Invoker HP and Mana Orbs

Summary

Built the HP and Mana orbs from scratch — went from no resource HUD at all to a fully shader-driven dual-orb display with 1D wave-sim surfaces, dithered cloud-noise interiors, scale + rotation hit feedback, and per-orb visual tuning. The orbs now sit in the bottom-left and bottom-right corners with [, ], ,, . debug binds for damaging and refreshing them.

Orientation + initial direction:

Read the previous session log (orblike-to-invoker.md) plus design.md and plan.md to understand Phase 1 status. Sub-phases 1-3 (primitives, cast HUD, spell catalog) were done; sub-phase 6 (HP / Mana orbs) was untouched.
User picked "HP + mana orbs first" over the alternative options (Fireball-first, tutorial-first), with mouse-cursor aim queued for later.
Surfaced the design conflict between design.md's "ARPG corner orbs" spec and mockup_ui_kit_1.html's top-bar stat-chip canon. User said "try both styles."

First HUD experiment — corner orbs vs top-bar chips:

Built ui_resource_orb (PoE-style stencil-clipped liquid disc) and ui_stat_chip (compact ♥ 100/100 panel) as side-by-side comparison primitives in ui.lua.
Added hp/hp_max/mp/mp_max = 100/100 to run_new() in run.lua.
Hit a bad argument #5 to 'circle' error on the stencil mask draw — the locked-palette colors are tables that need to be called (color()) to pack into rgba ints for the engine. Switched color(255,255,255) to white().
User chose: drop the orbs, move the chips to bottom-left stacked vertically (HP on top, MP below, 8 px margin, 2 px gap between).

Pivot back to orbs (with shader):

User: "let's try the orbs again, orblike_hud has a visual example. Try to make it look like liquid with a shader instead of just a flat circle."
Searched E:/a327ex for a "something something wave" project the user remembered for a 1D wave sim — no folder/code matched (only next_wave enemies in BYTEPATH, Waver.lua, shockwave shaders, and the shake.lua modules in every project). User said they'd find the archive.
Wrote assets/orb_liquid.frag: dedicated orb_layer = layer_new('orbs') rendered between game and UI, full-pass post-process shader that procedurally renders up to 2 orbs from per-frame uniforms.
First shader version had directional brightness gradient + specular highlight band + edge vignette. User: "the gradients you have for the liquid feel a bit incongruent... the game only has gradients for the walls but they're way more subtle and over longer distances. The various polychrome, etc effects are also noise that's more subtle than this."

SNKRX flat-color rewrite:

Read reference/snkrx_style_effects.md design philosophy: flat colors, no gradients, layered springs, rectangles not sprites.
Stripped the directional gradient + specular + edge vignette. Added subtle organic cloud noise lifted from wall_gradient.frag's organic_field function (3-term sin/cos sum at different frequencies + phase offsets).
Added a thin lighter strip directly on the waterline so the surface reads as a discrete pixel line.
Border ring kept as 1.5-px constant-thickness ring in darkened base color.

Edition noise mode catalog:

Wired u_noise_mode (cycled by n key — F11 conflicted with fullscreen on user's setup) through 6 modes from the wall_gradient edition palette:
- 0: static RGB scaled by organic_field
- 1: HSL lightness modulation
- 2: HSL saturation modulation
- 3: HSL L+S combined
- 4: two-octave organic_field (mixes a fast small-scale field into the main one)
- 5: hue-locked dynamic shimmer
First version had two bugs: (a) int uniform via layer_shader_set_int wasn't binding cleanly — switched to float-cast-to-int matching wall_gradient.frag's u_edition convention; (b) noise looked completely static across the orb because organic_field was sampled at pix * 0.15 (works at wall scale, but only 0.43 cycles across a 56-px orb). Bumped to pix * 1.2 for ~3 visible noise lumps.
User selected mode 0 (static RGB) for HP and mode 5 (hue shimmer) for Mana.

Animated drift:

User: "Can we have the noise move instead of being static for those?" — organic_field's internal time divisors (/143.634, /99.4324 etc.) drift extremely slowly at orb scale. Added explicit XY drift vec2(u_time * 10.0, u_time * 6.0) to the field input so the cloud pattern slides at ~8 px/sec horizontally, 5 px/sec vertically.

1D wave simulation port (tidal_waver.lua:130-185):

User added reference/tidal_waver.lua to the project. Found the splash function at line 240: arena:splash(f, x, width) perturbs width adjacent springs at once (3-5 wide) and sets (not adds) their velocity to f. Comment: "The width represents how many springs will be pulled at the same time, as that creates a more realistic water effect."
Wrote waves.lua: 16 spring-points per orb, packed into 4 vec4 uniforms for the shader (u_orbN_surface_0..3).
Per-frame: spring physics step (a = -k*x - d*v), then sub_iters passes of neighbor velocity coupling (spread * (self.x - neighbor.x) injected into neighbor velocity).
Endpoints clamp instead of wrap (waves bounce off "the orb wall").
Shader gained surface_offset_px helper that unpacks the 4 vec4s and lerps between the two nearest points based on (local_x + 1) * 0.5 * 15. Replaced the placeholder sine ripple with surface_disp_px / radius.
Added waves_splash(w, t01, force, width) matching tidal_waver's set-not-add semantics.

Wave tuning iterations:

Round 1: f=8, bounce=0.65, spread=0.45, sub_iters=4, impulse 80. User: "the current bounce is too weak and not visible at all." (Predicted peak displacement ~0.8 px on 28-px-radius orb — math worked out: peak ≈ v / (2π·f).)
Round 2: f=3, bounce=0.85, spread=0.50, sub_iters=6, impulse 150. Big visible improvement.
Round 3: User: "Higher initial impulse, lower bounce, slower spread." → impulse 350, bounce 0.60, spread 0.20.
Round 4: User: "Should impulse up always, the wave should spread across the orb's surface, such that each impulse should always move the whole orb in some way." → f=2, bounce=0.85, spread=0.55, sub_iters=8, width=6.
Round 5: User: "initial impulse is too big" → 400 → 200.
Round 6: User: "0.75 bounce, 0.4 spread coupling".
Round 7: User: "the impulse doesn't look like it's spreading properly... I think there was some additional way I made it spread." Found arena:splash's width parameter — switched to waves_splash(width=6) with set-not-accumulate semantics. Final values: f=2 Hz, bounce=0.75, spread=0.40, sub_iters=8, impulse 200, width 6.

Hit / refresh model with scale + rotation springs:

User wanted a orb_hit(orb, location, damage_pct) function: damage at a specific surface location, with wobble + rotation scaled to damage %. Refresh should be smaller impulse, no scale/rotation.
Created orb.lua wrapping a waves state + a spring container with two springs (main for scale at rest 1.0, rotation for rotation at rest 0).
orb_hit(o, t01, damage, max_value): wave splash at t01 with force scaled by damage/max_value, scale-spring impulse, rotation-spring impulse with random direction.
orb_refresh(o, t01, refresh, max_value): only a small upward wave impulse, no springs.
Updated run_reset to call orb_reset on each.
Final base values: wave impulse -200 * pct (width 6), scale displacement -0.3 * pct (max 30% squish at full damage, was 0.5 — reduced after user feedback "scale can be a bit less strong on high percentages"), rotation kick ±0.7 rad * pct (~40° max, was 0.4 — bumped after user feedback "rotation should be a bit more obvious").

Position to bottom corners:

Moved orbs from screen center (width/2 ± 50, height/2) to corners: HP at (36, height-36), MP at (width-36, height-36) with radius 28 and 8 px margin.
Removed the bottom-left HP/MP comparison chips (replaced by orbs).
Verified layout: bottom-left strip x=0..80 (cast HUD's primitive HUD starts at 80) — HP orb spans 8..64 with 16 px margin to primitive HUD. Bottom-right strip x=316..480 — MP orb spans 416..472.

Border color matching UI:

Switched orb border from hardcoded base * 0.30 (darkened orb hue) to u_border_color uniform set from the locked-palette panel_border (56, 60, 65) so orbs match every other UI element's chrome.

Color muting + per-orb tuning:

Added c key to cycle global color-mute variants (sat -25%, lit -12% per step, 4 variants) and v key to cycle mana hue-shimmer strength (1.0× → 0.5× → 0.25× → 0×).
Top-of-screen labels showed active variant indices.
User picked: HP halfway between full bright and one mute step (mute_sat = 0.875, mute_lit = 0.94), MP unchanged at full brightness, mana hue shimmer locked at × 0.25.
Removed cycle keys + labels, hardcoded 0.25 strength for mode 5, switched mute uniforms from global to per-orb (u_orbN_mute_sat, u_orbN_mute_lit).

Bayer dither:

User: "the orb effect visual itself looks a bit too off compared to how the rest of the game will look. Can we try something like a dither effect on the noise?"
Added 4×4 Bayer matrix (BAYER4) and bayer4(pix) helper. Quantized field to 4 discrete levels per pixel: field = clamp(floor(field*4 + bayer), 0, 3) / 3. The eye averages it into the same gradient at distance, but each pixel snaps to one of 4 discrete colors up close.
Created the SNKRX/Balatro pixel-art crosshatch look the user wanted.

Damage values randomized for testing:

User: "Give me random values for the hit instead of just 10%, since I wanna see what higher/lower values look like too."
[ and , keys: math.random(5, 100) damage (visible effect range from subtle to lethal). ] and . keys: math.random(5, 40) refresh.
Visual effects use intended magnitude even if HP/MP can't actually drop that much (so a 90-damage hit on 30 HP still gives the full ~45° rotation kick).

Text chrome:

Iterated through several positions for the "100" + "/ 100" readout. Color of "/ 100" changed from text_muted to white (palette token, 200,200,200) per user request.
Final spacing: big (header_l) at cy - 12, small (body) at cy + 2. Bounding boxes overlap by 2 rows but visible glyphs come out tight due to font padding. Text block sits a few px below orb center.
Wrapped chrome draws in layer_push(layer, cx, cy, rotation, scale, scale) so text wobbles + tilts with the orb on hits.

Rotation/scale applied to shader interior:

User: "because it's an orb and it wasn't rotating the contents inside I wasn't seeing the effect. Change it back and apply the rotation/scale effect to the contents of the orb too."
Reverted rotation spring (f=8, bounce=0.5 — fast snap-back).
In sample_orb, after the rotation step, computed noise_uv = center + local * radius and passed to apply_noise instead of pix. Math: at rest (rotation=0, scale=1), local * radius = d so noise_uv = pix — identical to before. On a hit, local is rotated and the noise + dither rotate/scale with the disc.
Mode 4's pix * 2.8 in apply_noise hadn't been updated → shader compile error undefined variable "pix". Fixed by switching to noise_uv * 2.8.

Brazilian keyboard fix:

User pointed out ; and ' shift around between keyboard layouts. Switched mana keys to , and . (universal across all layouts including Brazilian ABNT).

Files created/modified this session:

New: waves.lua (1D surface wave simulation), orb.lua (per-orb container + hit/refresh model), assets/orb_liquid.frag (procedural orb shader)
Modified: main.lua (orb_layer, orb_liquid_shader, hp_orb/mp_orb globals, debug binds, render wiring), ui.lua (ui_orbs_apply, ui_orbs_chrome, removed ui_resource_orb + ui_stat_chip), run.lua (hp/mp fields + waves_reset → orb_reset), design.md and plan.md (orb spec updated, sub-phase 6 status reflected in "Shipped").

Phase 1 status at session end:

Done: Sub-phases 1, 2, 6 (HP / Mana orbs).
Partial: Sub-phase 3 (catalog done, dispatch placeholder).
Partial: Sub-phase 5 (mana display orb done; pool/regen/cost not wired).
Not started: 4, 7, 8, 9, 10.

Next session: real spell-effect dispatch (probably Fireball as discussed at the start), or Mana cost gating now that the visual orb exists, or the tutorial map content.

Let's continue with Invoker. Read the summary of last session (Orblike -> Invoker) plus documents to get oriented and suggest next steps.

I'll get oriented by reading the core Invoker docs and finding the last session summary.

Read (E:\a327ex\Invoker\design.md)

1    # Invoker — Design
2   
3   What this game is, at a high level. Pair with `spells.md` for spell-system details and `plan.md` for the implementation roadmap.
4   
5   ---
6   
7   ## Concept
8   
9   Roguelike action game built on two novel-but-simple systems:
10  
11  1. **Live combinatorial spell-casting.** Player chords primitives (elements + formations) into spells in real time. Casting is a 3-key sequence; the same vocabulary recombines to produce dozens of distinct spells. No inventory, no orbs, no equipment — the build is which primitives you have, what modifiers are attached to them, and what passives + items you've accumulated.
12  
13  2. **Minesweeper-style meta-dungeon.** Between combats, the player navigates a procedurally generated grid of rooms with classic Minesweeper rules. Some rooms are mines (HP loss), some are combat encounters, some are shops or forges or treasure. Adjacent-mine numbers help deduce safe paths. The dungeon view replaces the older walkable hideout entirely.
14  
15  A run = beat the dungeons → reach the final boss. Death wipes everything (no meta-progression except a tutorial-skip flag).
16  
17  This is a **practice project**. Big-project ambition lives in `Psyonia/` — the Tales-of-Psyonia framing is deferred to a future production. When in doubt, cut rather than expand.
18  
19  ---
20  
21  ## Core loop
22  
23  1. Spawn into the **dungeon view** (Minesweeper grid).
24  2. Move room-to-room. Adjacent-mine numbers help you deduce safe paths.
25  3. Enter rooms:
26     - **Combat** — fight an encounter with your chord-cast spells.
27     - **Shop** — spend gold on new primitives / modifiers / passives / items.
28     - **Forge** — spend resources to upgrade what you have.
29     - **Treasure** — free reward.
30     - **Mine** — HP loss.
... [119 more lines]

Read (E:\a327ex\Invoker\plan.md)

1    # Invoker — Plan
2   
3   Implementation roadmap for the new direction (chord-cast spells + Minesweeper meta-dungeon). Pair with `design.md` for what we're building and `spells.md` for spell-system mechanics.
4   
5   ---
6   
7   ## Strategy
8   
9   **Tutorial-first.** Build the spell-cast tutorial and the dungeon-meta tutorial before any other content. The tutorial is the smallest complete game — if we can't teach it cleanly, the design has problems and we discard early rather than after building the rest. Once both tutorials work end-to-end, all the core systems exist; everything after is content expansion + lifecycle + polish.
10  
11  **Validate twice.** Two hard validation gates:
12  1. After Phase 1: does chord-casting feel satisfying? If no → pivot the casting design.
13  2. After Phase 2: does the Minesweeper meta feel engaging? If no → pivot the meta design.
14  
15  Either pivot is far cheaper at the gate than after building enemies, leagues, save systems, and content for the (now wrong) design.
16  
17  **Pacing.** Incremental. One sub-deliverable at a time. User tests between deliverables. No chaining without check-ins.
18  
19  ---
20  
21  ## Current state
22  
23  **Shipped (still relevant after the unwind):**
24  - v2 starter on Anchor 2 (`main.lua`, `anchor/` framework copy)
25  - Player movement (WASD)
26  - Wall pipeline + palette/edition cycling system
27  - Hideout + tutorial corridor map shapes (geometry only — entity content is from the old orbs design and will be replaced)
28  - UI primitive substrate (`ui.lua`): `ui_panel`, `ui_text`, `ui_slot`, `ui_item_icon`, drag-drop framework
29  - F2 debug screen, F3-F5 memory tracker, F9-F10 palette cycling
30  - Per-run state lifecycle (`run.lua`) — schema reset to placeholders for the new spell system
... [190 more lines]

Let me find the most recent session summary and check the current state of main.lua.

Bash (ls -t E:/a327ex/anchor.blot.im/logs/ 2>/dev/null | head -10; echo "---"; ls -t E...)

emoji-aimer-1.md
anchor-app-phase-1-part-1.md
website-redesign.md
orblike-redesign.md
orblike-ui-2.md
orblike-ui-1.md
codex-tests.md
codex-setup.md
anchor-app-ui-discussion.md
orblike-tutorial-map-and-wall-layers.md
---
emoji-aimer-5.md
orblike-to-invoker.md
emoji-aimer-4.md
anchor-app-phase-1-part-3.md
emoji-aimer-3.md
anchor-app-phase-1-part-2.md
emoji-aimer-2.md
codex-tests.md
emoji-aimer-1.md
kimi-code-setup.md
---

Read (E:\a327ex\a327ex-site\logs\orblike-to-invoker.md)

File content (106485 tokens) exceeds maximum allowed tokens (25000). Use offset and limit parameters to read specific portions of the file, or search for specific content instead of reading the whole file.

