchors → sanitizer in `apply_update` + "no leading +" + dual examples. **Recall pollution:** `_is_hollows_words` (>60% overlap) + `repetition_penalty 1.3`. Keep both.
+- **One `@spaces.GPU` per turn.** NO threading around `run_turn`; a `TextIteratorStreamer` thread INSIDE the GPU call is the supported stream pattern.
+- **transformers 5.x:** `apply_chat_template(...,return_tensors="pt")`→`BatchEncoding`; `dtype=` not `torch_dtype`; thinking OFF.
+- **Don't edit `.py`/`.css`/`.md` via PowerShell string ops** — mangles UTF-8 (`❯`, `↺`, `✦`, em-dashes, accents, emoji). Use the Edit tool.
+- **Cap CPU math threads before torch loads** (`OMP/OPENBLAS/MKL=4`). **Backend safety net:** `run_turn` failures → "fog swallowed your words"; msgs >500 chars declined. **Ollama 500 = ROCm `cudaMalloc` corruption** → restart Ollama (the Space is immune).
 
 ## Working style & conventions
-- **Everything in English:** code, prompts, UI, dialogue.
-- Phase by phase; explain *why* before code; no new deps/complexity without asking; push back on scope creep.
+- **Everything in English:** code, prompts, UI, dialogue. Phase by phase; explain *why* before code; no new deps/complexity without asking; push back on scope creep.
 - Big changes flow **spec → plan (writing-plans) → execution**. Opus writes specs/plans; the executor implements task-by-task (read `docs/superpowers/AGENT-EXECUTION-BRIEF.md` before executing a plan).
 - **One task/phase per commit**, exact message from the plan + trailer: `Co-Authored-By: Claude <model> <noreply@anthropic.com>` (the model you actually are).
-- Keep the test suite green (`pytest -q`); `import app` must stay OK after every change.
+- Keep `pytest -q` green (276) and `import app` OK after every change. **Verify visuals on the Space, never locally.**

CLAUDE.md

@@ -8,37 +8,41 @@ A horror NPC chatbot for the **Build Small Hackathon** (HF / Gradio). Track:
 **Pitch:** A lost child at the edge of a dead wood has no memories of its own. It asks for yours, stores them in a "treasure", later claims them as its own in the first person (the **recall** — core wow). How you *treat* it branches three endings.
 
 ## WHERE WE ARE (read first)
-- **Branch `feat/child-redesign`** (off `feat/voice-awareness`), **NOT deployed**. Both Spaces frozen at the pre-voice commit `4deeed5`. **268 tests green.**
-- **Done this redesign:** menu → cinematic **intro** (lore cards) → **game-view rebuilt as a letterboxed foggy scene** (Direction 3b). The child is a **cut-out silhouette** standing in fog (not an arch); corner-drawer collections; conversation as a **fading subtitle transcript** with a labeled two-voice split; **amber = memory / red = cruelty** color language; **whispered in-voice feedback cues**; full **lore** (Caor / the Gaunt / brother Edren); "What it remembers" weaves the memories it stole from you; **O2** scene (edges dissolve to shadow/fog); **F1** stone command-box input; deterministic dev seed; finale pacing/convulse/sigh fixes.
-- **Next:** (1) live-verify the redesign in `.venv-tts` + Ollama with Pablo; (2) replicate the GOOD finale's snappy pacing + **silent convulse beat** to the **loop & bad** finales (only good has it); (3) optional: blur the central frame edges MORE via the scene **mask + lower contrast** (the heavy frame-edge shadow approach FAILED — see gotchas); (4) merge `feat/child-redesign` → `feat/voice-awareness`, then deploy **mirror-first**; (5) deliverables (demo video, LinkedIn, Field Notes) are Pablo's.
+- **DEPLOYED + LIVE on BOTH Spaces** (no longer frozen at `4deeed5`): mirror `Pabloler21/hollow` (remote `space`) + submission `build-small-hackathon/hollow` (remote `hackathon`, ZeroGPU). The full redesign works on the Space — menu → intro → game (**chat + voice + recall + 3 finales + end screen**). **276 tests green.**
+- **Current work branch `feat/game-chrome-polish`** — game-view chrome polish: controls icon-cluster ✅, drawer counts ✅, dialogue band IN PROGRESS. Pending (plan `docs/superpowers/plans/2026-06-15-dialogue-fixes.md`): **center the player line** (right-align collapses in Gradio's user bubble), **remove the dialogue scrim rectangle**, **shrink the input box**.