Read (E:\a327ex\a327ex-site\logs\orblike-to-invoker.md)

1    Title: Orblike -> Invoker
2   Date: 2026-05-12 20:21:19
3   
4   # Orblike -> Invoker
5   
6   ## Summary
7   
8   Renamed the practice project from Orblike to Invoker (it's no longer about orbs — built on the Invoker/Magicka chord-cast mechanic), then spent the rest of the session building out the cast HUD substrate: cast slots, primitive slots, key labels, modifier hover, info text, spell catalog, and a Space-key repeat replay.
9   
10  **Project rename — Orblike to Invoker:**
11  - Renamed GitHub repo via `gh repo rename Invoker`; updated local `origin` URL to `git@github.com:a327ex/Invoker.git`.
12  - Renamed local folder `E:\a327ex\Orblike\` → `E:\a327ex\Invoker\` via PowerShell (MINGW bash held a stale handle on the old path even after `cd`, so `mv` failed with "Device or resource busy" until I switched to `Rename-Item`).
13  - Updated text references in active code files alongside `main.lua` (window title in `main.lua`, header comments in `ui.lua`/`wall_patterns.lua`/`wall_symbols.lua`, doc headers in `design.md`/`plan.md`/`spells.md`).
14  - Per user instruction, did NOT update `reference/` subfolder docs/assets (PNG filenames like `orblike_ui_kit_*.png` still on disk).
15  - Swept Claude's working context: `E:\a327ex\.claude\CLAUDE.md`, `MEMORY.md` plus all per-memory feedback files, `Anchor2/reference/anchor2_plan.md`, `reference_disciplines.md`, `reference_taxonomy.md`, `arena/ARENA_PROGRESS.md`, `arena/main.lua`, `docs/FRAMEWORK_API.md`.
16  - Renamed memory file `feedback_orblike_anchor_divergence.md` → `feedback_invoker_anchor_divergence.md` and fixed the link in MEMORY.md.
17  - Updated `end-session/SKILL.md` project table row to point at `invoker` / `E:/a327ex/Invoker`.
18  - Flagged but did NOT touch `.claude/commands/cross-check.md` and `.claude/skills/add-ability/skill.md` — both reference the deprecated tags/interactions/abilities system and are functionally dead for the new spell design.
19  - Flagged the Psyonia folder (`Psyonia/orblike.md` and related) as separate-decision territory since "future Invoker" there is a Psyonian-era reskin, not the practice game.
20  
21  **Cast HUD substrate — three central slots:**
22  - Built `ui_cast_hud(lyr, center_x, bottom_y, chord)` in `ui.lua` using the existing `ui_panel` chrome primitive.
23  - Initial design was 3 slots in world space below the player at `(p.x, p.y + 16)` on `game_layer` inside `camera_attach`.
24  - Pivoted to screen-space bottom-center on `ui_layer` at `(width/2, height - 8)` per user request — "fixed to bottom center of the screen instead of below player".
25  - **Critical bug fix**: `layer_image` centers on `(x, y)` per the C engine (`anchor.c:4328`: "Image is centered at (x, y)"). My original code passed `-img.width/2, -img.height/2` as offset (assuming top-left semantics), causing icons to render at world `(0, 0)` instead of inside the slot. Discovered by reading `process_sprite` source.
26  - Slot geometry iterations: 16w×24h → 44×52 → 22×26 (half) → 30×36 (slightly bigger). Settled at 30×36, with PNG icons scaled to fit (≈23×34 at native 40×60 source).
27  - Made the HUD **permanent** — three slot frames always render, even when chord is empty.
28  - Added per-slot springs (`cast_hud_springs[1..3]`) with pulse on chord insertion. Spring wraps slot chrome + icon via outer `layer_push(scale=spring.main.x)`.
29  - Label panel above each slot showing primitive name. Iterations on font: FatPixel 8 (chunky) → Lana 11 (final).
30  - Label color: started at `text` (240,240,235 ≈ white) → switched to `text_muted` (140,140,145 light grey) for both cast labels and primitive-slot keycaps. User wanted consistent "light grey instead of white" treatment.
... [20 more lines]

Bash (wc -l E:/a327ex/a327ex-site/logs/orblike-to-invoker.md)

4634 E:/a327ex/a327ex-site/logs/orblike-to-invoker.md

Read (E:\a327ex\a327ex-site\logs\orblike-to-invoker.md)

50   **Action log (`log.lua`) — built then disabled:**
51  - Bottom-left scrolling action log. 5 visible lines (down from 6 after user feedback).
52  - Per-entry animated `rank` (float) tweens toward `target_rank` (int) via critically-damped exponential ease (`TWEEN_RATE = 20`).
53  - On `log_add`, every existing entry's target shifts up by 1; new entry enters at `rank = -1` with target 0 (slides up from below).
54  - Old entries past `max_visible` tween off the top and fade out; removed when settled.
55  - Alpha gradient: `VISIBLE_FADE = 0.2` (newest=1.0, oldest visible=0.8) — most of the fade burns off during scroll-out per user spec.
56  - Font research: user mentioned "Ark Pixel 10px monospaced". The `ark.ttf` files in `super-emoji-*` reference projects turned out to be "Ark Pixel 12px proportional lat" (verified via `PrivateFontCollection.AddFontFile`). Found the actual file the user wanted: `fusion-pixel-10px-monospaced-latin.ttf` (sister font, same maintainer). Copied to assets, registered as `ui_font_log`.
57  - **Disabled at session end**: user said "keep this log in the code but comment it out, maybe save it to its own file". Commented out at 6 call sites tagged `[log]` (require, font register, log_font assignment, log_update, log_draw, log_add in cast.lua, log_reset in run.lua). `log.lua` file kept intact.
58  
59  **Info text (`info_text.lua`) — floating damage-number-style:**
60  - Spawned by `cast_attempt` to announce spell name (success) or "Fizzle" (failure).
61  - World-space, anchored to spawn position — doesn't follow player after spawn.
62  - Floats up 22 px over 0.9 s lifetime; alpha holds full for first 50%, then linear fade to 0.
63  - Dark `bg`-colored rect with 0.85 base alpha behind text. Content-fit width. Text in effect color.
64  - Drawn on `game_layer` inside `camera_attach` block (above player draw).
65  - `info_text_reset()` clears on `run_reset`.
66  - Error word for fizzles chosen: **"Fizzle"** (D&D heritage, evocative).
67  
68  **Spell catalog (`spells.lua`):**
69  - Canonical-chord-to-spell lookup. `spell_canonical_chord(chord)` sorts ids and joins with commas; `spell_lookup(chord)` returns the entry or nil.
70  - Initial scope: all 36 Fire-section combinations transcribed from `spells.md`, plus Airborne (Wind+Self), Flicker Strike (Wind+Self+Line), Dash (Force+Self).
71  - Narrowed at user request: "The only element I want is fire, no wind force or anything else." Dropped the 3 non-Fire entries, dropped Wind/Earth/Water/Force from `primitives.defs`. Catalog is now strictly Fire combinations.
72  - Multiple chord paths can map to the same name when spells.md declares them so (Fire+Fire and Fire+Fire+Self are both "Fire Pillar" — the slot/behavior modifier doesn't change the display name).
73  
74  **Primitive slot system (left of cast HUD):**
75  - 4 owned-primitive slots replacing the previous "primitives_known/primitive_keys" model.
76  - `PRIMITIVE_SLOT_COUNT = 4`, `PRIMITIVE_SLOT_KEYS = {'q','e','r','t'}` (slot index → key, fixed).
77  - `run.primitive_slots[1..4]` — nil means empty. Pressing Q/E/R/T with empty slot is a no-op.
78  - `primitives_acquire(prim_id)` fills the earliest empty slot. Returns false if unknown/already-owned/all-full.
79  - Debug binds: key:1 acquires Fire, key:2 acquires Projectile (debug-only; real pickup events replace later).
... [170 more lines]

I have the session summary. Now let me check the current state of the actual code files.

Bash (ls E:/a327ex/Invoker/*.lua E:/a327ex/Invoker/assets/ 2>&1 | head -40)

E:/a327ex/Invoker/cast.lua
E:/a327ex/Invoker/door.lua
E:/a327ex/Invoker/info_text.lua
E:/a327ex/Invoker/log.lua
E:/a327ex/Invoker/main.lua
E:/a327ex/Invoker/map.lua
E:/a327ex/Invoker/map_generation.lua
E:/a327ex/Invoker/player.lua
E:/a327ex/Invoker/primitives.lua
E:/a327ex/Invoker/run.lua
E:/a327ex/Invoker/spells.lua
E:/a327ex/Invoker/ui.lua
E:/a327ex/Invoker/wall.lua
E:/a327ex/Invoker/wall_flow.lua
E:/a327ex/Invoker/wall_palettes.lua
E:/a327ex/Invoker/wall_patterns.lua
E:/a327ex/Invoker/wall_symbols.lua

E:/a327ex/Invoker/assets/:
ARROW_Hit_Body_stereo.ogg
ARROW_Hit_Body_stereo.wav
ARROW_Hit_Body_stereo_lightmed.ogg
ARROW_Hit_Wood_Shield_stereo.ogg
ARROW_Hit_Wood_Shield_stereo.wav
ARROW_Hit_Wood_Shield_stereo_lightmed.ogg
Arrow Impact wood 1.ogg
Arrow Impact wood 1.wav
Arrow Impact wood 1_lightmed.ogg
Arrow Impact wood 3.ogg
Arrow Impact wood 3.wav
... [10 more lines]

Read (E:\a327ex\Invoker\main.lua)

1    require('anchor')({
2     width = 480,
3     height = 270,
4     title = "Invoker",
5     scale = 3,
6     filter = "rough",
7   })
8   
9   -- ============================================================================
10  -- LOCKED PALETTE (per reference/ui_primitives_spec.md and the four
11  -- reference/mockup_ui_kit_*.html files). 18 canonical tokens — 9 chrome
12  -- (grayscale ladder, dark to light) + 9 accents (roughly by hue).
13  -- ============================================================================
14  
15  -- Chrome (grayscale ladder, dark → light)
16  bg = color(22, 28, 34)
17  slot_fill = color(27, 31, 34)
18  panel_fill = color(31, 33, 38)
19  charcoal = color(45, 48, 51)
20  panel_border = color(56, 60, 65)
21  grey = color(74, 77, 80)
22  text_muted = color(140, 140, 145)
23  white = color(200, 200, 200)
24  text = color(240, 240, 235)
25  
26  -- Accents (roughly by hue)
27  red = color(210, 49, 56)
28  orange = color(233, 108, 27)
29  yellow = color(234, 171, 28)
30  lime = color(148, 176, 41)
... [233 more lines]

Read (E:\a327ex\Invoker\cast.lua)

1    --[[
2     cast.lua — spell chord input + execution.
3   
4     Tap-sequence chord building: primitive keys append to run.cast_slots
5     (cap = 3). LMB attempts a cast; RMB/Esc cancels. Pure-formation chords
6     silently noop (the at-least-one-element rule from spells.md). Actual
7     spell-effect dispatch is a placeholder until the catalog lands in the
8     next sub-phase.
9   
10    Inputs (bound in main.lua):
11      prim_q/e/r/t/f  the five primitive slot keys (acquisition order)
12      cast            mouse:1 — attempt cast
13      cast_cancel     mouse:3 + key:escape — clear chord
14  
15    See design.md, plan.md (Phase 1, sub-phase 3), and spells.md.
16  ]]
17  
18  --[[
19    cast_input_update(dt)
20  
21    Per-frame chord state machine + HUD spring driver. Called from update()
22    before the player update so input from this frame is reflected
23    immediately. Also advances the cast HUD's per-slot springs (used for
24    the pop-in feedback when a primitive is added).
25  
26    Discipline notes:
27      - Slot-full presses are dropped (no FIFO/LIFO replacement). The user
28        must RMB to clear and rebuild.
29      - Mid-chord removal is not supported.
30      - Order of simultaneous keypresses is whatever Lua's pairs() gives.
... [214 more lines]

Read (E:\a327ex\Invoker\run.lua)

1    --[[
2     run.lua — per-run state namespace and lifecycle.
3   
4     The `run` global holds everything that should die when a run ends. On
5     death, boss kill, or a manual reset (F12), run_reset() walks the old
6     table for resources that own engine handles, frees them, then assigns
7     a fresh `run = run_new()`. Anything reachable only through the old
8     `run` table is GC'd.
9   
10    Schema (placeholders — populated as the new spell system lands):
11      primitive_slots     fixed-size array of length PRIMITIVE_SLOT_COUNT
12                          (4). Each index is a primitive_id string or nil
13                          (empty). Slot index → key is fixed by
14                          PRIMITIVE_SLOT_KEYS = {'q','e','r','t'}, so
15                          primitive_slots[1] is the Q-bound primitive.
16                          primitives_acquire fills the earliest empty slot.
17      modifiers_owned     array of modifier ids
18      primitive_modifiers map primitive_id → array of modifier_ids attached
19      cast_slots          current chord (array of primitive_ids); cleared
20                          on cast or cancel
21      last_cast_chord     copy of the last successfully-cast chord; the
22                          Repeat button (Space / right-side button) fires
23                          this chord again. nil until first cast.
24  
25    What does NOT live in `run` (persists across runs):
26      settings, tutorial_completed, save serialization
27      UI scratch state (drag_state, cast UI visibility) — defensively cleared
28        in run_reset where it could reference per-run data.
29  
30    Discipline: the destroy pass runs BEFORE the table is replaced, so any
... [66 more lines]

Read (E:\a327ex\Invoker\primitives.lua)

1    --[[
2     primitives.lua — data table for the 8 spell primitives.
3   
4     Each primitive has:
5       id      string key; appears in run.cast_slots, run.primitive_keys, etc.
6       name    human-readable label for tooltips and tutorial prompts
7       short   (optional) HUD-fit abbreviation when `name` is too long for
8               the cast-slot label panel. Falls back to `name` when nil.
9               Example: 'Projectile' -> 'Proj.', 'Conjuration' -> 'Conj.',
10              'Lightning' -> 'Elec.', 'Duration' -> 'Time'. Short forms
11              should target ~5 chars so they fit Lana 11 inside a 30px-wide
12              label panel.
13      sprite_path (optional) Filename override when the asset on disk
14              doesn't match `assets/<id>.png`. Used e.g. when Duration's
15              sprite is named time.png to match its display abbreviation.
16      kind    'element' or 'formation'. Pure-formation chords don't fire —
17              cast_attempt enforces the at-least-one-element rule.
18      color   tint color from the locked palette. The PNG icon art is
19              (intentionally) light/grayscale so the layer_image color
20              tint paints in each primitive's identity. Elements get
21              their natural hue; formations get distinct operator hues
22              that read against the four element colors.
23      sprite  image object loaded from assets/<id>.png. Populated lazily
24              by primitives_load_sprites() — call once after engine init.
25  
26    Launch subset is 4 elements + 4 formations (8 total); the spec also
27    lists Lightning, Force, Duration, Conjuration as expansions. Sprites
28    for those land when their primitives do. See spells.md.
29  ]]
30  
... [106 more lines]

Read (E:\a327ex\Invoker\spells.lua)

1    --[[
2     spells.lua — spell catalog (chord → spell name + element).
3   
4     Lookup table keyed by *canonical* chord: primitive ids sorted
5     alphabetically and joined by commas. The spell entry carries:
6       name     display name shown in the info_text on successful cast
7       element  primitive id whose color tints the info_text. Picked
8                as the "primary" element of the spell (e.g. Fire spells
9                carry the Fire color even when Wind or another element
10               also appears in the chord).
11  
12    The catalog currently covers everything spelled out in spells.md's
13    Fire section (36 combinations). Non-Fire spells from spells.md
14    (Wind+Self Airborne, Wind+Self+Line Flicker Strike, Force+Self Dash)
15    are intentionally not yet wired up — current scope is Fire only.
16  
17    Multiple chords can map to the same display name when spells.md
18    declares them as the same effect with a behavior modification (e.g.
19    Fire+Fire vs Fire+Fire+Self both ARE "Fire Pillar", differing only
20    in where the pillar lands). The catalog records the name as-is.
21  
22    Any chord not in this table fizzles via cast_attempt.
23  ]]
24  
25  spells = {}
26  
27  --[[
28    spell_canonical_chord(chord) -> string
29  
30    Canonical key from a chord array. Multiset semantics — order doesn't
... [74 more lines]

Let me briefly check the tutorial map and spells.md to round out my understanding before suggesting next steps.

Bash (ls E:/a327ex/Invoker/maps/ 2>&1; echo "---"; wc -l E:/a327ex/Invoker/spells.md E...)

hideout.lua
tutorial.lua
---
  588 E:/a327ex/Invoker/spells.md
  126 E:/a327ex/Invoker/maps/tutorial.lua
  714 total

Read (E:\a327ex\Invoker\maps\tutorial.lua)

1    --[[
2     Tutorial — opening linear corridor. ~2 screens long. Walls have
3     organic top/bottom edges driven by stacked sine waves (centerline
4     meander + width variation + per-edge detail), with rounded west/east
5     bays (half-circles) on the ends. Single region; same wall pipeline
6     as the hideout.
7   
8     Spawn at west end. exit_x/exit_y recorded for the future exit entity.
9   ]]
10  
11  maps.tutorial = {
12    bg_color = color(17, 17, 17),
13    palette = 'snkrx_blue',
14    wall_edge_width = 2,
15    wall_symbol_size = 6,
16    wall_symbol_width = 3,
17    wall_gradient_fade_dist = 200,
18  
19    map_w = 960,
20    map_h = 270,
21    -- West/east bay centers sit this far in from each map end. Bay radius
22    -- comes from the corridor's actual half-height at that x (so the bay
23    -- auto-fits whatever the waves are doing locally).
24    bay_inset = 100,
25    base_half_height = 70,
26  
27    -- Door-gate pinch: corridor narrows sharply at pinch_x and re-expands.
28    -- pinch_depth is subtracted from the half-height; falls off via cosine
29    -- bumps with separate left/right widths so the approach and recovery
30    -- can have different visual pacing.
... [97 more lines]

Read the orblike-to-invoker session log, design.md, plan.md, main.lua, cast.lua, run.lua, primitives.lua, spells.lua, and tutorial map. Here's where we are and what I'd do next.

Where we left off

Project context: Invoker (formerly Orblike) — chord-cast spell roguelike. Last session was 50% rename sweep + 50% cast HUD substrate.

Phase 1 status (spell-cast tutorial):

Sub-phase	Status
1. Primitives data layer	✅ Done (Fire-only element + 6 formations: self/duration/projectile/conjuration/line/rune)
2. Cast HUD substrate	✅ Done (3 cast slots + primitive HUD + Q/E/R/T keycaps + Space-repeat button + info_text)
3. Spell execution engine	⚠️ Half — catalog of 36 Fire combinations exists, lookup works, but cast_attempt only spawns floating text. No real effects, no mana gating, no dispatch.
4. Spell content (effects)	❌
5. Mana system	❌
6. Combat HP / orbs	❌
7. Tutorial map (entities)	❌ Geometry exists (maps/tutorial.lua — corridor with pinch gate + east/west bays) but no entities live in it
8. Tutorial entities (breakables)	❌
9. Tutorial scripting	❌
10. Door exit → dungeon	❌

Final note from previous session log: "Next session: actual tutorial."

Phase 1 validation gate: "Does chord-casting feel satisfying?" — unanswered, because there's no payoff yet. The chord builds beautifully (HUD, springs, repeat button, fizzles), but casting only prints the spell name. The loop hasn't been validated.

What I'd suggest

The bottleneck for the validation gate is at least one spell with a real visible effect. Without it, "is the chord-cast rhythm fun?" is unanswerable — the player builds a chord and gets a text label. Until something fires, we can't tell whether the rhythm is right.

So I'd lead with Fireball (Fire + Projectile) as the first real effect:

One spell that actually does something — Fireball seems best: simple to author (spawn projectile, travel, collide, vanish), unambiguously the iconic spell, and a projectile in flight gives the chord-build → cast → recover rhythm a visual heartbeat.
A breakable crate to receive the projectile — the smallest target entity (HP, collider tagged so projectiles can find it, destroy on zero, maybe a tiny puff). Spawned in the tutorial corridor.
HP + mana orbs (sub-phase 6) — ARPG corner orbs. Even before there's an enemy that damages you, the mana orb is what makes mana costs feel like something. This unlocks the next item:
Mana cost on Fireball — cost stored on the spell entry, deducted in cast_attempt, failure-to-cast feedback if pool < cost. This is the cheapest version of sub-phase 5.

That sequence answers the validation question before any tutorial scripting is invested. If casting Fireball at a crate three times in a row feels good — chord, hover-cast, recover, repeat — we've validated. If not, we pivot before writing prompts/highlights/door transitions/etc.

After that, the tutorial map content (sub-phases 7-9) and the door exit (10) come together fast — the systems exist, we're authoring content.

Order I'd suggest within "first real spell"

Breakable crate entity first (target needs to exist before the projectile has anywhere to land)
Fireball effect (projectile spawn, velocity from cast direction, collision with crate + walls)
HP + mana orbs (screen-space, bottom corners)
Mana cost gating

Or — alternative ordering — skip mana entirely until the spell-effect loop feels good. Avoids tuning a cost against a spell whose feel isn't validated.

Open questions for you

AskUserQuestion

User has answered your questions: "What should we start on first?"="HP + mana orbs first", "Cast aiming direction — how should projectile spells aim?"="Toward mouse cursor (Recommended)". You can now continue with the user's answers in mind.

Good — HP + mana orbs first, mouse-cursor aim for later projectile work. Before I start coding, I need to look at the UI kit mockups since they're canonical for orb visual design.

Grep (orb|HP|mana|health)

5:<title>Orblike UI Kit 1 — Color Mockup</title>
11:     ORBLIKE UI KIT 1 — locked palette per user feedback (2026-04-30).
208:  /* --- ORB COMPOSITION -------------------------------------------- */
353:  <h1>Orblike UI Kit 1</h1>
455:     02 ORB / SKILL ICONS
458:  <div class="section-label"><span class="num">02</span> ORB / SKILL ICONS
476:     03 SHOP / 04 ORB COMPOSITION / 05 INVENTORY
498:    <div class="section-label"><span class="num">04</span> ORB COMPOSITION</div>
574:      <span class="desc">health, resources, and quick stats</span>
594:        <div class="desc">conjure a whirling orb that orbits around you</div>
642:      <div class="desc">hurls a burning orb<br>on hit, 4 damage<br>+ ignite</div>

Read (E:\a327ex\Invoker\reference\mockup_ui_kit_1.html)

560          wave <span class="green-text">5</span><span class="muted">/15</span>
561         <div class="wave-pips">
562           <div class="pip filled"></div><div class="pip filled"></div><div class="pip filled"></div>
563           <div class="pip filled"></div><div class="pip filled"></div>
564           <div class="pip"></div><div class="pip"></div><div class="pip"></div>
565           <div class="pip"></div><div class="pip"></div>
566         </div>
567       </div>
568       <div class="muted" style="font-size: 11px; margin-top: 6px;">progress through the round</div>
569     </div>
570   </div>
571 
572   <div>
573     <div class="section-label"><span class="num">07</span> COMBAT HUD
574       <span class="desc">health, resources, and quick stats</span>
575     </div>
576     <div class="panel">
577       <div class="row tight">
578         <div class="nested-panel"><span class="red-text">&hearts;</span> 10/10</div>
579         <div class="nested-panel"><span class="light-blue-text">&diams;</span> 120</div>
580         <div class="nested-panel"><span class="gold-text">&diams;</span> 34</div>
581         <div class="nested-panel">&#9760; 125</div>
582       </div>
583     </div>
584   </div>
585 
586   <div>
587     <div class="section-label"><span class="num">08</span> TOOLTIPS
588       <span class="desc">clear, informative, and tier aware</span>
589     </div>
... [73 more lines]

Found something important — let me check the "orb composition" section the mockup references and the dungeon HUD too.

Read (E:\a327ex\Invoker\reference\mockup_ui_kit_1.html)

449      <div class="icon-btn">&#128214;</div>
450     <div class="icon-btn">&#9881;</div>
451   </div>
452 </div>
453 
454 <!-- ====================================================================
455      02 ORB / SKILL ICONS
456      ==================================================================== -->
457 <div class="section">
458   <div class="section-label"><span class="num">02</span> ORB / SKILL ICONS
459     <span class="desc">white symbols on color tiles, distinct meaning</span>
460   </div>
461   <div class="row">
462     <div class="tile fire">&#11044;</div>
463     <div class="tile ice">&#10043;</div>
464     <div class="tile shoot">&uarr;</div>
465     <div class="tile sword">&#9876;</div>
466     <div class="tile target">&#9678;</div>
467     <div class="tile shield">&#9960;</div>
468     <div class="tile gold">&#10026;</div>
469     <div class="tile skull">&#9760;</div>
470     <div class="tile spark">&#10038;</div>
471     <div class="tile psion">&#10042;</div>
472   </div>
473 </div>
474 
475 <!-- ====================================================================
476      03 SHOP / 04 ORB COMPOSITION / 05 INVENTORY
477      ==================================================================== -->
478 <div class="section-row cols-3">
... [85 more lines]

There's a design contradiction I want to surface. design.md says PoE-style corner orbs ("HP orb bottom-left, mana orb bottom-right"), but the canonical mockup_ui_kit_1.html shows the Combat HUD as a compact row of stat panels (♥ 10/10 ♦ 120 ♦ 34 ☠ 125) — no orbs at corners. Let me check the dungeon top-bar mockup to see if there's a unified language.

Read (E:\a327ex\Invoker\reference\minesweeper_mockup_1.png)

[
  {
    "type": "image",
    "source": {
      "type": "base64",
      "data": "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",
      "media_type": "image/png"
    }
  }
]

This is exactly the contradiction. The minesweeper mockup shows the dungeon view top bar as a horizontal row of stat panels: ♥ 10/10 ♦ 23 gold 🚩 23/28 flags. And UI Kit 1's "Combat HUD" matches that style. design.md's "ARPG corner orbs" spec was written before the mockups were locked, and CLAUDE.md says the mockups are canonical.

A practical consequence: the current bottom strip is already busy (primitive HUD ← cast HUD → repeat button). Corner orbs at the bottom would crowd that strip. A top-bar HUD avoids the crowding and matches the dungeon view's HUD language.

AskUserQuestion

User has answered your questions: "Which HUD style for HP and mana?"="Let's try both styles for now so I can see which one looks better initially.", "Starting values?"="HP 100 / Mana 100". You can now continue with the user's answers in mind.

OK — building both side-by-side, HP 100 / Mana 100. Let me check what UI primitives I have to work with and find available screen real estate.

Bash (wc -l E:/a327ex/Invoker/ui.lua E:/a327ex/Invoker/info_text.lua; echo "---"; grep...)

  765 E:/a327ex/Invoker/ui.lua
  123 E:/a327ex/Invoker/info_text.lua
  888 total
---
23:local function with_alpha(c, mul)
51:function ui_panel(lyr, x, y, w, h, opts)
77:function ui_text(lyr, text, x, y, opts)
146:function ui_slot(lyr, x, y, size, state, opts)
221:function ui_item_icon(lyr, x, y, size, fill, glyph, opts)
258:function ui_cast_hud_update(dt)
262:function ui_cast_hud_pulse(slot_idx)
267:function ui_primitive_hud_update(dt)
271:function ui_primitive_hud_pulse(slot_idx)
293:function ui_repeat_button_update(dt)
297:function ui_repeat_button_pulse()
305:local function repeat_button_rect()
312:function ui_repeat_button_hovered(mx, my)
318:function ui_repeat_button(lyr, image)
388:function ui_cast_hud(lyr, center_x, bottom_y, chord)
478:function ui_primitive_hud(lyr, right_x, bottom_y, slots, mouse_x, mouse_y)
601:function drag_active()
605:function drag_start(source_id, item_id, mx, my, icon_x, icon_y)
612:function drag_clear()
625:function slot_at(mx, my, slots)
646:function drag_cancel_pressed()
657:function ui_drag_ghost(lyr, mx, my, size, fill, glyph, opts)
673:function ui_tooltip_for_item(lyr, mx, my, item)
708:function ui_draw_test(lyr)

Read (E:\a327ex\Invoker\ui.lua)

1    --[[
2     ui.lua — UI primitives for Invoker.
3   
4     Pure render functions. The caller owns state (hover, active, focus, drag) and
5     passes it in; primitives never hide internal state. Widget IDs (when needed)
6     are explicit caller-provided strings.
7   
8     Discipline (per the action-vs-rules dichotomy applied to UI):
9       - Data (inventory contents, equipped items, drag-in-progress) lives in
10        retained models elsewhere. Action-based.
11      - Rendering and per-frame interaction (hit-testing, hover, valid-target
12        computation) is rules-based and recomputed each frame from input + data.
13      - Primitives in this file are pure draw — no state, no closures over data.
14  
15    See reference/ui_primitives_spec.md for the design tokens these implement.
16  ]]
17  
18  -- F2-toggled debug screen for live-tuning chrome against the kit.
19  ui_test_active = false
20  
21  -- Clone `c` with alpha multiplied by `mul`. New color per call; only call from
22  -- UI-frequency draw paths. Used for the `disabled` slot state.
23  local function with_alpha(c, mul)
24    local out = color_clone(c)
25    out.a = math.floor(out.a*mul + 0.5)
26    return out
27  end
28  
29  --[[
30    ui_panel(lyr, x, y, w, h, opts)
... [130 more lines]

Bash (grep -nE "circle|ellipse|arc|rounded_rectangle|polygon" E:/a327ex/Anchor2/docs/E...)

13:layer_circle(layer, x, y, radius, color)
14:layer_circle_line(layer, x, y, radius, color, line_width?)
20:layer_polygon(layer, vertices, color)
21:layer_polygon_line(layer, vertices, color, line_width?)
22:layer_rounded_rectangle(layer, x, y, w, h, radius, color)
23:layer_rounded_rectangle_line(layer, x, y, w, h, radius, color, line_width?)
133:physics_add_circle(body, tag, radius, opts?) -> shape
136:physics_add_polygon(body, tag, vertices, opts?) -> shape
203:physics_query_circle(x, y, radius, tags) -> [body, ...]
207:physics_query_polygon(x, y, vertices, tags) -> [body, ...]

Bash (grep -nE "scissor|clip|stencil|mask" E:/a327ex/Anchor2/docs/ENGINE_API_QUICK.md ...)

28:layer_stencil_mask(layer)
29:layer_stencil_test(layer)
30:layer_stencil_test_inverse(layer)
31:layer_stencil_off(layer)
333:clipboard_get() -> string | nil
334:clipboard_set(text) -> bool
335:clipboard_has_text() -> bool

Bash (grep -nE "stencil" E:/a327ex/Anchor2/docs/ENGINE_API.md | head -30)

220:### layer_stencil_mask
222:`layer_stencil_mask(layer)`
224:Starts writing to the stencil buffer. Subsequent draws write to stencil only (not visible on screen). Use to define a mask shape.
227:layer_stencil_mask(layer)
228:layer_rectangle(layer, 100, 100, 50, 50, rgba(255, 255, 255))  -- draws to stencil only
229:layer_stencil_test(layer)
230:-- subsequent draws only appear where stencil was set
233:### layer_stencil_test
235:`layer_stencil_test(layer)`
237:Starts testing against the stencil buffer. Subsequent draws only appear where stencil has been written.
240:layer_stencil_mask(layer)
242:layer_stencil_test(layer)
244:layer_stencil_off(layer)
247:### layer_stencil_off
249:`layer_stencil_off(layer)`
251:Disables stencil, returns to normal drawing.
254:layer_stencil_off(layer)
2619:### layer_stencil_test_inverse
2621:`layer_stencil_test_inverse(layer)`
2623:Starts inverse stencil testing — subsequent draws only appear where the stencil was NOT written. Complement of `layer_stencil_test`.
2626:layer_stencil_mask(layer)
2628:layer_stencil_test_inverse(layer)
2630:layer_stencil_off(layer)

I have everything I need to design. Here's my plan — code snippets below, then I'll wait for your OK before writing.

Plan — render both HUD styles simultaneously

A) run.lua schema additions — add hp/hp_max/mp/mp_max to the run table:

return {
  id = (run and run.id or 0) + 1,
  hp     = 100, hp_max = 100,
  mp     = 100, mp_max = 100,
  primitive_slots     = {},
  modifiers_owned     = {},
  primitive_modifiers = {},
  cast_slots          = {},
  last_cast_chord     = nil,
}

B) ui.lua — corner orb (PoE-style)

-- ui_resource_orb(lyr, cx, cy, radius, cur, max, fill_color)
-- Filled circle with stencil-clipped liquid fill rising from bottom.
-- Centered text "cur/max" overlay in white via Lana 11.
function ui_resource_orb(lyr, cx, cy, radius, cur, max, fill_color)
  local pct = max > 0 and math.clamp(cur/max, 0, 1) or 0
  local fill_h = 2*radius*pct
  local fill_top_y = cy + radius - fill_h

  -- Empty background
  layer_circle(lyr, cx, cy, radius, slot_fill())

  -- Liquid fill (stencil-clipped to circle)
  layer_stencil_mask(lyr)
  layer_circle(lyr, cx, cy, radius, color(255,255,255))
  layer_stencil_test(lyr)
  layer_rectangle(lyr, cx - radius, fill_top_y, 2*radius, fill_h, fill_color())
  layer_stencil_off(lyr)

  -- Border
  layer_circle_line(lyr, cx, cy, radius, panel_border(), 1)

  -- Text
  ui_text(lyr, tostring(math.floor(cur)) .. '/' .. tostring(math.floor(max)),
    cx, cy - 6, {align = 'center', color = text})
end

C) ui.lua — top-bar mockup-style stat panel

-- ui_stat_chip(lyr, x, y, dot_color, text_str) -> x_end
-- Compact panel: dot icon + value. Returns x_end for chaining.
function ui_stat_chip(lyr, x, y, dot_color, text_str)
  local w, h = 50, 14
  ui_panel(lyr, x, y, w, h, {radius = 2})
  layer_rectangle(lyr, x + 4, y + 4, 6, 6, dot_color())
  ui_text(lyr, text_str, x + 14, y + 2, {color = text})
  return x + w
end

D) main.lua — wiring and debug binds

-- Near the other bind() calls:
bind('debug_hp_down', 'key:[')   -- subtract 10 HP for visual tuning
bind('debug_mp_down', 'key:]')   -- subtract 10 MP for visual tuning

-- In update():
if run and is_pressed('debug_hp_down') then run.hp = math.max(0, run.hp - 10) end
if run and is_pressed('debug_mp_down') then run.mp = math.max(0, run.mp - 10) end

-- In draw(), inside the `if run then` block, before the cast HUD:
-- Corner orbs (PoE style)
ui_resource_orb(ui_layer, 32, height - 32, 22, run.hp, run.hp_max, red)
ui_resource_orb(ui_layer, width - 32, height - 32, 22, run.mp, run.mp_max, light_blue)

-- Top-bar chips (mockup style)
local x = 8
x = ui_stat_chip(ui_layer, x, 4, red,        'HP ' .. run.hp .. '/' .. run.hp_max)
x = ui_stat_chip(ui_layer, x + 4, 4, light_blue, 'MP ' .. run.mp .. '/' .. run.mp_max)

Layout check: Orbs at radius 22, centered (32, height-32) and (width-32, height-32). Bottom-left orb spans x=10 to x=54 — fits in the 0-80 free strip left of the primitive HUD. Bottom-right orb spans x=426 to x=470 — fits in the 316-480 free strip right of the repeat button. Top-bar chips at y=4-18, x=8-112 — empty space; no conflicts.