+- **Next:** finish those dialogue fixes (verify on the Space); Pablo's deliverables (demo video, LinkedIn, Field Notes) link from the README.
 
 ## Stack (decided — do not reopen)
-- **UI:** Gradio 6.x `gr.Blocks` + `gr.Chatbot`. Three full-viewport views toggled by visibility: `menu_view` (Fear & Hunger front door, modes Tester/Full + How-to-Play) → `intro_view` (client-driven typewriter lore cards) → `game_view` (the scene). Horror visuals are pure CSS; the `<head>` JS controller (`_HEAD_JS`) owns menu music, the chime→greeting ritual, the intro cards, mute, idle timing, and lifting the tone/cue classes onto `#game-view`.
-- **Model:** `Qwen/Qwen3-8B` via `transformers`+`@spaces.GPU(duration=90)` on Space; `qwen3:8b` via Ollama (`localhost:11434`) locally. Switch: `IS_SPACE = bool(os.environ.get("SPACE_ID"))`.
-- **Voice:** Kokoro-82M TTS (`voice.py`, `af_nicole`, speed 0.88, "28% fog" DSP), **CPU**, import-guarded → **silent if unavailable** (3.13 `.venv` has no kokoro; voice lives in 3.12 `.venv-tts`, and on the Space at py3.11). `_clean_for_tts` strips dashes (Kokoro verbalizes them).
-- **State:** plain dict in `gr.State`. **Parsing:** Pydantic v2 + fallback. **Pkg:** `uv`. Space build needs `packages.txt` (espeak-ng) + `python_version: "3.11"`.
+- **UI:** Gradio **6.18.0** (PINNED: README `sdk_version: 6.18.0` + `requirements gradio==6.18.0`). `gr.Blocks` + `gr.Chatbot`. Three full-viewport views by visibility: `menu_view` → `intro_view` (typewriter lore cards) → `game_view` + a top-level `#end-overlay`. **SSR forced OFF** (`os.environ["GRADIO_SSR_MODE"]="False"` at the TOP of app.py — see gotcha). Horror visuals are pure CSS; the `<head>` controller (`_HEAD_JS`, via `launch(head=...)`) owns menu music, the chime→greeting ritual, the intro cards, mute, idle, the tone/cue/end markers, AND the Google-Fonts `<link>`.
+- **Model:** `Qwen/Qwen3-8B` via `transformers`+`@spaces.GPU(duration=90)` on Space; `qwen3:8b` via Ollama (`localhost:11434`) locally. Switch: `IS_SPACE = bool(os.environ.get("SPACE_ID"))`. Model loads on CPU at import; `.to("cuda")` happens INSIDE the GPU call.
+- **Voice:** Kokoro-82M (`af_nicole`, speed 0.94, "28% fog" numpy DSP), **CPU**. Runs in a SEPARATE subprocess (`voice_worker.py`, `CUDA_VISIBLE_DEVICES=""`) — MANDATORY on ZeroGPU (gotcha below). `voice.py` drives it over stdio and does the DSP in-process. Import-guarded/silent if kokoro absent (3.13 `.venv`); `HOLLOW_VOICE_OFF=1` disables.
+- **Assets:** `assets/*.webp` tracked via **Git LFS** (HF rejects plain-git binaries — Xet policy; see gotcha). **State:** plain dict in `gr.State`. **Parsing:** Pydantic v2 + fallback. **Pkg:** `uv`. Space build: `packages.txt` (espeak-ng); runtime py3.10.
 - **Rejected (don't reopen):** Gemma 4 12B (VRAM); Qwen3.5/3.6 (multimodal breaks the loader); LangGraph, Qdrant, cloud APIs, fine-tuning, sponsor-model swap.
 
 ## Submission (Field Guide REQ-06)
-Space in the org + complete README IS the entry (no form). README frontmatter has track + badges + write-up. **Pending:** demo video, one LinkedIn post (both linked FROM README), Field Notes post. Models ≤32B. Badges: 🔌 Off-the-Grid, 🎨 Off-Brand, 📓 Field Notes. Deploy remotes: `space` (mirror), `hackathon` (org submission).