Open question I haven't asked: the top-bar mockup uses ♥ and ♦ glyphs. I'm using a small colored square as the icon since font glyph coverage is uncertain. Easy to swap later if you'd prefer trying the heart/diamond chars.

Does this look right? If yes I'll write it.

Go ahead.

Bash (grep -nE "layer_rectangle\b|math.clamp" E:/a327ex/Anchor2/docs/ENGINE_API_QUICK...)

E:/a327ex/Anchor2/docs/ENGINE_API_QUICK.md:9:layer_rectangle(layer, x, y, w, h, color)
E:/a327ex/Invoker/anchor/math.lua:121:  return math.clamp(t - math.floor(t/length)*length, 0, length)
E:/a327ex/Invoker/anchor/math.lua:145:  return source + dt*math.clamp(t, 0, 1)
E:/a327ex/Invoker/anchor/math.lua:211:    math.clamp(5, 0, 10)   -> 5
E:/a327ex/Invoker/anchor/math.lua:212:    math.clamp(-5, 0, 10)  -> 0
E:/a327ex/Invoker/anchor/math.lua:213:    math.clamp(15, 0, 10)  -> 10
E:/a327ex/Invoker/anchor/math.lua:222:function math.clamp(value, min, max)
E:/a327ex/Invoker/anchor/math.lua:248:    - To clamp, combine with math.clamp on input or output

Both exist. Writing the changes.

Edit (E:\a327ex\Invoker\run.lua)

The file E:\a327ex\Invoker\run.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Now add the UI primitives to ui.lua. Let me insert them after ui_item_icon and before the cast HUD section.

Read (E:\a327ex\Invoker\ui.lua)

221  function ui_item_icon(lyr, x, y, size, fill, glyph, opts)
222   opts = opts or {}
223   local f  = opts.font        or ui_font_header_l
224   local gc = opts.glyph_color or white
225   local r  = opts.radius      or 2
226 
227   layer_rounded_rectangle(lyr, x, y, size, size, r, fill())
228   ui_text(lyr, glyph, x + size/2, y + (size - f.height)/2, {font = f, color = gc, align = 'center'})
229 end
230 
231 --[[
232   Cast HUD + primitive HUD spring state.
233 
234   cast_hud_springs[i=1..3]        spring per cast-chord slot; pulses on
235                                   chord insertion via ui_cast_hud_pulse.
236   primitive_hud_springs[i=1..4]   spring per primitive slot (Q/E/R/T);
237                                   pulses on acquisition via
238                                   ui_primitive_hud_pulse.
239 
240   PRIMITIVE_SLOT_COUNT is hardcoded as 4 here because primitives.lua
241   loads after ui.lua and the constant isn't defined yet at module-load
242   time. Keep in sync if the count ever changes.
243 ]]
244 cast_hud_springs = {}
245 for i = 1, 3 do cast_hud_springs[i] = spring_new() end
246 
247 primitive_hud_springs = {}
248 for i = 1, 4 do primitive_hud_springs[i] = spring_new() end
249 
250 -- Per-slot sticky hover state. Set true when the mouse enters the slot
... [15 more lines]

Edit (E:\a327ex\Invoker\ui.lua)

The file E:\a327ex\Invoker\ui.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Now wire it into main.lua — debug binds, update handling, and the two render calls.

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Per CLAUDE.md, the engine builds itself but you compile and run the game. Code's ready to test — launch and tell me how the two styles look. [ decrements HP by 10, ] decrements MP by 10 so you can see the orb fill drop and the chip readout tick down.

Open questions for after you compare:

Orbs: radius 22 OK, or want bigger/smaller?
Chips: width is a fixed 50 px; readouts like HP 100/100 may run close to the right edge. If text overflows, I'll shrink the readout to just the current value, or widen the chip.
Top-bar position: y=4 puts chips ~4 px from the top edge. Move down if it feels cramped.

Lua draw() error: anchor/layer.lua:109: bad argument #5 to 'circle' (number expected, got table)

The signature is different than what I assumed — let me check.

Read (E:\a327ex\Invoker\anchor\layer.lua)

90   --- Create a layer state table and optionally register in global `layers`.
91  function layer_new(name)
92    local lyr = {
93      name = name,
94      handle = eng.create(name),
95      parallax_x = 1,
96      parallax_y = 1,
97    }
98    if layers then
99      layers[name] = lyr
100   end
101   return lyr
102 end
103 
104 function layer_rectangle(lyr, x, y, w, h, color)
105   eng.rectangle(lyr_handle(lyr), x, y, w, h, color)
106 end
107 
108 function layer_circle(lyr, x, y, radius, color)
109   eng.circle(lyr_handle(lyr), x, y, radius, color)
110 end
111 
112 function layer_rectangle_line(lyr, x, y, w, h, color, line_width)
113   eng.rectangle_line(lyr_handle(lyr), x, y, w, h, color, line_width or 1)
114 end
115 
116 function layer_circle_line(lyr, x, y, radius, color, line_width)
117   eng.circle_line(lyr_handle(lyr), x, y, radius, color, line_width or 1)
118 end
119 
... [10 more lines]

Bash (grep -n "^function color|color.__call|setmetatable.*color|function color_clon...)

94:function color_new(r, g, b, a)
111:function color_from_hsl(h, s, l, a)
123:function color_to_hsl(c)
131:function color_clone(c)
140:function color_mix(a, b, t)
155:function color_darken(c, factor)
170:function color_lighten(c, factor)
178:function color_invert(c)

Read (E:\a327ex\Invoker\anchor\color.lua)

70       return p
71    end
72    local r = math.floor(hue_to_rgb(h + 1/3)*255 + 0.5)
73    local g = math.floor(hue_to_rgb(h)*255 + 0.5)
74    local b = math.floor(hue_to_rgb(h - 1/3)*255 + 0.5)
75    return r, g, b
76  end
77  
78  -- Metatable for colors: __call returns the packed rgba integer.
79  local color_mt = {}
80  
81  color_mt.__call = function(c)
82    return rgba(
83      math.floor(c.r + 0.5),
84      math.floor(c.g + 0.5),
85      math.floor(c.b + 0.5),
86      math.floor(c.a + 0.5)
87    )
88  end
89  
90  --[[
91    color_new(r, g, b, a)
92    Create a new color. Defaults to white (255, 255, 255, 255).
93  ]]
94  function color_new(r, g, b, a)
95    return setmetatable({
96      r = r or 255,
97      g = g or 255,
98      b = b or 255,
99      a = a or 255,
... [20 more lines]

Found it — colors must be invoked (()) to pack into rgba ints for the engine. My stencil-mask circle uses color(255, 255, 255) without the call. Fix:

Edit (E:\a327ex\Invoker\ui.lua)

The file E:\a327ex\Invoker\ui.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Fixed. The stencil-mask draw doesn't render visibly (it only writes to the stencil buffer), so the choice of color doesn't matter — white() is just a convenient packed-rgba source. Try again.

OK, let's have the ones that are on the top left be on the bottom left stacked vertically, and then remove the orbs.

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Also removing the now-unused ui_resource_orb function from ui.lua so it's not dead code.

Edit (E:\a327ex\Invoker\ui.lua)

The file E:\a327ex\Invoker\ui.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Orbs gone, chips moved to bottom-left stacked vertically (HP on top, MP below, 8 px margin from bottom and left, 2 px gap between them).

Hmmm, let's try the orbs again, orblike_hud has a visual example. Try to make it look like liquid with a shader instead of just a flat circle, and try to keep it in the general style of the UI kit if you can. Show both orbs to me in the center of the screen, we'll iterate them over time.

Let me view the reference image first.

Bash (ls E:/a327ex/Invoker/reference/orblike_hud.png E:/a327ex/Invoker/reference/*.fra...)

Exit code 2
ls: cannot access 'E:/a327ex/Invoker/reference/*.frag': No such file or directory
E:/a327ex/Invoker/reference/orblike_hud.png

Read (E:\a327ex\Invoker\reference\orblike_hud.png)

[
  {
    "type": "image",
    "source": {
      "type": "base64",
      "data": "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",
      "media_type": "image/png"
    }
  }
]

Looking at the reference: large round orbs with bright liquid fill, gradient on the liquid (brighter at top), flat waterline, large centered current value, "/ max" below, icon at bottom, dark border. Let me check how existing shaders are wired so I can add a liquid shader.

Bash (cat E:/a327ex/Invoker/assets/wall_gradient.frag 2>/dev/null | head -30; echo "--...)

// Per-pixel wall gradient coloring shader.
//
// Applied as a post-process to the wall area layer. The wall triangulation
// is drawn in placeholder white (alpha=1); this shader replaces those
// pixels with the pre-baked gradient color sampled at the pixel's world
// position. Pixels with alpha=0 (outside the wall triangles, i.e., over
// the corridor or outside the map) pass through unchanged.
//
// Editions: when u_edition > 0, the gradient color is further transformed
// per the edition table below. The edition transform happens AFTER the
// gradient sample so the shimmer/shift operates on the final wall color
// (not on the placeholder white that gets replaced).
//
// Edition table (mirrors draw_shader.frag's apply_edition):
//   2..11 = old fixed-hue shimmer (ruby/rose/amber/gold/lime/emerald/cyan/sapphire/indigo/violet)
//   12    = old continuous-HSL polychrome
//   13    = silver
//   14    = white
//   23    = NEW dynamic-hue shimmer (center derived from gradient color)
//   24    = NEW palette-cycling polychrome (discrete cycle through 9 new palette accents)

in vec2 TexCoord;
out vec4 FragColor;

uniform sampler2D u_texture;       // the wall area layer FBO (auto-bound, unit 0)
uniform sampler2D u_gradient;      // pre-baked region-aware wall color (unit 1)
uniform vec2 u_cam;                // camera center in world space
uniform vec2 u_screen;             // screen size in world units
uniform vec2 u_grad_origin;        // gradient texture's world-space origin (top-left)
uniform vec2 u_grad_size;          // gradient texture's world-space extent (w, h)
... [8 more lines]

Bash (grep -nE "layer_shader_apply|shader_set|attach_shader|layer_render_to|render_tar...)

276:layer_shader_set_int(layer, shader, "u_texture", tex)
323:### shader_set_float_immediate
325:`shader_set_float_immediate(shader, name, value)`
330:shader_set_float_immediate(blur_shader, "u_radius", 5.0)
333:### shader_set_vec2_immediate
335:`shader_set_vec2_immediate(shader, name, x, y)`
340:shader_set_vec2_immediate(outline_shader, "u_pixel_size", 1/480, 1/270)
343:### shader_set_vec4_immediate
345:`shader_set_vec4_immediate(shader, name, x, y, z, w)`
350:shader_set_vec4_immediate(tint_shader, "u_color", 1.0, 0.5, 0.0, 1.0)
353:### shader_set_int_immediate
355:`shader_set_int_immediate(shader, name, value)`
360:shader_set_int_immediate(effect_shader, "u_mode", 2)
778:### layer_shader_set_float
780:`layer_shader_set_float(layer, shader, name, value)`
785:layer_shader_set_float(layer, blur_shader, "u_radius", 5.0)
788:### layer_shader_set_vec2
790:`layer_shader_set_vec2(layer, shader, name, x, y)`
795:layer_shader_set_vec2(layer, shader, "u_resolution", 480, 270)
798:### layer_shader_set_vec4

Read (E:\a327ex\Anchor2\docs\ENGINE_API.md)

258  
259 `layer_draw(layer, x?, y?)`
260 
261 Queues the layer to be drawn to screen at the given offset.
262 
263 ```lua
264 layer_draw(game_layer)
265 layer_draw(ui_layer, 0, 0)
266 ```
267 
268 ### layer_get_texture
269 
270 `layer_get_texture(layer) -> texture_id`
271 
272 Returns the layer's current texture handle (for shader uniforms).
273 
274 ```lua
275 local tex = layer_get_texture(layer)
276 layer_shader_set_int(layer, shader, "u_texture", tex)
277 ```
278 
279 ### layer_reset_effects
280 
281 `layer_reset_effects(layer)`
282 
283 Clears the layer's contents and resets effect state.
284 
285 ```lua
286 layer_reset_effects(layer)
287 ```
... [60 more lines]

Grep (wall_gradient_shader|wall_symbols_shader|layer_shader_set|layer_draw_from)

E:\a327ex\Invoker\map.lua:141:  apply_wall_gradient_shader(self, wall_area_layer)
E:\a327ex\Invoker\map.lua:156:  apply_wall_symbols_shader(self, symbol_layer)
E:\a327ex\Invoker\wall.lua:402:  apply_wall_gradient_shader after to convert white pixels to the gradient
E:\a327ex\Invoker\wall.lua:423:  apply_wall_gradient_shader(m, layer)
E:\a327ex\Invoker\wall.lua:429:function apply_wall_gradient_shader(m, layer)
E:\a327ex\Invoker\wall.lua:432:  layer_shader_set_vec2(layer, wall_gradient_shader, 'u_cam', main_camera.x, main_camera.y)
E:\a327ex\Invoker\wall.lua:433:  layer_shader_set_vec2(layer, wall_gradient_shader, 'u_screen', width, height)
E:\a327ex\Invoker\wall.lua:434:  layer_shader_set_vec2(layer, wall_gradient_shader, 'u_grad_origin', g.bx0, g.by0)
E:\a327ex\Invoker\wall.lua:435:  layer_shader_set_vec2(layer, wall_gradient_shader, 'u_grad_size', g.bw, g.bh)
E:\a327ex\Invoker\wall.lua:436:  layer_shader_set_texture(layer, wall_gradient_shader, 'u_gradient', g.gradient_texture, 1)
E:\a327ex\Invoker\wall.lua:441:  layer_shader_set_float(layer, wall_gradient_shader, 'u_edition', m.wall_edition or 0)
E:\a327ex\Invoker\wall.lua:442:  layer_shader_set_float(layer, wall_gradient_shader, 'u_time', time)
E:\a327ex\Invoker\wall.lua:443:  layer_apply_shader(layer, wall_gradient_shader)
E:\a327ex\Invoker\wall.lua:522:  apply_wall_symbols_shader(m, layer)
E:\a327ex\Invoker\wall.lua:528:function apply_wall_symbols_shader(m, layer)
E:\a327ex\Invoker\wall.lua:531:  layer_shader_set_vec2(layer, wall_symbols_shader, 'u_cam', main_camera.x, main_camera.y)
E:\a327ex\Invoker\wall.lua:532:  layer_shader_set_vec2(layer, wall_symbols_shader, 'u_screen', width, height)
E:\a327ex\Invoker\wall.lua:533:  layer_shader_set_vec2(layer, wall_symbols_shader, 'u_df_origin', g.bx0, g.by0)
E:\a327ex\Invoker\wall.lua:534:  layer_shader_set_vec2(layer, wall_symbols_shader, 'u_df_size', g.bw, g.bh)
E:\a327ex\Invoker\wall.lua:535:  layer_shader_set_texture(layer, wall_symbols_shader, 'u_symbol_color', g.symbol_gradient_texture, 1)
E:\a327ex\Invoker\wall.lua:540:  layer_shader_set_float(layer, wall_symbols_shader, 'u_edition', m.wall_edition or 0)
E:\a327ex\Invoker\wall.lua:541:  layer_shader_set_float(layer, wall_symbols_shader, 'u_time', time)
E:\a327ex\Invoker\wall.lua:542:  layer_apply_shader(layer, wall_symbols_shader)
E:\a327ex\Invoker\main.lua:102:wall_symbols_shader = shader_load_file('assets/wall_symbols.frag')
E:\a327ex\Invoker\main.lua:103:wall_gradient_shader = shader_load_file('assets/wall_gradient.frag')
E:\a327ex\Invoker\main.lua:138:  layer_shader_set_float(layer, draw_shader, 'u_edition', mode or 0)
E:\a327ex\Invoker\main.lua:139:  layer_shader_set_float(layer, draw_shader, 'u_time', time)
E:\a327ex\Invoker\anchor\layer.lua:66:  shader_set_float = layer_shader_set_float,
E:\a327ex\Invoker\anchor\layer.lua:67:  shader_set_vec2 = layer_shader_set_vec2,
E:\a327ex\Invoker\anchor\layer.lua:68:  shader_set_vec4 = layer_shader_set_vec4,
... [12 more lines]

Bash (grep -nE "layer_apply_shader" E:/a327ex/Anchor2/docs/ENGINE_API.md E:/a327ex/Inv...)

E:/a327ex/Anchor2/docs/ENGINE_API.md:818:### layer_apply_shader
E:/a327ex/Anchor2/docs/ENGINE_API.md:820:`layer_apply_shader(layer, shader)`
E:/a327ex/Anchor2/docs/ENGINE_API.md:826:layer_apply_shader(layer, blur_shader)
E:/a327ex/Anchor2/docs/ENGINE_API.md:2612:layer_apply_shader(game, shaders.composite)
E:/a327ex/Invoker/anchor/layer.lua:65:  apply_shader = layer_apply_shader,
E:/a327ex/Invoker/anchor/layer.lua:211:function layer_apply_shader(lyr, shader)

Read (E:\a327ex\Anchor2\docs\ENGINE_API.md)

818  ### layer_apply_shader
819 
820 `layer_apply_shader(layer, shader)`
821 
822 Applies a shader to the layer's current contents (ping-pong rendering).
823 
824 ```lua
825 layer_shader_set_float(layer, blur_shader, "u_radius", 4.0)
826 layer_apply_shader(layer, blur_shader)
827 ```
828 
829 ---
830 
831 ## Physics: World & Bodies
832 
833 ### physics_init
834 
835 `physics_init()`
836 
837 Initializes the physics world. Call once at startup.
838 
839 ```lua
840 physics_init()
841 ```
842 
843 ### physics_set_gravity
844 
845 `physics_set_gravity(gx, gy)`
846 
847 Sets world gravity in pixels/sec².
... [20 more lines]

Let me see the full wall_gradient.frag so I mirror the shader template exactly.

Read (E:\a327ex\Invoker\assets\wall_gradient.frag)

1    // Per-pixel wall gradient coloring shader.
2   //
3   // Applied as a post-process to the wall area layer. The wall triangulation
4   // is drawn in placeholder white (alpha=1); this shader replaces those
5   // pixels with the pre-baked gradient color sampled at the pixel's world
6   // position. Pixels with alpha=0 (outside the wall triangles, i.e., over
7   // the corridor or outside the map) pass through unchanged.
8   //
9   // Editions: when u_edition > 0, the gradient color is further transformed
10  // per the edition table below. The edition transform happens AFTER the
11  // gradient sample so the shimmer/shift operates on the final wall color
12  // (not on the placeholder white that gets replaced).
13  //
14  // Edition table (mirrors draw_shader.frag's apply_edition):
15  //   2..11 = old fixed-hue shimmer (ruby/rose/amber/gold/lime/emerald/cyan/sapphire/indigo/violet)
16  //   12    = old continuous-HSL polychrome
17  //   13    = silver
18  //   14    = white
19  //   23    = NEW dynamic-hue shimmer (center derived from gradient color)
20  //   24    = NEW palette-cycling polychrome (discrete cycle through 9 new palette accents)
21  
22  in vec2 TexCoord;
23  out vec4 FragColor;
24  
25  uniform sampler2D u_texture;       // the wall area layer FBO (auto-bound, unit 0)
26  uniform sampler2D u_gradient;      // pre-baked region-aware wall color (unit 1)
27  uniform vec2 u_cam;                // camera center in world space
28  uniform vec2 u_screen;             // screen size in world units
29  uniform vec2 u_grad_origin;        // gradient texture's world-space origin (top-left)
30  uniform vec2 u_grad_size;          // gradient texture's world-space extent (w, h)
... [152 more lines]

Now writing the shader and wiring. Starting with the fragment shader.

Write (E:\a327ex\Invoker\assets\orb_liquid.frag)

File created successfully at: E:\a327ex\Invoker\assets\orb_liquid.frag (file state is current in your context — no need to Read it back)

Now wire the shader + dedicated layer into main.lua, and add a Lua function in ui.lua.

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Now the ui.lua draw function — replaces ui_resource_orb with a shader-driven ui_orbs_apply + chrome overlay. Adding after ui_stat_chip.

Edit (E:\a327ex\Invoker\ui.lua)

The file E:\a327ex\Invoker\ui.lua has been updated successfully. (file state is current in your context — no need to Read it back)

There's no vec3 setter — only vec4. The shader declares color as vec3. Let me change it to vec4 (alpha unused) for the transport.

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated. All occurrences were successfully replaced. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated. All occurrences were successfully replaced. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated. All occurrences were successfully replaced. (file state is current in your context — no need to Read it back)

Read (E:\a327ex\Invoker\assets\orb_liquid.frag)

35   uniform float u_orb2_radius;
36  uniform float u_orb2_fill;
37  uniform vec4  u_orb2_color;
38  
39  vec4 sample_orb(vec2 pix, vec2 center, float radius, float fill, vec4 base4) {
40      vec3 base = base4.rgb;
41      if (radius <= 0.0) return vec4(0.0);
42  
43      vec2 d = pix - center;
44      float dist = length(d);
45      if (dist > radius) return vec4(0.0);
46  
47      // Local coords in [-1, 1] across both axes; y grows downward.
48      vec2 local = d / radius;
49

Good. Now wire into draw().

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Now render and draw the orb_layer between game_layer and ui_layer in draw().

Read (E:\a327ex\Invoker\main.lua)

228    layer_draw(wall_area_layer)
229   layer_render(wall_layer)
230   layer_draw(wall_layer)
231   layer_render(symbol_layer)
232   layer_draw(symbol_layer)
233 
234   camera_attach(main_camera, game_layer)
235   p:draw(game_layer)
236   -- Floating effect text (spell names, fizzles, future damage numbers
237   -- and status applications). World-space, anchored to spawn position.
238   info_text_draw(game_layer)
239   camera_detach(main_camera, game_layer)
240   layer_render(game_layer)
241   layer_draw(game_layer)
242 
243   -- Cast HUD: screen-space, pinned to bottom-center with an 8px margin.
244   -- Permanent — empty slots render as chrome-only when no chord is being
245   -- built. Drawn on ui_layer (no camera attach) so it stays fixed
246   -- regardless of player position.
247   if run then
248     -- Resource orbs centered in the middle of the screen for iteration.
249     -- The shader paints the orb interior + border; the chrome overlay
250     -- (numeric readouts) is queued on the same layer after the shader
251     -- pass so it composites cleanly on top.
252     local orbs = {
253       {cx = width/2 - 50, cy = height/2, radius = 28, fill = run.hp/run.hp_max, color = red,        cur = run.hp, max = run.hp_max},
254       {cx = width/2 + 50, cy = height/2, radius = 28, fill = run.mp/run.mp_max, color = light_blue, cur = run.mp, max = run.mp_max},
255     }
256     ui_orbs_apply(orb_layer, orbs)
257     ui_orbs_chrome(orb_layer, orbs)
... [10 more lines]

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

The orb_layer needs its commands queued BEFORE its render call. Let me restructure so the orb block is above the render+draw, and the ui_layer block stays where it is.

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Quick check that the orb chrome uses an appropriate font for big numbers — Awesome 9 (header_l) may render digits as icons.

Bash (grep -nE "font.*[0-9]|font_register|ui_font" E:/a327ex/Invoker/info_text.lua E:/...)

E:/a327ex/Invoker/info_text.lua:75:  local f = log_font or ui_font_body
E:/a327ex/Invoker/anchor/memory.lua:164:  local line_h = (font.height or 11) + 1

Ready to compile/run. Quick summary of what's in:

assets/orb_liquid.frag — Fragment shader, full-pass on orb_layer. Per pixel: passthrough outside both orbs, dark border ring on outermost 1.5 px, animated sine ripple at the waterline, vertical gradient + specular highlight + edge vignette inside the liquid, dark backdrop + edge vignette inside the empty section.
main.lua — New orb_layer between game and ui; loads orb_liquid_shader; queues ui_orbs_apply + ui_orbs_chrome per frame.
ui.lua — ui_orbs_apply(layer, orbs) sets uniforms (positions in screen coords, radius, fill 0–1, base color as vec4) and applies the shader. ui_orbs_chrome(layer, orbs) draws the cur value (header_l) + "/ max" (body, muted) centered on each orb.
Both orbs sit at screen center, radius 28, 100 px apart. Bottom-left chips kept as a comparison anchor.
[ and ] still decrement HP and MP so you can watch the surface drop and ripple.

The numeric readout uses Awesome 9 for the big number — if it renders digits as icons we'll swap to body and add a bigger Lana variant.

Give me keys to decrease/increase HP and mana.

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Done. Each key step is 10 HP/MP, clamped to [0, max].

HP: [ decrease, ] increase
Mana: ; decrease, ' increase

Use other buttons for mana, I'm on a Brazilian keyboard. Let's also improve the visuals. Try to match the visuals of SNKRX more, most of the UI kits get this right, although some go in higher-res directions. The current gradients you have for the liquid feel a bit incongruent with the rest of the game. The game only has gradients for the walls but they're way more subtle and over longer distances. The various polychrome, etc effects are also noise that's more subtle than this. We want the liquid to look like maybe one of those effects, subtle cloud noise kind of things, instead of these abrupt gradients around the orb direction that try to give it a glass effect. As for the liquid itself, we could do a 1D wave simulation that actually behaves like a liquid, I have this going on in one of my older projects, it was called something something wave, it's probably in one of the reference projects in E:/a327ex, if you can find it tell me because I can look for it in my archives.

Switching mana keys first (commas are universal on any layout), then I'll search for the wave project.

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Now hunting for the wave project.

Bash (ls E:/a327ex/ 2>&1)

Anchor/
Anchor-YueScript/
Anchor-lua/
Anchor2/
BYTEPATH/
Danpenzero/
Invoker/
Null/
Psyonia/
RETICLE/
RTFX/
Ricochet/
SNKRX/
a327ex-site/
a327ex.com-blog/
anchor.blot.im/
cute_action.mp3
cute_action.wav
emoji-aimer/
emoji-ball-battles/
emoji-ball-battles-generated/
emoji-ball-bounce/
it-follows/
lane-battle/
ld54-remake/
meal_prep.md
music-theory/
neural-networms/
notes/
petri/
... [8 more lines]

Grep (wave)

Found 50 files limit: 50
Invoker\wall_symbols.lua
Invoker\wall_patterns.lua
emoji-aimer\anchor\shake.lua
petri\anchor\shake.lua
neural-networms\anchor\shake.lua
Invoker\maps\tutorial.lua
ld54-remake\vertical_stack.lua
ld54-remake\color_pop.lua
Invoker\map_generation.lua
emoji-ball-bounce\anchor\shake.lua
Ricochet\anchor\shake.lua
RETICLE\anchor\shake.lua
Invoker\v1\main.lua
thalien-lune\anchor\shake.lua
Anchor2\app\anchor\shake.lua
lane-battle\anchor\shake.lua
Invoker\v1\assets\shockwave.frag
Invoker\v1\assets\wall_gradient.frag
Invoker\anchor\shake.lua
Anchor2\arena\anchor\shake.lua
Anchor2\framework\anchor\shake.lua
ld54-remake\anchor\shake.lua
Anchor2\reference\Anchor_v1\anchor_v1\shake.lua
Anchor\framework\anchor\shake.lua
Danpenzero\anchor\shake.lua
Invoker\v1\anchor\shake.lua
Invoker\assets\shockwave.frag
Invoker\reference\yue\anchor\shake.yue
emoji-ball-battles\anchor\shake.lua
... [21 more lines]

Found 17 files
emoji-aimer\anchor\shake.lua
petri\anchor\shake.lua
neural-networms\anchor\shake.lua
emoji-ball-bounce\anchor\shake.lua
Ricochet\anchor\shake.lua
RETICLE\anchor\shake.lua
Invoker\v1\main.lua
thalien-lune\anchor\shake.lua
Anchor2\app\anchor\shake.lua
lane-battle\anchor\shake.lua
Invoker\anchor\shake.lua
Anchor2\arena\anchor\shake.lua
Anchor2\framework\anchor\shake.lua
ld54-remake\anchor\shake.lua
Anchor-YueScript\reference\examples\examples.yue
Anchor-lua\reference\examples\examples.yue
Anchor\reference\examples\examples.yue

Grep (wave)

156-
157-if #arena\A 'enemy' == 0
158:  arena\next_wave!
159-
160-

Grep (wave)

Found 10 files
BYTEPATH\globals.lua
BYTEPATH\love\changes.txt
BYTEPATH\objects\effects\ShockwaveDisplacement.lua
BYTEPATH\objects\effects\WaverPreAttackEffect.lua
BYTEPATH\objects\enemies\Waver.lua
BYTEPATH\objects\Explosion.lua
BYTEPATH\objects\Player.lua
BYTEPATH\rooms\Classes.lua
BYTEPATH\rooms\Director.lua
BYTEPATH\rooms\Stage.lua

Bash (ls "E:/a327ex/private/" 2>&1 | head -40; echo "---"; ls "E:/a327ex/reference/" 2...)

gemini-personality-conversation-2-summary.md
website-migration-session.md
---
experimental-visual-games-steam-research.md
noita-wand-composition-system.md
personal-knowledge-capture-research.md
poe2-meta-modifier-mechanics.md
star-of-providence-research.md
switch_sports_goal.mp4
switch_sports_goal_2.mp4
switch_sports_goal_2_frames/
switch_sports_goal_frames/
switch_sports_goal_frames_full/
tomodachi_life_border.mp4
tomodachi_life_border_frames/
tomodachi_life_button_1.mp4
tomodachi_life_button_1_frames/
tomodachi_life_button_2.mp4
tomodachi_life_button_2_frames/
visual_effects.md

Found 20 files limit: 20
Invoker\main.lua
Invoker\ui.lua
emoji-aimer\main.lua
emoji-ball-bounce\reference\main.lua
Invoker\v1\main.lua
emoji-ball-battles\main.lua
emoji-ball-battles\reference\yue\main.yue
Invoker\reference\Balatro\localization\en-us.lua
Invoker\reference\Balatro\functions\common_events.lua
Invoker\reference\Balatro\game.lua
Invoker\reference\Balatro\challenges.lua
Invoker\reference\Balatro\card.lua
Invoker\reference\Balatro\localization\zh_TW.lua
Invoker\reference\Balatro\localization\zh_CN.lua
Invoker\reference\Balatro\localization\ru.lua
Invoker\reference\Balatro\localization\pt_BR.lua
Invoker\reference\Balatro\localization\pl.lua
Invoker\reference\Balatro\localization\nl.lua
Invoker\reference\Balatro\localization\ko.lua
Invoker\reference\Balatro\localization\ja.lua

Grep (tension|spread.*neighbor|wave_pts|surface_pts|wave_y)

No matches found

Bash (find E:/a327ex -maxdepth 3 -type d -iname "wave" 2>/dev/null)

(Bash completed with no output)

Bash (find E:/a327ex/Anchor/reference E:/a327ex/Anchor/framework -maxdepth 4 -type f...)

(Bash completed with no output)

Couldn't find a "something wave" project in E:/a327ex — searched folder names, top-level files, code patterns (spring/spread/tension/neighbor on surface arrays), and the reference subdirs. The closest hits were enemy-wave names (next_wave, Waver.lua in BYTEPATH) and shockwave/shake modules, none of which are 1D liquid surface sims. If you can point me at the archive folder once you find it, I'll lift the simulation directly.

In the meantime, the visual fix. Reading SNKRX-style effects + the existing wall_gradient organic_field pattern, then rewriting the shader to drop the directional gradient and add subtle cloud noise instead.

Read (E:\a327ex\Invoker\reference\snkrx_style_effects.md)

1    # SNKRX-Style Visual Effects Guide
2   
3   A comprehensive catalog of visual effects from 8 projects in the SNKRX lineage, synthesized into a unified style guide with Anchor framework translations.
4   
5   **Original SNKRX-likes** (gray background, drop shadow): SNKRX, BUILDRX2, 112020, 001-No-title
6   **Dark variant** (black background, no shadow): Antitower, 004, 239A5, Rekoyam Tsohg
7   
8   ---
9   
10  ## Table of Contents
11  
12  1. [Design Philosophy](#1-design-philosophy)
13     - [Core Principles (All Projects)](#core-principles-all-projects)
14     - [Dark Variant Differences](#dark-variant-differences)
15  2. [Effect Primitives](#2-effect-primitives)
16     - [HitCircle](#2a-hitcircle)
17     - [HitParticle](#2b-hitparticle)
18     - [Hit Effect (Spritesheet)](#2c-hit-effect-spritesheet)
19     - [Spring](#2d-spring)
20     - [Flash](#2e-flash)
21     - [Trail](#2f-trail)
22     - [Blink-out](#2g-blink-out)
23     - [Image Particle](#2h-image-particle)
24     - [Wall Hit Particle](#2i-wall-hit-particle)
25     - [Block Death Effect](#2j-block-death-effect)
26     - [Fading Shoot Capsule](#2k-fading-shoot-capsule)
27     - [Shoot Effect (Muzzle Tracker)](#2l-shoot-effect-muzzle-tracker)
28  3. [Composite Formulas](#3-composite-formulas)
29     - [Hit Formula](#3a-hit-formula)
30     - [Death Formula](#3b-death-formula)
... [50 more lines]

Rewriting the shader: dropping the directional gradient + specular + edge vignette; adding subtle organic cloud noise (same family as wall_gradient.frag) plus a clean waterline strip. Keeping the existing sine ripple as a placeholder for the wave sim you'll port over.