+Space in the org + complete README IS the entry (no form). README frontmatter has track + badges + write-up. **Pending:** demo video, one LinkedIn post (both linked FROM README), Field Notes post. Models ≤32B. Badges: 🔌 Off-the-Grid, 🎨 Off-Brand, 📓 Field Notes.
 
 ## File structure
 ```
-app.py        # thread-cap (OMP/OPENBLAS/MKL=4) at top; 3 views (menu/intro/game); chat() generator;
-              #   _start_turn (stashes msg in state["_pending_msg"], clears box); 3 _play_finale_*; _enter_game/_show_intro;
-              #   _voice_html / _voice_and_pause; _recovered(state); _render_title/_render_bond; _on_idle; _HEAD_JS controller
-voice.py      # speak(text)->base64 wav|None — Kokoro whisper + locked DSP, CPU, import-guarded; _clean_for_tts; thread-cap
-engine.py     # run_turn + run_turn_stream generators; _EXTRACTION_SYSTEM; IS_SPACE switch; repetition_penalty. Gen+extract in ONE @spaces.GPU
-character.py  # build_system_prompt(...); _WORLD note; OWN_FRAGMENTS (5-line Caor arc); OPENING_LINE; INTRO_CARDS/INTRO_SEQUENCES; FRAGMENT_TONES
-memory.py     # PACING (tester/full); get_tier; apply_update (+sanitize/Hollow-words filter); should_recall; decide_ending (force_ending bypass); style_signal; pick_aware_memory — pure
-finale.py     # finale_steps (loop) / _bad (wound) / _good (redemption) — pure data; good has a SILENT frenzy beat + Edren recovery
-render.py     # render_entity(affinity, mode, seq, tone, cue) — emits .tone-now + .cue-now markers; cut-out silhouette + finale modes;
-              #   render_recovered(fragments_told, claimed) — its arc + the stolen memories; render_treasure(...)
+app.py        # thread-cap (OMP/OPENBLAS/MKL=4) + os.environ["GRADIO_SSR_MODE"]="False" at top; 3 views + #end-overlay;
+              #   chat() generator; _start_turn (stashes msg in state["_pending_msg"]); 3 _play_finale_*;
+              #   _enter_game/_show_intro; _reset/_to_menu; _render_title(state, show_end); _render_bond; _on_idle;
+              #   _HEAD_JS controller (font <link> + JS); HOLLOW_DEBUG per-turn telemetry; launch(head=, ssr_mode=False)
+voice.py      # speak(text)->base64 wav|None — drives voice_worker.py over stdio, then numpy DSP (_fog); _clean_for_tts
+voice_worker.py # SEPARATE process: Kokoro with CUDA_VISIBLE_DEVICES=""; protocol on a clean dup'd fd (lib noise -> stderr)
+engine.py     # run_turn + run_turn_stream generators; _EXTRACTION_SYSTEM; IS_SPACE switch. Gen+extract in ONE @spaces.GPU
+character.py  # build_system_prompt(...); _WORLD note; OWN_FRAGMENTS (5-line Caor arc); OPENING_LINE;
+              #   INTRO_CARDS (5; tester=[0,4,3] uses a condensed teaching card)/INTRO_SEQUENCES; FRAGMENT_TONES
+memory.py     # PACING (tester/full); get_tier; apply_update (+sanitize/Hollow-words filter); should_recall;
+              #   decide_ending (force_ending bypass); style_signal; pick_aware_memory — pure
+finale.py     # finale_steps (loop) / _bad (wound) / _good (redemption) — pure data; ALL THREE have a SILENT frenzy beat
+render.py     # render_entity(affinity, mode, seq, tone, cue) — .tone-now/.cue-now markers; render_recovered (count);
+              #   render_treasure (count) — drawer headers show "· N"
 sting.py      # numpy audio synth (sting/chime/heartbeat/sigh/flatline/menu_loop/type_tick) — HF rejects committed .wav
-styles.css    # §19-22 finale/progress/typing; §23 menu; §24 how-to; §25 intro; §26 GAME SCENE (3b)
-assets/       # hollow_{base,terror,almost,end,rage,peace}.webp + *_cut.webp (rembg alpha silhouettes); intro_*.webp; background.webp