Write (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Done. Changes:

Mana keys → , and . (universal across all keyboard layouts).
Wave project search: nothing matched in E:/a327ex — checked top-level project names, all subdirectories (no folder with wave/liquid/ripple/water), and code patterns (1D surface sims using neighbor-spring propagation). The "wave" hits I found were unrelated: enemy-wave systems in BYTEPATH (Waver.lua, next_wave), shockwave radial pulse shaders, and the shake.lua modules in every project. Point me at the archive when you find it and I'll wire the sim in.
Shader rewrite — SNKRX flat-color discipline:
- Dropped the vertical brightness gradient
- Dropped the specular highlight band under the surface
- Dropped the edge vignette
- Added subtle organic cloud noise (lifted organic_field from wall_gradient.frag), modulating the liquid color by ~±6%
- The empty section also uses the same noise so the texture reads continuous through the surface
- Surface is now a thin lighter strip directly on the waterline so it reads as a discrete pixel line
- Border ring unchanged (1.5 px dark ring)
- Sine ripple still in as the placeholder surface displacement — wave sim swap goes there

Yes, this looks better, now let's just try the different types of noise effects we have going for the walls/abilities in the game, the edition shaders. Let's try the different variations we have there. Not in terms of color, but in terms of how they make the otherwise flat color look a bit noisy/cloudy with what's going on in the background. I also added tidal_waver.lua to reference, which contains the wave simulation code from that game.

Reading the wave sim first, then the edition shaders to see what noise variations exist.

Bash (ls E:/a327ex/Invoker/reference/tidal_waver.lua && wc -l E:/a327ex/Invoker/refere...)

E:/a327ex/Invoker/reference/tidal_waver.lua
1084 E:/a327ex/Invoker/reference/tidal_waver.lua

5:  an:anchor_start('Tidal Waver', 480, 270, 3, 3, default_display(), 'tidal_waver')
133:  self.water_spread_value = 0.45
140:        left_water_spring_deltas[index] = self.water_spread_value*(self.water_springs.children[index].y - self.water_springs.children[back_index].y)
142:        right_water_spring_deltas[index] = self.water_spread_value*(self.water_springs.children[index].y - self.water_springs.children[next_index].y)
149:    self.water_surface = {}
152:      table.insert(self.water_surface, spring.x - spring.w/2)
153:      table.insert(self.water_surface, left_spring.y)
156:        table.insert(self.water_surface, spring.x + spring.w/2)
157:        table.insert(self.water_surface, right_spring.y)
160:    water_particles:gradient_polyline(self.water_surface, false, 10, an.colors.blue[0])
277:  self.previous_vx, self.previous_vy = 0, 0
280:  self.last_gained_velocity_from_diving = an.time
292:  self:stats_set('movement_speed', 7, 0, 20) -- base velocity under water
293:  self:stats_set('acceleration', 10, 0, 20) -- how much velocity the player gains on various triggers
299:  self.movement_speed_to_base_velocity = {6, 28, 40, 50, 57, 63, 68, 72, 76, 79.5, 82.5, 85.35, 88, 90, 92.5, 94.5, 96.5, 98.32, 100, 101.69}
300:  self.movement_speed_to_base_velocity[0] = 0
301:  self.base_v = self.movement_speed_to_base_velocity[self.stats.movement_speed.x]
305:  self.acceleration_to_dive_velocity_gain = {0.025, 0.0712, 0.098, 0.1174, 0.1323, 0.1444, 0.1547, 0.1636, 0.1714, 0.1785, 0.1848, 0.1906, 0.1959, 0.2009, 0.205, 0.21, 0.2138, 0.2179, 0.2213, 0.2247}
306:  self.acceleration_to_dive_velocity_gain[0] = 0
311:  self.acceleration_to_waving_velocity_gain = {0.0277, 0.0439, 0.0554, 0.0643, 0.0716, 0.0778, 0.0831, 0.0878, 0.0921, 0.1, 0.105, 0.11, 0.115, 0.12, 0.125, 0.13, 0.135, 0.14, 0.145, 0.15}
312:  self.acceleration_to_waving_velocity_gain[0] = 0
318:    self:shoot(math.angle_to_point(self.x, self.y, target.x, target.y))
331:    local vx, vy = self:collider_get_velocity()
332:    local v = math.length(vx, vy)
334:    local time_elapsed_since_last_in_water = an.time - self.last_in_water_time -- use this to control added velocity, shouldn't add anything if has been in water recently at certain angles
339:      local dot = math.dot(math.cos(self.target_r), math.sin(self.target_r), math.normalize(vx, vy))
340:      if dot < 0.5 then -- if velocity vector and target vector aren't aligned up to a point then it's coded as a bad landing, which enters the player into a recovery animation (no input while on it)
342:        self.bad_landing_vx, self.bad_landing_vy = vx/2, vy/2
346:        self:timer_tween(recovery_time*0.25, self, {bad_landing_vx = 0, bad_landing_vy = self.bad_landing_vy/2}, math.linear, function()
351:            self.bad_landing_vx, self.bad_landing_vy = self.v*math.cos(target_r), self.v*math.sin(target_r)
... [32 more lines]

Read (E:\a327ex\Invoker\reference\tidal_waver.lua)

110    self:add(solid(an.w/2, an.h + 20, {w = 2*an.w, h = 40}))
111   self:add(object('players'))
112   self:add(object('enemies'))
113   self:add(object('projectiles'))
114 
115   self.players:add(player(an.w/2, 40))
116   self:timer()
117 
118   self:timer_every({2, 4}, function()
119     local x = an:random_float(20, an.w - 20)
120     if an:random_bool(20) then
121       for i = 1, an:random_int(2, 4) do
122         self.enemies:add(seeker(x, -64))
123       end
124     else
125       self.enemies:add(seeker(x, -64))
126     end
127   end)
128 --}}}
129 
130 --{{{ water
131   -- Water (https://code.tutsplus.com/make-a-splash-with-dynamic-2d-water-effects--gamedev-236t)
132   self.water_springs_count = 52
133   self.water_spread_value = 0.45
134   self:action(function(self, dt) -- advance water simulation as an action
135     local left_water_spring_deltas = {}
136     local right_water_spring_deltas = {}
137     for k = 1, 8 do
138       for i = 1, self.water_springs_count do
139         local index, back_index, next_index = i, array.get_circular_buffer_index(self.water_springs.children, i-1), array.get_circular_buffer_index(self.water_springs.children, i+1)
... [45 more lines]

Bash (cat E:/a327ex/Invoker/anchor/spring.lua 2>&1 | head -80)

--[[
  spring — procedural damped spring animation.

  Usage:
    self.spring = spring_new()                       -- in constructor
    spring_add(self.spring, 'scale', 1, 5, 0.5)      -- name, initial value, frequency, bounce
    spring_pull(self.spring, 'scale', 0.3)           -- apply impulse
    spring_update(self.spring, dt)                   -- in update

    -- Read current value:
    local s = self.spring.scale.x

  A default 'main' spring at value 1 is created on spring_new() for convenience.
  You typically use spring.main.x as a single pulsing value for hit flashes,
  click feedback, etc.

  Spring parameters:
    frequency - oscillations per second (higher = faster)
    bounce    - bounciness 0-1 (0=no overshoot, 1=infinite oscillation)
    bounce=0.5 is moderate overshoot, common for hit reactions.
]]

--[[
  spring_new()
  Creates a new spring container with a default 'main' spring at value 1.
]]
function spring_new()
  local s = {
    _names = {},
  }
... [49 more lines]

Bash (grep -n "spring_update|spring_pull|spring_set|target_x|k =|d =|.v " E:/a3...)

7:    spring_pull(self.spring, 'scale', 0.3)           -- apply impulse
8:    spring_update(self.spring, dt)                   -- in update
47:  local k = (2*math.pi*frequency)^2
48:  local d = 4*math.pi*(1 - bounce)*frequency
51:    target_x = x,
53:    k = k,
54:    d = d,
59:  spring_pull(s, name, force, [frequency], [bounce])
63:function spring_pull(s, name, force, frequency, bounce)
67:    sp.k = (2*math.pi*frequency)^2
68:    sp.d = 4*math.pi*(1 - (bounce or 0.5))*frequency
74:  spring_set_target(s, name, value)
77:function spring_set_target(s, name, value)
78:  if s[name] then s[name].target_x = value end
86:  threshold = threshold or 0.01
89:  local dx = sp.x - sp.target_x
97:  spring_update(s, dt)
101:function spring_update(s, dt)
105:    local a = -sp.k*(sp.x - sp.target_x) - sp.d*sp.v
106:    sp.v = sp.v + a*dt

Read (E:\a327ex\Invoker\anchor\spring.lua)

95   
96  --[[
97    spring_update(s, dt)
98    Advances all springs by dt. Call once per frame for each spring container you own.
99    Standard damped spring equation: a = -k*(x - target) - d*v
100 ]]
101 function spring_update(s, dt)
102   local names = s._names
103   for i = 1, #names do
104     local sp = s[names[i]]
105     local a = -sp.k*(sp.x - sp.target_x) - sp.d*sp.v
106     sp.v = sp.v + a*dt
107     sp.x = sp.x + sp.v*dt
108   end
109 end
110

Found the wave sim — tidal_waver.lua:130-185. Row of N springs (52 there) with per-frame velocity-propagation to left/right neighbors (spread_value = 0.45, 8 sub-iterations/frame). Standard damped-harmonic springs with neighbor coupling. I'll port that as the next step — for now let me add the edition-noise variations with a cycle key.

Write (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Now Lua wiring: cycle key, mode global, uniform setter, and an on-screen indicator.

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Now thread the mode into the shader uniform inside ui_orbs_apply, and add a small mode-label readout above the orbs.

Edit (E:\a327ex\Invoker\ui.lua)

The file E:\a327ex\Invoker\ui.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Adding a label above the orbs so you know which mode you're looking at while cycling.

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Ready to test. Press F11 to cycle through 6 noise modes:

0 — static RGB (current default — flat base ± organic_field scaled noise)
1 — HSL L breath (lightness cycles via res cosine; smooth brighter/darker patches over time)
2 — HSL S breath (saturation cycles; more/less vivid patches)
3 — HSL L+S combined (both modulated at lower amplitude — richest cloudy look)
4 — two-octave field (mixes a faster smaller-scale field into the main one — more granular detail)
5 — hue-locked dynamic shimmer (closest to the wall_gradient edition 23 noise; small hue wobble + sat boost)

The mode name is rendered above the orbs as you cycle so you don't lose track.

The 1D wave-sim port is queued — tidal_waver.lua:130-185 has everything: 52-spring row, spread_value = 0.45, 8 sub-iterations per frame, neighbor-velocity propagation, spring physics frequency 77 / damping 6. Once you've picked a noise mode I'll port it (downscaled to ~16-24 points per orb, swapped in for the placeholder sin(local.x * 6.0 + ...) surface term).

F11 is also fullscreen, use another key, but also it doesn't seem to be changing the insides of the orb at all.

Bash (grep -n "shader_set_int|set_int" E:/a327ex/Invoker/anchor/layer.lua)

69:  shader_set_int = layer_shader_set_int,
227:function layer_shader_set_int(lyr, shader, name, value)
228:  eng.shader_set_int(lyr_handle(lyr), shader, name, value)

Two fixes: change F11 → n (free, mnemonic), and switch the uniform to float-cast-to-int (matches the codebase convention in wall_gradient.frag, which uses uniform float u_edition + int(u_edition + 0.5)). I also suspect the visual modulations may all be reading as "subtle = identical" because the amplitudes are too close; bumping them so modes are clearly distinguishable.

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\ui.lua)

The file E:\a327ex\Invoker\ui.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Done. Press n to cycle. Amplitudes are now ~2.5–3× higher than before so modes are clearly distinguishable. The mode-name label above the orbs still updates as you cycle, so even if the visual difference is small you can confirm the cycle is firing. If you press n and the label changes but the orb interior still looks identical, that means the uniform isn't binding and we'll need to debug the shader pipeline next.

The colors now change slightly but I see no actual cloud/noise variation within the orb itself.

The bug is the noise frequency. organic_field was being sampled at pix * 0.15, which works for walls (which span the full 480 px screen → ~3.7 noise cycles). The orb is only 56 px wide → 0.43 cycles → the noise looks effectively constant across the orb, so all modes just shift the whole orb's color uniformly. Fix: scale the input by ~8× so 2-3 noise lumps actually fit inside the disc.

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Try again. Three changes in this revision:

Field scale 0.15 → 1.2 — was the root cause. The orb now samples 2-3 noise lumps across its diameter instead of less than half a cycle.
Modes 1-3 now drive on field directly instead of res. res = 0.5 + 0.5*cos(... + field*π) smooths out a lot of the spatial variation; using field directly preserves the cloud pattern.
Mode 4 high-frequency layer bumped to pix * 2.8 so the small-scale field is actually small-scale.

You should now see distinct cloudy patches inside each orb when cycling through modes 0-4. Mode 5 (hue shimmer) keeps res for the temporal breath since hue wobbles benefit from the smoother cosine pulse.

Static RGB and hue shimmer look the best. Can we have the noise move instead of being static for those?

The organic_field function does take u_time but its internal divisors (/143.634 etc.) make the drift extremely slow — at the wall's huge spatial scale that looks correct, but for a 56-px orb it reads as static. Fix: add an explicit time-driven offset to the sample position so the noise field slides visibly across the orb.

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Should now visibly drift across the orb. Both mode 0 and mode 5 will animate (they both consume field); the other modes get the motion too, but that's probably an improvement for them as well. If the drift speed feels wrong, the constants are u_time * 10.0 (x) and u_time * 6.0 (y) — easy to tune.

Let's do static RGB for HP and hue shimmer for mana. Also implement the actual wave simulation too.

Let me check what's available for passing array data into shaders — that's the key architectural question for the wave sim.

39:shader_set_vec2_immediate(shader, name, x, y)
40:shader_set_vec4_immediate(shader, name, x, y, z, w)
105:layer_shader_set_vec2(layer, shader, name, x, y)
106:layer_shader_set_vec4(layer, shader, name, x, y, z, w)

Two changes — per-orb noise mode, then the wave sim. Starting with the wave sim module since it's the bigger piece.

Write (E:\a327ex\Invoker\waves.lua)

File created successfully at: E:\a327ex\Invoker\waves.lua (file state is current in your context — no need to Read it back)

Now wire it into main.lua — require, per-orb state globals, update each frame, and trigger ripples when HP/MP change.

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\run.lua)

The file E:\a327ex\Invoker\run.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Now the orb spec in main.lua — add per-orb noise_mode and wave. And drop the global noise-mode cycler since the user's locked in.

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Now update ui_orbs_apply to set per-orb noise mode + wave uniforms.

Edit (E:\a327ex\Invoker\ui.lua)

The file E:\a327ex\Invoker\ui.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Now the shader — per-orb noise mode + surface array lookup replacing the placeholder sine ripple.

Write (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Quick tuning pass — the spring impulse force at 8 will be barely visible (k=2526 makes the spring quite stiff). Bumping to 80 so the first wave is clearly visible.

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Done. Summary:

Per-orb noise mode locked in: HP = mode 0 (static RGB), Mana = mode 5 (hue shimmer). The N cycle key is gone.

Wave simulation (waves.lua):

16 spring-points per orb, packed into 4 vec4 uniforms for the shader
Spring physics: f=8 Hz, bounce=0.65 (underdamped, rings visibly without sustaining forever)
Per-frame: spring step → 4 sub-iterations of neighbor velocity coupling (spread = 0.45)
Endpoints clamp instead of wrapping — waves bounce off "the orb wall" rather than circulating

Wiring:

hp_wave and mp_wave allocated once at boot
Updated each frame in update()
Reset on run_reset (flat surface on death/restart)
HP/MP debug keys ([ ] , .) now also drop a velocity impulse at a random surface position so you can watch the ripple propagate

Shader:

New surface_offset_px helper unpacks the 4 vec4s and lerps between the two nearest points
Surface displacement (in px) is added to the waterline position, divided by radius for local-space conversion
Old sin-ripple placeholder removed

Press [ ] , . to drop ripples and watch the wave bounce around inside each orb. If the wave amplitude is too aggressive or too subtle, the impulse force (currently 80) and spring frequency (8 Hz) in waves.lua are the two main knobs.

The current bounce is too weak and not visible at all.

Numbers are the issue. With f=8 Hz (k≈2526) and an impulse of 80, peak displacement at the impulse point works out to ~0.8 px on a 28-px-radius orb — that's 2.8% of orb height, basically invisible. Three knobs to turn: lower stiffness, less damping, bigger impulse.

Edit (E:\a327ex\Invoker\waves.lua)

The file E:\a327ex\Invoker\waves.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

New tuning:

	before	after
spring frequency	8 Hz (k≈2526)	3 Hz (k≈355)
bounce	0.65	0.85
spread	0.45	0.50
sub-iterations	4	6
impulse force	80	150

Predicted peak displacement: ~8 px (~28% of orb radius). The wave should now be loud and clearly bounce around for ~1 s before settling. If it's now too aggressive, the impulse force in main.lua (150) is the easiest dial — drop to 80-100. If the oscillation feels jittery, raise frequency back to 4-5 Hz.