-tests/        # 268 tests: test_{memory,render,finale,schemas,engine,character,app,sting,voice}.py
-docs/superpowers/{specs,plans}/   # latest: 2026-06-15-lore-and-scene-polish, -game-ui-polish, -game-match-mockup
+styles.css    # §19-22 finale/progress/typing; §23 menu; §24 how-to; §25 intro; §26 GAME SCENE; §27 END SCREEN
+assets/       # Git LFS *.webp: hollow_{base,terror,almost,end,rage,peace}{,_cut}.webp; intro_*.webp; background.webp
+tests/        # 276 tests: test_{memory,render,finale,schemas,engine,character,app,sting,voice}.py
+docs/superpowers/{specs,plans}/   # AGENT-EXECUTION-BRIEF.md (how to execute a plan here) + dated plans
 ```
 
 ## State shape (gr.State) — read NEW fields with `.get(...)`
@@ -49,57 +53,60 @@ docs/superpowers/{specs,plans}/   # latest: 2026-06-15-lore-and-scene-polish, -g
   "wounds": list[str],  # cruel quotes (cap 10), shown redacted ▮
   "history": list[dict], "turn": int, "last_recall_turn": int|None,
   "fragments_told": int,        # OWN_FRAGMENTS surfaced (redemption arc)
-  "ended": bool, "ending": str|None,             # "good"|"loop"|"bad"
-  "force_ending": str|None,     # dev seed forces this ending (deterministic)
+  "ended": bool, "ending": str|None, "force_ending": str|None,   # "good"|"loop"|"bad"; dev seed forces
   "msg_lengths": list[int], "last_aware_memory": str|None,
-  "last_activity": float, "idle_count": int,
-  "greeted": bool, "mode": str, # "tester"|"full" pacing
-  "chosen_name": str|None, "named": bool,        # the child gropes toward a name
-  "_pending_msg": str }         # transient — _start_turn stashes the sent msg, chat() pops it
+  "barren_turns": int,          # consecutive turns with no capture (B4 plea); read with .get
+  "last_activity": float, "idle_count": int, "greeted": bool, "mode": str,  # "tester"|"full"
+  "chosen_name": str|None, "named": bool, "_pending_msg": str }   # _pending_msg: _start_turn stashes, chat() pops
 ```
-Dev seed: `HOLLOW_FAST_FINALE=good|loop|bad` (`1`/`neutral`→good/loop) — seeds state AND `force_ending` so the named ending fires deterministically on the next message.
+Env flags: `HOLLOW_FAST_FINALE=good|loop|bad` (`1`/`neutral`→good/loop; seeds state+`force_ending` so the ending fires on the next message); `HOLLOW_DEBUG=1` (per-turn `aff/tone/treasure/claimed/...` log); `HOLLOW_VOICE_OFF=1` (silence).
 
 ## Affinity tiers (base voice; tone modulates on top)
-0–25 Hollow (broken, barely speaks) · 26–50 Curious (repeats you) · 51–75 Too Human (claims memories) · 76–100 Almost (possessive, `almost` face, no sway).
+0–25 Hollow (broken) · 26–50 Curious (repeats you) · 51–75 Too Human (claims memories) · 76–100 Almost (possessive, `almost` face, no sway).
 
 ## Endings (`decide_ending`, turn start, fully scripted, no GPU)
-`force_ending` (dev seed) wins first. Else, one tone-branched gate (no lower threshold races a warm run):
-- **tester:** `bad` at `tone≤−30 ∧ turn≥4 ∧ wounds≥2`; `good/loop` at `aff≥50 ∧ claimed≥2` branched by `tone≥20`.
+`force_ending` (dev seed) wins first. Else one tone-branched gate. Tester is tuned SHORT for judges:
+- **tester:** `bad` at `tone≤−30 ∧ turn≥3 ∧ wounds≥2`; `good/loop` at `aff≥35 ∧ claimed≥2` branched by `tone≥20` (recall_cooldown 1).
 - **full:** `bad` at `tone≤−30 ∧ turn≥6 ∧ wounds≥2`; `good/loop` at `aff≥68 ∧ claimed≥3` branched by `tone≥20`.
-- **bad** (Wound Loop): recites cruelties (redacted→un-redacted), frenzy + triple full-screen stab + scream, reveals rage face, "see you... soon."
-- **good** (Redemption): warmth surfaced its real past; confesses (Caor/the Gaunt), recovers brother **Edren**, RETURNS the treasure (un-claimed, lit), first smile (`peace`→`peace_dissolve`), resolves in peace — NO sequel/quest hook.
+- **bad** (Wound Loop): recites cruelties (redacted→un-redacted), SILENT convulse + frenzy/stab/scream, rage face, and a **collective lore threat** (Caor/the Gaunt — "we are all still here… we will teach it back to you"), ends "see you... soon."
+- **good** (Redemption): warmth surfaced its past; confesses (Caor/Gaunt), recovers brother **Edren**, RETURNS the treasure (lit), first smile (`peace`→`peace_dissolve`), peace — NO sequel hook.
 - **loop** (Visitor Loop): fed but never loved → stays a predator, consumes you, ends with YOU speaking its opening line.
+- The finale's FINAL frame emits an `ended-now[data-ending]` marker → `_HEAD_JS applyEnd()` reveals the **end screen** (§27 layout B: epitaph hero + footer credits + begin-again/leave-the-wood).
 - Prompt tone bands (separate from routing): warm ≥15 / wounded −22..−1 / hostile ≤−22 (injects wounds). Mimicry always on.
 
 ## Lore (canon)
-The child is a dead tithe of **Caor**, a salt-fen hamlet that, in famine, gave **one** child to **the Gaunt** (the old hunger breathing in the wood) to spare the rest. Its family paid the debt; reed-masked bearers bound it at the treeline and left; it starved unburied and became the Gaunt's hook — the thing that waits at the edge and begs memories to fill the hollow where its own used to be. It had a **brother (Edren)** who braided its hair. Recovered as the 5-line `OWN_FRAGMENTS` arc; `_WORLD` note in the system prompt keeps improvised dialogue consistent.
+The child is a dead tithe of **Caor**, a salt-fen hamlet that, in famine, gave **one** child to **the Gaunt** (the old hunger breathing in the wood) to spare the rest. Reed-masked bearers bound it at the treeline and left; it starved unburied and became the Gaunt's hook — the thing that waits at the edge and begs memories. It had a **brother (Edren)** who braided its hair. The **bad ending reveals a collective** — it was not the last the Gaunt was fed; the wood holds many of the given. Recovered as the 5-line `OWN_FRAGMENTS` arc; `_WORLD` note keeps improvised dialogue consistent.
 
-## Game-view scene anatomy (§26)
-- `#game-view` is `position:fixed; inset:0` and defines the **frame geometry as CSS vars** (`--fw/--fh/--band-v/--band-h`). Every in-frame region (drawers, `#game-entity`, `#game-dialogue` subtitle, `#game-groundfog`) is `position:fixed` and anchored to those vars so it sits INSIDE the centered 16:9 frame. `#game-inputbar` is the stone command box at the viewport base, below the frame.
-- Layers: `#game-bg` (blurred forest) → `#game-scene` (sharp forest, radial-mask feathered = O2) → `#game-vig`/`#game-frame-edge` → `#game-entity` silhouette → `#game-groundfog`.
-- **Voice/Presence:** every reply + finale line is spoken (Kokoro). Reply text streams; each finished sentence → a **client-side audio queue** in `_HEAD_JS` (sentences never cut). **Mute is client-side** (the controller owns `{muted}` in localStorage; the server always synthesizes). Awareness: B3 `style_signal`, B2 `pick_aware_memory` (every 3rd non-recall turn), B1 idle (client-timed 60 s → hidden `#idle-trigger` → `_on_idle`, escalating, −3 aff/−2 tone), B4 `_memory_plea_level` (barren turns plead harder). Feedback cues: capture→Treasure chip pulse + "— it keeps this —"; recall→amber line + claimed mark; recover→"— it remembers —".
+## Game-view scene anatomy (§26) + end screen (§27)
+- `#game-view` is `position:fixed; inset:0` and defines the **frame geometry as CSS vars** (`--fw/--fh/--band-v/--band-h`). In-frame regions (drawers, `#game-entity`, `#game-dialogue`, `#game-groundfog`) are `position:fixed` anchored to those vars so they sit INSIDE the centered 16:9 frame. `#game-inputbar` is the stone command box at the viewport base.
+- Layers: `#game-bg` (blurred forest) → `#game-scene` (sharp, radial-mask feathered) → `#game-vig`/`#game-frame-edge` → `#game-entity` silhouette → `#game-groundfog`.