Higher initial impulse, lower bounce, slower spread.

Edit (E:\a327ex\Invoker\waves.lua)

The file E:\a327ex\Invoker\waves.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated. All occurrences were successfully replaced. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated. All occurrences were successfully replaced. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated. All occurrences were successfully replaced. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated. All occurrences were successfully replaced. (file state is current in your context — no need to Read it back)

Done. New tuning:

	before	after
impulse force	150	350
bounce	0.85	0.60
spread	0.50	0.20

So: a much bigger initial drop (~18 px peak displacement at the impact point), settling faster without prolonged ringing, with the wave traveling more slowly across the surface — should read more like "a single big drop hits and ripples out gradually" instead of the surface vibrating uniformly.

Hmmm, the tidal waver example had some code that made sure the wave looked correct, I think by spawning multiple waves or by doing something to the surrounding of the impulse, but I don't remember what. Currently the impulse doesn't look like it's spreading properly, and I think there was some additional way I made it spread, please look in the source and try to figure it out.

131:  -- Water (https://code.tutsplus.com/make-a-splash-with-dynamic-2d-water-effects--gamedev-236t)
132:  self.water_springs_count = 52
135:    local left_water_spring_deltas = {}
136:    local right_water_spring_deltas = {}
138:      for i = 1, self.water_springs_count do
139:        local index, back_index, next_index = i, array.get_circular_buffer_index(self.water_springs.children, i-1), array.get_circular_buffer_index(self.water_springs.children, i+1)
140:        left_water_spring_deltas[index] = self.water_spread_value*(self.water_springs.children[index].y - self.water_springs.children[back_index].y)
141:        self.water_springs.children[back_index].springs.water.v = self.water_springs.children[back_index].springs.water.v + left_water_spring_deltas[index]
142:        right_water_spring_deltas[index] = self.water_spread_value*(self.water_springs.children[index].y - self.water_springs.children[next_index].y)
143:        self.water_springs.children[next_index].springs.water.v = self.water_springs.children[next_index].springs.water.v + right_water_spring_deltas[index]
146:  end, 'water_springs_update')
150:    for i, spring in ipairs(self.water_springs.children) do
151:      local left_spring = self.water_springs.children[array.get_circular_buffer_index(self.water_springs.children, i-1)]
154:      if i == #self.water_springs.children then
155:        local right_spring = self.water_springs.children[array.get_circular_buffer_index(self.water_springs.children, i+1)]
161:  end, 'water_springs_late_update')
164:  local spring_w = an.w/self.water_springs_count
165:  self.water_spring_w = spring_w
166:  self:add(object('water_springs'))
167:  for i = 1, self.water_springs_count do
168:    self.water_springs:add(object():build(function(self)
181:      local left_spring = self.water_springs.children[array.get_circular_buffer_index(self.water_springs.children, self.index-1)]
213:function arena:get_closest_water_spring(x)
215:  for i, spring in ipairs(self.water_springs.children) do
222:  return self.water_springs.children[min_i]
236:  Creates a splash with a given width around position x. f should be a value in the ~[100, 1000] range for better results.
240:function arena:splash(f, x, width)
241:  local springs = self.water_springs.children
242:  local i = self:get_closest_water_spring(x).index
281:  self.last_spawned_splash = an.time
... [52 more lines]

Read (E:\a327ex\Invoker\reference\tidal_waver.lua)

230      end
231   end
232   return enemies
233 end
234 
235 --[[
236   Creates a splash with a given width around position x. f should be a value in the ~[100, 1000] range for better results.
237   The width represents how many springs will be pulled at the same time, as that creates a more realistic water effect.
238   Springs to be pulled will be chosen based on position x. Width values in the ~[3, 5] range tend to have better results.
239 --]]
240 function arena:splash(f, x, width)
241   local springs = self.water_springs.children
242   local i = self:get_closest_water_spring(x).index
243   if width % 2 == 0 then
244     local w = width/2
245     for j = i-w+1, i+w do
246       springs[array.get_circular_buffer_index(springs, j)].springs.water.v = f
247     end
248   else
249     local w = math.floor(width/2)
250     for j = i-w, i+w do
251       springs[array.get_circular_buffer_index(springs, j)].springs.water.v = f
252     end
253   end
254 end
255 --}}}
256 
257 --{{{ player 1
258 player = class:class_new(object)
259 function player:new(x, y, args)
... [25 more lines]

Found it. In tidal_waver.lua:240, arena:splash(f, x, width) perturbs width adjacent springs at once (3-5 wide), not a single point. And it sets their velocity to f instead of adding to it. Quote from the comment: "The width represents how many springs will be pulled at the same time, as that creates a more realistic water effect... Width values in the ~[3, 5] range tend to have better results."

A single-point impulse looks pinched because the propagation step alone can't widen it fast enough — splashes already start wide. Porting that as waves_splash.

Edit (E:\a327ex\Invoker\waves.lua)

The file E:\a327ex\Invoker\waves.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Three things now match tidal_waver:

Multi-point splash (width = 4) — 4 adjacent springs perturbed at once. This is the missing piece that was making your ripple look pinched.
Velocity assignment, not accumulation — v = force rather than v += force. Rapid presses don't compound; each splash overrides the impacted region.
Force range — dropped from ±350 to ±250 since 4 springs now share the energy.

The old waves_pull_at is still there if you want single-point impulses for a different effect later (e.g. a tiny droplet vs a big slosh). Spring tuning (f=3, bounce=0.6, spread=0.2, sub_iters=6) is unchanged.

Should impulse up always, the wave should spread across the orb's surface, such that each impulse should always move the whole orb in some way, so whatever parameters need to be changed, do that. I want more spread, slower resolution, changing the entire surface of the orb.

Two-pronged change. The wave was failing to reach the edges because (a) spread was too low so energy propagated slowly, and (b) damping was too high so it died before getting there. Combined with a wider initial splash:

Edit (E:\a327ex\Invoker\waves.lua)

The file E:\a327ex\Invoker\waves.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Five knobs moved in the same direction (toward "loud, slow, spreads everywhere"):

	before	after
impulse force	250	400
splash width	4	6 (~37% of the orb)
spring frequency	3 Hz	2 Hz (slower oscillation)
bounce	0.6	0.85 (less damping)
spread coupling	0.20	0.55
sub-iterations/frame	6	8

Predicted: ~25-30 px peak at the impact point, wave reaching the orb edges within ~3-4 frames at still ~half amplitude, and surface motion sustained for ~1.5 s. The whole orb should clearly slosh on every press. If it feels too dramatic, the easiest dial is the impulse force.

Other values seem right, initial impulse is too big.

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated. All occurrences were successfully replaced. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated. All occurrences were successfully replaced. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated. All occurrences were successfully replaced. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated. All occurrences were successfully replaced. (file state is current in your context — no need to Read it back)

Halved to ±200. Predicted peak ~16 px (~57% of orb radius) instead of ~32 px. The other knobs stay where they are.

0.75 bounce, 0.4 spread coupling

Edit (E:\a327ex\Invoker\waves.lua)

The file E:\a327ex\Invoker\waves.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Done. Bounce 0.85→0.75, spread 0.55→0.40.

Great, let's make the border for both grey like other UI elements, let's place the orbs on bottom left and right of the screen, let's add a function that "deals damage" to each orb. This does damage on a certain location of it, and the impulse is pulled from that location. The orb also wobbles a little and rotate a little (spring on scale and rotation) when this hit happens, let's try the hit effect as well. Impulse force and wobble force should also be tied to how big the hit is based on a percentage of life/mana it takes away. Give me two keys, one to hit HP one to hit mana. HP/mana being refreshed up should not follow these same rules and should just move up directly with lower impulse but without hit effects.

Big change set — wraps everything per-orb in a new state container. Showing the design first; if it looks right I'll write it all.

orb.lua (new module) — wraps wave + scale spring + rotation spring per orb: ```lua function orb_new() local o = {wave = waves_new(), springs = spring_new()} spring_add(o.springs, 'rotation', 0, 8, 0.5) -- f=8 bouncy snap return o end

function orb_update(o, dt) waves_update(o.wave, dt) spring_update(o.springs, dt) end

function orb_reset(o) waves_reset(o.wave) o.springs.main.x, o.springs.main.v = 1, 0 o.springs.rotation.x, o.springs.rotation.v = 0, 0 end

-- Hit: damage at location, with wobble/rotation scaled to damage%. function orb_hit(o, t01, damage, max_value) local pct = damage/max_value waves_splash(o.wave, t01, -200pct, 6) spring_pull(o.springs, 'main', -0.5pct) local dir = math.random() < 0.5 and -1 or 1 spring_pull(o.springs, 'rotation', 0.4pctdir) end

-- Refresh: small upward wave impulse, no wobble/rotation. function orb_refresh(o, t01, refresh, max_value) local pct = refresh/max_value waves_splash(o.wave, t01, 80*pct, 4) end ```

main.lua — bind the four keys, repurpose:

[ → orb_hit(hp_orb, math.random(), 10, 100) (10% hit)
] → orb_refresh(hp_orb, math.random(), 10, 100) (10% refresh, no wobble)
, → orb_hit(mp_orb, ...)
. → orb_refresh(mp_orb, ...)

Position to bottom corners: cx = 36, cy = height - 36 (left), and cx = width - 36 for right.

shader — three new uniforms:

u_border_color (global vec4) — set from panel_border in Lua
u_orbN_scale (per orb) — multiplies effective radius
u_orbN_rotation (per orb) — rotates local coords inside sample_orb

Layout check: Bottom-left free strip is x=0..80 (cast HUD starts at 80). Orb at cx=36, radius 28 spans 8..64. Fits with 16 px margin to the primitive HUD. Bottom-right has way more room (x=316..480 free; orb at cx=444 spans 416..472). Cast HUD vertical center ≈ height-25 sits between the orbs cleanly.

Going to write this now.

Write (E:\a327ex\Invoker\orb.lua)

File created successfully at: E:\a327ex\Invoker\orb.lua (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

<tool_use_error>String to replace not found in file.
String:   -- HP/MP debug adjust. Each change drops a 4-wide splash at a
  -- pseudo-random spot on the surface — the splash semantics (set,
  -- not accumulate, across multiple adjacent springs) come from
  -- tidal_waver.lua's arena:splash and are what makes the wave
  -- shape read as "a drop hit the surface" rather than a pinched
  -- single-point spike.
  if run and is_pressed('debug_hp_down') then run.hp = math.max(0,          run.hp - 10); waves_splash(hp_wave, math.random(), -200, 6) end
  if run and is_pressed('debug_hp_up')   then run.hp = math.min(run.hp_max, run.hp + 10); waves_splash(hp_wave, math.random(),  200, 6) end
  if run and is_pressed('debug_mp_down') then run.mp = math.max(0,          run.mp - 10); waves_splash(mp_wave, math.random(), -200, 6) end
  if run and is_pressed('debug_mp_up')   then run.mp = math.min(run.mp_max, run.mp + 10); waves_splash(mp_wave, math.random(),  200, 6) end

  waves_update(hp_wave, dt)
  waves_update(mp_wave, dt)
(note: Edit also tried swapping \uXXXX escapes and their characters; neither form matched, so the mismatch is likely elsewhere in old_string. Re-read the file and copy the exact surrounding text.)</tool_use_error>

Read (E:\a327ex\Invoker\main.lua)

205  -- starts with a flat surface and no residual wobble.
206 hp_orb = orb_new()
207 mp_orb = orb_new()
208 
209 run_reset()  -- bootstrap the first run (loads tutorial; spell-system seeding TBD)
210 
211 function update(dt)
212   sync_engine_globals()
213   camera_update(main_camera, dt)
214   if is_pressed('reload_map') then load_map('tutorial') end
215   if is_pressed('toggle_mem') then memory_tracker_toggle(mem_tracker) end
216   if is_pressed('baseline_mem') then memory_tracker_capture_baseline(mem_tracker) end
217   if is_pressed('cycle_symbol') then cycle_wall_symbol() end
218   if is_pressed('cycle_pattern') then cycle_wall_pattern() end
219   if is_pressed('cycle_flow') then cycle_wall_flow() end
220   if is_pressed('cycle_palette') then cycle_wall_palette() end
221   if is_pressed('cycle_palette_back') then cycle_wall_palette_back() end
222   if is_pressed('toggle_ui_test') then ui_test_active = not ui_test_active end
223   if is_pressed('reset_run') then run_reset() end
224   -- HP/MP debug adjust. Each change drops a 4-wide splash at a
225   -- pseudo-random spot on the surface — the splash semantics (set,
226   -- not accumulate, across multiple adjacent springs) come from
227   -- tidal_waver.lua's arena:splash and are what makes the wave
228   -- shape read as "a drop hit the surface" rather than a pinched
229   -- single-point spike.

Read (E:\a327ex\Invoker\main.lua)

228    -- shape read as "a drop hit the surface" rather than a pinched
229   -- single-point spike.
230   if run and is_pressed('debug_hp_down') then run.hp = math.max(0,          run.hp - 10); waves_splash(hp_wave, math.random(), -200, 6) end
231   if run and is_pressed('debug_hp_up')   then run.hp = math.min(run.hp_max, run.hp + 10); waves_splash(hp_wave, math.random(),  200, 6) end
232   if run and is_pressed('debug_mp_down') then run.mp = math.max(0,          run.mp - 10); waves_splash(mp_wave, math.random(), -200, 6) end
233   if run and is_pressed('debug_mp_up')   then run.mp = math.min(run.mp_max, run.mp + 10); waves_splash(mp_wave, math.random(),  200, 6) end
234   waves_update(hp_wave, dt)
235   waves_update(mp_wave, dt)
236   primitives_input_update()
237   cast_input_update(dt)
238   info_text_update(dt)
239   -- [log] DISABLED. log_update(dt) advances the per-entry rank tween.
240   -- log_update(dt)
241   if current_map then current_map:update(dt) end
242   p:update(dt)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Now update orb positions, render data, and the run_reset hook.

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Also drop the bottom-left HP/MP chips — orbs replace them now.

Read (E:\a327ex\Invoker\main.lua)

277      }
278     ui_orbs_apply(orb_layer, orbs)
279     ui_orbs_chrome(orb_layer, orbs)
280   end
281   layer_render(orb_layer)
282   layer_draw(orb_layer)
283 
284   -- Cast HUD: screen-space, pinned to bottom-center with an 8px margin.
285   -- Permanent — empty slots render as chrome-only when no chord is being
286   -- built. Drawn on ui_layer (no camera attach) so it stays fixed
287   -- regardless of player position.
288   if run then
289     -- HP / MP stat chips, stacked vertically in the bottom-left corner
290     -- (HP on top, MP below). 8 px margin from the bottom and left edges,
291     -- 2 px gap between chips. Each chip is 14 px tall.
292     ui_stat_chip(ui_layer, 8, height - 38, red,        'HP ' .. run.hp .. '/' .. run.hp_max)
293     ui_stat_chip(ui_layer, 8, height - 22, light_blue, 'MP ' .. run.mp .. '/' .. run.mp_max)
294 
295     ui_cast_hud(ui_layer, width/2, height - 8, run.cast_slots)
296 
297     -- Primitive HUD: 4 owned-primitive slots directly to the left of
298     -- the cast HUD, bottom-aligned with the cast-slot row. Row total

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\run.lua)

The file E:\a327ex\Invoker\run.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Now ui.lua: add scale + rotation per orb, border color global, and read from state instead of wave.

Edit (E:\a327ex\Invoker\ui.lua)

The file E:\a327ex\Invoker\ui.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Now the shader — adds scale, rotation, border color, and applies the transforms.

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Grep (hp_wave|mp_wave)

No files found

Clean. Summary of what changed:

Border → grey: u_border_color uniform set from panel_border (matches every other UI element).

Position: orbs at bottom-left (cx=36) and bottom-right (cx=width-36), 8 px margin from screen edges. Old comparison chips removed.

Hit semantics ([ for HP, , for MP):

Wave splash impulse: -200 × damage_pct (negative = surface drops)
Scale wobble: -0.5 × damage_pct (squish — 50% at full damage)
Rotation kick: ±0.4 rad × damage_pct, random sign
All three scale linearly with damage / max

Refresh semantics (] for HP, . for MP):

Wave splash: +80 × refresh_pct (positive = surface lifts)
No scale wobble, no rotation kick

Spring tuning:

Scale spring: f=5, bounce=0.5 (default)
Rotation spring: f=8, bounce=0.5 (snappier)

Per single key press (10% of max), expected effects: ~5% squish, ~2.3° rotation kick, small splash. Spam to see scaling. The shader's hit-test now uses eff_radius = radius * scale so the disc actually shrinks when squished, and rotation is applied to local coords so the surface line tilts with the orb.

Give me random values for the hit instead of just 10%, since I wanna see what higher/lower values look like too.