+- **End screen (§27):** `#end-overlay` is a top-level `gr.Column` (`position:fixed; inset:0`), always mounted, `opacity:0/pointer-events:none` until JS adds `.shown`. Epitaph set per ending by `applyEnd()`; two real `gr.Button`s → `_reset` (begin again) / `_to_menu` (leave). On dismiss the JS **removes the `.ended-now` marker** so the poll can't re-reveal (was the double-click bug).
+- **Voice/Presence:** every reply + finale line is spoken (Kokoro subprocess). Reply text streams; each finished sentence → a **client-side audio queue** in `_HEAD_JS`. **Mute is client-side**. Awareness: B3 `style_signal`, B2 `pick_aware_memory` (every 3rd non-recall turn), B1 idle (client-timed 60 s → `#idle-trigger` → `_on_idle`, −3 aff/−2 tone), B4 `_memory_plea_level`. Cues: capture→chip pulse + "— it keeps this —"; recall→amber line; recover→"— it remembers —".
 
 ## Gotchas (DON'T re-learn the hard way)
-- **`position:fixed` is the layout primitive here** (menu/intro/game). A `transform`/`filter` on ANY ancestor traps fixed children → they fall into Gradio flow. `#game-stage` has `transform:none`; `#game-entity` centers via `left:calc(50% - var(--fw)*0.15)` (NO transform) so the full-screen `.entity-ghost` finale stab isn't trapped in the narrow column.
-- **Don't dissolve scene edges with a heavy full-viewport inset shadow** — it darkens the viewport border and leaves a bright-center RECTANGLE (the "cut"). Dissolve via the `#game-scene` radial **mask + low brightness contrast** vs the blur. (210px/90px frame-edge is the accepted ceiling; 250/120 broke it.)
-- **Verify CSS against the RUNNING app, never the static harness.** Headless screenshots are fragile in this shell; `.venv` for fast layout probes, `.venv-tts` + Pablo's eyes/ear for the real playthrough. Audio can only be confirmed by ear.
-- **Gradio replaces the DOM each update:** CSS *animations* restart (the flash/screamer mechanism); *transitions* don't animate across updates. `prefers-reduced-motion`: disable sway/fog only, NEVER opacity (killed the terror flash once).
-- **Cut-out silhouettes:** `hollow_*_cut.webp` (rembg alpha). Child stands free, materializes via brightness/opacity + a bottom fade mask into the fog — NO `blur` (reads as broken image), NO `scale`. Finale/flash overlays (`.entity-flick` etc.) MUST use `object-fit:contain`+bottom like the base, or they balloon past the frame.
-- **Finale audio:** one-shot cues (sigh/flatline/sting) carry `class="cue-audio"` (NOT autoplay) → routed through the client voice queue so they don't talk over the playing line; the looping heartbeat stays autoplay. `_voice_and_pause` paces spoken steps to `audio_len + 0.25` (silent beats keep their scripted pause). Good finale splits the convulsion into a spoken line + a **silent `frenzy` beat** so the queue drains and the relief sigh lands exactly on the smile.
-- **Input clears on send via state stash:** `_start_turn` sets `state["_pending_msg"]` + clears the box; `chat()` pops it. Do NOT recover from chatbot history — Gradio normalizes message `content` to a list (`'list' has no attribute 'strip'`).
-- **The opening always renders the base face** (`_enter_game` → `render_entity(20)`), never the seeded Almost tier — it's the vulnerable first contact.
-- **Tone & cue classes ride a marker pattern:** `render_entity` emits hidden `.tone-now[data-tone]` and `.cue-now[data-cue][data-seq]`; `applyTone()`/`applyCue()` (polled in `_HEAD_JS` `wire()`) lift them onto `#game-view` so the WHOLE scene tints (the markers are siblings of `#game-bg`, not ancestors).
-- **Qwen emits `+2` (invalid JSON)** without positive anchors → bond froze. Sanitizer in `apply_update` + "no leading +" + dual examples. Keep both.
-- **Recall pollution:** extraction grabbed Hollow's own recalled line as a "new memory". `_is_hollows_words` (>60% overlap dropped) + `repetition_penalty 1.3`. Keep both.
-- **One `@spaces.GPU` per turn** (gen+extract). NO threading around `run_turn` (breaks ZeroGPU); a `TextIteratorStreamer` thread INSIDE the GPU call is the supported stream pattern (verify on the Space; fall back to non-stream if it breaks).
-- **transformers 5.x:** `apply_chat_template(...,return_tensors="pt")`→`BatchEncoding` (`hasattr(t,"input_ids")`); `dtype=` not `torch_dtype`; thinking OFF (`enable_thinking=False` / `"think":False`).
-- **HF rejects committed binaries (.wav):** synth at runtime (`sting.py`). New non-image binaries: same.
-- **Don't edit `.py`/`.css` via PowerShell string ops** — mangles UTF-8 (em-dashes, `❯`, accents). Use the Edit tool.
-- **Cap CPU math threads before torch loads** (`OMP/OPENBLAS/MKL=4`) or OpenBLAS exhausts the CPU allocator → Kokoro silently disabled (NOT real RAM exhaustion).
-- **Backend safety net:** `run_turn` failures → in-character "fog swallowed your words" reply. Msgs >500 chars declined in character. **Ollama 500 = ROCm `cudaMalloc` corruption** (every reply becomes the fallback) → restart Ollama; the Space is immune.
-- **Gradio 6 moved `js=`/`head=` to `launch()`.** Inject JS via `launch(head=...)`. `gr.HTML` initial values are client-rendered (in `/config`, not page HTML).
+- **ZeroGPU: NEVER touch CUDA in the main process.** Any CUDA init in main (even Kokoro/torch on CPU probing `cuda.is_available()`) poisons the `@spaces.GPU` worker fork → `RuntimeError: No CUDA GPUs are available` → EVERY gen falls to the "fog swallowed your words" fallback. Fix: voice (Kokoro) runs in a SEPARATE process (`voice_worker.py`, `CUDA_VISIBLE_DEVICES=""`). In-process patches (`is_available=False`) were NOT enough. Confirm with `HOLLOW_VOICE_OFF=1` (gen recovers → voice was the poison).
+- **Subprocess stdio protocol:** Kokoro/torch/tqdm print to stdout and corrupt a line protocol — `voice_worker.py` dups the real stdout for the protocol and redirects fd 1→stderr (lib noise to stderr).
+- **SSR must be OFF on the Space.** HF defaults Gradio 6 to SSR (Node proxy → the app runs in a SUBPROCESS on :7861), which breaks the `_HEAD_JS`/CSS client DOM (blank intro, broken layout) AND ZeroGPU. `launch(ssr_mode=False)` is IGNORED when HF imports the app instead of running `__main__`; force it with `os.environ["GRADIO_SSR_MODE"]="False"` at the top of app.py (gradio reads it in `_resolve_ssr_mode`).
+- **Fonts: `@import` is rejected on the Space** (Gradio injects styles.css after its own, so the `@import` isn't the first rule → "not allowed here"). Load Google Fonts via a `<link>` in `<head>` (in `_HEAD_JS`), not `@import` in styles.css.
+- **HF rejects plain-git binaries (Xet policy):** webp assets are **Git LFS**. The LFS migration rewrote history, so deploys are **`git push --force`** (`git lfs migrate import --include="*.webp"`; `git stash -u` first — migrate refuses on a dirty tree, even untracked).
+- **Local rendering ≠ the Space** (Gradio version, SSR, font load all differ). The Space is the visual source of truth — **verify there, not locally**. Agents must NOT judge visuals locally; that's human-only (Pablo, on the mirror).
+- **Gradio chatbot gives user messages a narrow right bubble** → right-aligning/clamping the player line collapses it to a 1-char vertical column. CENTER it like the bot; distinguish "you" by the label + muted color.
+- **Intro typewriter races the server view-flip:** poll for `#intro-text` (it's rendered after the `_show_intro` round-trip; on the Space the latency loses a fixed `setTimeout`).
+- **`position:fixed` is the layout primitive** (menu/intro/game/end). A `transform`/`filter` on ANY ancestor traps fixed children. `#game-entity` centers via `left:calc(...)` (NO transform) so the full-screen finale stab isn't trapped. **Gradio wraps each component in a full-width `.block`** — `align-items:center` centers the block, not the content; neutralize wrappers to truly center an overlay.
+- **Don't dissolve scene edges with a heavy full-viewport inset shadow** (bright-center rectangle). Use the `#game-scene` radial **mask + low brightness** (210px/90px frame-edge is the ceiling). A dialogue **scrim `::before` also reads as a dark rectangle** — avoid it; rely on text-shadow.
+- **Gradio replaces the DOM each update:** CSS *animations* restart (the flash mechanism); *transitions* don't animate across updates. `prefers-reduced-motion`: disable sway/fog only, NEVER opacity.
+- **Cut-out silhouettes** (`hollow_*_cut.webp`, rembg alpha): materialize via brightness/opacity + bottom fade — NO `blur`, NO `scale`. Finale overlays use `object-fit:contain`+bottom or they balloon.
+- **Marker pattern:** `render_entity`/`_render_title` emit hidden `.tone-now`/`.cue-now`/`.ended-now` spans; `applyTone/applyCue/applyEnd` (polled in `_HEAD_JS wire()`) lift them. Markers are siblings of `#game-bg`. Don't add outputs to the finale generator tuples (tuple-desync) — use the marker.
+- **Input clears on send via state stash:** `_start_turn` sets `state["_pending_msg"]`; `chat()` pops it. Do NOT recover from chatbot history (Gradio normalizes content to a list → `'list' has no attribute 'strip'`).
+- **The opening always renders the base face** (`render_entity(20)`), never the seeded tier.
+- **Qwen emits `+2` (invalid JSON)** without positive anchors → sanitizer in `apply_update` + "no leading +" + dual examples. **Recall pollution:** `_is_hollows_words` (>60% overlap) + `repetition_penalty 1.3`. Keep both.
+- **One `@spaces.GPU` per turn.** NO threading around `run_turn`; a `TextIteratorStreamer` thread INSIDE the GPU call is the supported stream pattern.
+- **transformers 5.x:** `apply_chat_template(...,return_tensors="pt")`→`BatchEncoding`; `dtype=` not `torch_dtype`; thinking OFF.
+- **Don't edit `.py`/`.css`/`.md` via PowerShell string ops** — mangles UTF-8 (em-dashes, `❯`, `↺`, `✦`, accents, emoji). Use the Edit tool.
+- **Cap CPU math threads before torch loads** (`OMP/OPENBLAS/MKL=4`). **Backend safety net:** `run_turn` failures → "fog swallowed your words"; msgs >500 chars declined. **Ollama 500 = ROCm `cudaMalloc` corruption** → restart Ollama (the Space is immune).
 
 ## Workflows
-**Deploy** (repo root, **mirror-first**): `git push space main` → wait ~3-4 min (Kokoro/spacy build) → Ctrl+F5, verify build+voice → only then `git push hackathon main`. Commit trailer: `Co-Authored-By: Claude <model> <noreply@anthropic.com>`.
-**Local voice test:** `$env:HOLLOW_FAST_FINALE='good'; ./.venv-tts/Scripts/python app.py` (PowerShell; `&&` and inline env-prefix don't work) with Ollama running. The 3.13 `.venv` is silent (fine for tests).
-**Working style:** phase by phase; explain *why* before code; everything **English** (code/UI/dialogue); no new deps/complexity without asking; push back on scope creep. Big changes via spec → plan (writing-plans) → subagent execution (Sonnet 4.6 has executed the last plans; Opus writes the plans).
+**Deploy** (repo root, **mirror-first**, history is LFS-rewritten so **force-push** the branch HEAD to `main`):
+`git push --force space HEAD:main` → wait ~3-4 min (model+Kokoro build) → Ctrl+F5, verify on the Space (build + chat + voice + layout) → only then `git push --force hackathon HEAD:main`. Commit trailer: `Co-Authored-By: Claude <model> <noreply@anthropic.com>`.
+**Local:** `pytest -q` (276) + `import app` are the only automated checks; the 3.13 `.venv` runs silent (no kokoro). `$env:HOLLOW_FAST_FINALE='good'; ./.venv-tts/Scripts/python app.py` for a local voice run (PowerShell; `&&`/inline env-prefix don't work) — but **visuals are validated on the Space, not local**.
+**Working style:** phase by phase; explain *why* before code; everything **English**; no new deps/complexity without asking; push back on scope creep. Big changes via spec → plan (writing-plans) → executor (Opus plans; an executing agent implements task-by-task — read `docs/superpowers/AGENT-EXECUTION-BRIEF.md`).