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Done. Each press now rolls a fresh damage amount:

Hits ([, ,): math.random(5, 100) → 5%–100% range. Subtle nudges through to lethal-feeling slams on the same key.
Refreshes (], .): math.random(5, 40) → 5%–40% range. Refreshes feel calmer by design.

The visual effect always uses the intended magnitude — a 90-damage hit on a 30-HP orb still gives the full ~45° rotation kick + ~45% squish + big splash, even though only 30 HP actually came off.

Perfect, functionally it feels good. I just feel like the orb effect visual itself looks a bit too off compared to how the rest of the game will look. Can we try something like a dither effect on the noise? So that the gradient doesn't look like it changes normally, but it fits the pixelized style of the game and changes as a dither instead. I don't know if you can or know how to do that exactly.

Yes, this is a classic pixel-art trick — Bayer-matrix ordered dither. The continuous noise field still drives the modulation, but instead of producing a smooth gradient it gets quantized to 4 discrete color levels per pixel, with a fixed 4×4 threshold pattern deciding which two adjacent levels each pixel takes. The eye averages it out into the same gradient at a distance, but up close every pixel snaps to one of N discrete colors — fits the pixel-art look.

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Done. What you should see:

The orb interior now has a visible crosshatch/checker pattern transitioning between 4 discrete brightness levels (HP) or hue shifts (mana)
The pattern is locked to the screen-pixel grid (not the orb), so as the orb wobbles or rotates, the dither stays stable rather than swimming
The underlying field still drifts over time, so individual pixels flip between adjacent levels as the field crosses thresholds — gives the same "alive" feel without the smooth gradient

If the dither is too obvious (too many pixels of pattern visible at once), bump n_levels to 6 or 8 — finer color granularity, less visible pattern. If it's too subtle, drop to 3 — more dramatic discrete-level steps. If you want a tighter pattern, swap bayer4 for bayer2 (would need to add — 2×2 Bayer is [[0,2],[3,1]]/4).

Perhaps I want the HP and mana colors to be a bit more muted, as they look too bright currently. Give me a key to try out some variations. Also is there a way for me to try variations on the delta between colors for mana only? It's a bit too bright sometimes.

Two debug cycles, easy to lock in once you find what you want. Adding u_color_mute_sat and u_color_mute_lit (global, affects both orbs), and u_mode5_strength (only mana's hue shimmer).

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Now add the labels somewhere visible — putting them on the orb_layer so they're near the orbs.

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Now ui.lua sends the new uniforms:

Edit (E:\a327ex\Invoker\ui.lua)

The file E:\a327ex\Invoker\ui.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Shader: declare uniforms, mute the base color before noise, scale the mode-5 hue delta.

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Now mute the base color inside sample_orb before noise application, and scale mode 5's hue delta.

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Done. Two new debug cycles:

c — global color mute (both orbs):

0: full bright (current)
1: 75% sat, 88% lit
2: 50% sat, 76% lit
3: 25% sat, 64% lit (very muted)

Each step pulls saturation by 25% (toward perceptual-luminance grey) and lightness by 12% (RGB multiplier).

v — mana hue shimmer strength:

0: 1.0× (current)
1: 0.5×
2: 0.25×
3: 0× (no hue shift — mana stays uniform light_blue)

Both labels at the top of the screen show the active variant. Once you find the combo you like, tell me the indices and I'll lock them in.

Between full bright and the single muting level for HP, mana remains the same. 0.25x hue shimmer strength for mana.

Locking it in. Per-orb color mute (since they differ now), mode5 strength fixed at 0.25, cycle keys + labels removed.

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\main.lua)

The file E:\a327ex\Invoker\main.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Now ui.lua — switch mute uniforms from global to per-orb, drop the mana hue strength uniform (will hardcode in shader).

Edit (E:\a327ex\Invoker\ui.lua)

The file E:\a327ex\Invoker\ui.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Shader: drop the global mute + mode5_strength uniforms, add per-orb mute uniforms, hardcode mana hue strength at 0.25.

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Update the two sample_orb calls in main() to pass the new args.

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Locked in:

HP: mute_sat = 0.875, mute_lit = 0.94 (halfway between full bright and the single mute step)
MP: mute_sat = 1.0, mute_lit = 1.0 (unchanged from before)
Mana hue shimmer: hardcoded to × 0.25

Cycle keys (c, v), variant globals (orb_color_variant, mana_hue_variant, MANA_HUE_STRENGTHS), and the top-of-screen debug labels are all gone. Easy to bring any of them back if you want to iterate again.

Color of / 100 should be white, spacing between top and bottom lines should be bigger, top line should be a bit up.

Edit (E:\a327ex\Invoker\ui.lua)

The file E:\a327ex\Invoker\ui.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Three changes:

/ 100 color: text_muted (140, 140, 145 — greyish) → white (200, 200, 200, the literal palette token)
Top line position: cy - 10 → cy - 16 (6 px higher)
Spacing: top line spans cy-16..cy; bottom line top edge at cy + 4, so there's a clear 4 px gap between them (used to overlap by ~4 px due to font heights)

They need to be closer together and the spacing between them should be in the very center of the orb.

Edit (E:\a327ex\Invoker\ui.lua)

The file E:\a327ex\Invoker\ui.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Big top moves from cy - 16 to cy - ui_font_header_l.height (same value but now adapts if the font height differs from 16), and small top moves from cy + 4 to cy + 1. Result: big ends at row cy-1, gap at row cy, small starts at row cy+1 — the 1-pixel gap is exactly at the orb's center.

They should be closer, and actually a bit below the vertical center of the orb.

Edit (E:\a327ex\Invoker\ui.lua)

The file E:\a327ex\Invoker\ui.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Big text top moved from cy - 16 to cy - 8 (down 8 px), small text from cy + 1 to cy + 5 (down 4 px). Bounding boxes now overlap by ~3 rows — the actual visible glyphs should sit tight together because both Awesome 9 and Lana 11 have padding around their cap/x heights. The block's visual center lands ~3 px below the orb center.

2 pixels up, and 1 extra pixel of spacing between them.

Edit (E:\a327ex\Invoker\ui.lua)

The file E:\a327ex\Invoker\ui.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Big top: cy-8 → cy-10. Small top: cy+5 → cy+4. Both lines shift up 2 px, and the small line gains 1 extra row of separation from the big (overlap goes from 3 rows to 2).

2 further pixels up.

Edit (E:\a327ex\Invoker\ui.lua)

The file E:\a327ex\Invoker\ui.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Both lines up 2 more px. Spacing between them stays the same.

When the orb is spring rotated and scaled it should also affect the text inside it.

Grep (layer_push|layer_pop)

E:\a327ex\Invoker\player.lua:90:  layer_push(layer, self.x, self.y, self.r + self.visual_r, self.sx*s, self.sy*s)
E:\a327ex\Invoker\player.lua:92:  layer_pop(layer)
E:\a327ex\Invoker\anchor\layer.lua:34:  layer_draw, layer_push, layer_pop, layer_clear, layer_get_texture, and every
E:\a327ex\Invoker\anchor\layer.lua:61:  push = layer_push,
E:\a327ex\Invoker\anchor\layer.lua:62:  pop = layer_pop,
E:\a327ex\Invoker\anchor\layer.lua:194:function layer_push(lyr, x, y, r, sx, sy)
E:\a327ex\Invoker\anchor\layer.lua:198:function layer_pop(lyr)
E:\a327ex\Invoker\ui.lua:420:  layer_push(lyr, cx, cy, rot, s_pop, s_pop)
E:\a327ex\Invoker\ui.lua:434:    layer_push(lyr, 0, 0, static_rot, s_fit, s_fit)
E:\a327ex\Invoker\ui.lua:436:    layer_pop(lyr)
E:\a327ex\Invoker\ui.lua:439:  layer_pop(lyr)
E:\a327ex\Invoker\ui.lua:474:    - Slot chrome + icon: drawn inside layer_push(cx, cy, scale) in
E:\a327ex\Invoker\ui.lua:476:    - Label chrome + text: drawn after layer_pop in world coordinates.
E:\a327ex\Invoker\ui.lua:513:    layer_push(lyr, cx, cy, 0, s_pop, s_pop)
E:\a327ex\Invoker\ui.lua:526:      layer_push(lyr, 0, 0, 0, s_fit, s_fit)

[Showing results with pagination = limit: 15]

Edit (E:\a327ex\Invoker\ui.lua)

The file E:\a327ex\Invoker\ui.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Done. layer_push(layer, cx, cy, rotation, scale, scale) sets up a transform centered on the orb, then the text is drawn at local coords (0, -12) and (0, 2). When the orb is at rest (scale=1, rotation=0) the text renders identically to before. On a hit, the text squishes and tilts with the disc.

Rotation should be a bit more obvious, scale can be a bit less strong on high percentages.

Edit (E:\a327ex\Invoker\orb.lua)

The file E:\a327ex\Invoker\orb.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Two knobs:

Rotation kick 0.4 rad → 0.7 rad at 100% damage (~23° → ~40°)
Scale squish -0.5 → -0.3 at 100% damage (scale drops to 0.7x instead of 0.5x)

10% hits scale proportionally: ~4° rotation, scale to 0.97x — barely visible, as before.

Make the bounce for rotation be like slower, it settles over a longer period.

Edit (E:\a327ex\Invoker\orb.lua)

The file E:\a327ex\Invoker\orb.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\orb.lua)

The file E:\a327ex\Invoker\orb.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Rotation spring: f=8, bounce=0.5 → f=3, bounce=0.85. Period goes from ~0.13 s to ~0.33 s and the damping ratio drops, so the orb should now visibly wobble through several tilt cycles before settling rather than snapping back fast.

Ah, I see, because it's an orb and it wasn't rotating the contents inside I wasn't seeing the effect. Change it back and apply the rotation/scale effect to the contents of the orb too, so the shader contents itself.

Two changes — rotation spring back to fast snap, and the shader noise+dither now sample in orb-local space so they rotate/scale with the disc.

Edit (E:\a327ex\Invoker\orb.lua)

The file E:\a327ex\Invoker\orb.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\orb.lua)

The file E:\a327ex\Invoker\orb.lua has been updated successfully. (file state is current in your context — no need to Read it back)

Now the shader. apply_noise takes a noise_uv instead of pix; in sample_orb we compute noise_uv = center + local*radius after the rotation step so noise + dither follow the orb's rotation and scale.

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Now compute noise_uv in sample_orb after the rotation step and pass it to apply_noise.

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Done. Summary:

Rotation spring reverted: f=3, bounce=0.85 → f=8, bounce=0.5 (snappy snap-back)
Shader noise + dither now sample in orb-frame coordinates: noise_uv = center + local * radius where local is rotated. The dithered cloud pattern and the surface waterline all rotate together with the disc on a hit. The math is set up so at rest (rotation=0, scale=1) the noise samples exactly the same positions as before — only hits change anything.

Shader compilation failed: 0(156) : error C1503: undefined variable "pix"

ERROR: main.lua:110: Failed to load effect shader: assets/orb_liquid.frag stack traceback: [C]: in function 'shader_load_file' main.lua:110: in main chunk

Read (E:\a327ex\Invoker\assets\orb_liquid.frag)

145          hsl.y = clamp(hsl.y + (field - 0.5) * 0.60, 0.0, 1.0);
146         return hsl2rgb(hsl);
147     }
148     if (mode == 3) {
149         vec3 hsl = rgb2hsl(base);
150         hsl.y = clamp(hsl.y + (field - 0.5) * 0.40, 0.0, 1.0);
151         hsl.z = clamp(hsl.z + (field - 0.5) * 0.25, 0.0, 1.0);
152         return hsl2rgb(hsl);
153     }
154     if (mode == 4) {
155         float field2 = organic_field(pix * 2.8 + drift, u_time * 0.7 + 11.0);
156         float combined = mix(field, field2, 0.4) - 0.5;
157         return base + base * combined * 0.50;
158     }
159     if (mode == 5) {
160         // Mana hue shimmer at 0.25× strength — locked in after the
161         // `v` cycle. Hue swings stay subtle so light_blue keeps
162         // identity even at the brighter dither levels.
163         vec3 hsl = rgb2hsl(base);
164         hsl.x = hsl.x + ((res - 0.5) * 0.10 + (field - 0.5) * 0.05) * 0.25;

Found it — mode 4 still references pix for its second-octave noise. Updating to use noise_uv for consistency.

Edit (E:\a327ex\Invoker\assets\orb_liquid.frag)

The file E:\a327ex\Invoker\assets\orb_liquid.frag has been updated successfully. (file state is current in your context — no need to Read it back)

Mode 4 (the unused two-octave variant) was the only remaining pix reference inside apply_noise. Should compile now.

Perfect, update any documentation files and let's end the session here, name it Invoker HP and Mana Orbs

Let me check what's in design.md and plan.md that needs updating.

Bash (grep -n "orb|HP|mana|Orb|Mana" E:/a327ex/Invoker/design.md E:/a327ex/Invoker...)

E:/a327ex/Invoker/design.md:11:1. **Live combinatorial spell-casting.** Player chords primitives (elements + formations) into spells in real time. Casting is a 3-key sequence; the same vocabulary recombines to produce dozens of distinct spells. No inventory, no orbs, no equipment — the build is which primitives you have, what modifiers are attached to them, and what passives + items you've accumulated.
E:/a327ex/Invoker/design.md:13:2. **Minesweeper-style meta-dungeon.** Between combats, the player navigates a procedurally generated grid of rooms with classic Minesweeper rules. Some rooms are mines (HP loss), some are combat encounters, some are shops or forges or treasure. Adjacent-mine numbers help deduce safe paths. The dungeon view replaces the older walkable hideout entirely.
E:/a327ex/Invoker/design.md:30:   - **Mine** — HP loss.
E:/a327ex/Invoker/design.md:47:- **Mana cost.** Spammable. Cost scales with chord complexity.
E:/a327ex/Invoker/design.md:61:- **Passives** apply globally to the player (Isaac-style). +HP, -mana cost, +damage, etc. Stack up over a run.
E:/a327ex/Invoker/design.md:77:- **Mine cells deal damage** on entry (dungeon HP, not combat HP — the two pools are separate).
E:/a327ex/Invoker/design.md:82:The dungeon view UI shows a legend (color-graded numbers + room icons), the selected/hovered room's details (type, status, coords, links, adjacent mine count), the player's loadout, floor progress, and the top-bar stats (floor, time, HP, gold, flags).
E:/a327ex/Invoker/design.md:90:Each Combat room in the dungeon launches a fight. Combat is **separate** from the dungeon view — different render path, different HP pool.
E:/a327ex/Invoker/design.md:92:- Player has combat HP and Mana (orbs at corners, ARPG-style).
E:/a327ex/Invoker/design.md:95:- Win condition: clear the encounter (kill all / survive timer / objective). Lose condition: HP hits zero.
E:/a327ex/Invoker/design.md:132:- **Lore.** The Psyonia-universe framing (orbs as mana-cleansing devices, the sigil-mage class fantasy, the diegetic-book meta-narrative) is deferred to a future production. Practice version ships simple.
E:/a327ex/Invoker/design.md:147:- `Invoker/archive/` — superseded code (inventory/equipment/items from the orbs-and-gems iteration).
E:/a327ex/Invoker/design.md:148:- `reference/archive/` — superseded design docs (`design_orbs_v1.md` is the previous design.md).
E:/a327ex/Invoker/plan.md:27:- Hideout + tutorial corridor map shapes (geometry only — entity content is from the old orbs design and will be replaced)
E:/a327ex/Invoker/plan.md:34:- `inventory.lua`, `inventory_ui.lua`, `equipment.lua`, `equipment_ui.lua`, `items.lua` — old orb/gem/equipment system
E:/a327ex/Invoker/plan.md:37:- Spell system (primitives, modifiers, cast HUD, spell execution, mana)
E:/a327ex/Invoker/plan.md:38:- Combat (HP/mana orbs, enemies, damage)
E:/a327ex/Invoker/plan.md:51:**Goal.** Fresh game launch → spell-cast tutorial completes → player walks into a door/hole that triggers transition to the dungeon tutorial. Player has internalized: chord input, casting, canceling, mana, the relationship between elements + formations and resulting spells.
E:/a327ex/Invoker/plan.md:59:3. **Spell execution engine.** Chord (multiset of primitives) → spell lookup (table by canonicalized chord) → effect dispatch. Mana cost gating (insufficient mana = cast fails with feedback).
E:/a327ex/Invoker/plan.md:61:5. **Mana system.** Pool with max + current. Regen over time. Cost per cast (scales with chord complexity per `spells.md`). Display via mana orb.
E:/a327ex/Invoker/plan.md:62:6. **Combat HP / orbs.** Player has combat HP (separate from future dungeon HP). HP orb bottom-left, mana orb bottom-right (ARPG-style, screen-space).
E:/a327ex/Invoker/plan.md:82:3. **Top bar + sidebars.** Floor / time / dungeon-HP / gold / flags counter. Legend (left). Selected-room panel (right) with type + status + adj-mines + coords + links. Loadout strip (always-visible per spec).
E:/a327ex/Invoker/plan.md:91:   - **Mine** → -1 dungeon HP + visual feedback.
E:/a327ex/Invoker/plan.md:187:- **Mana regen curve** (Phase 1). Constant rate vs. burst-on-empty vs. condition-driven.
E:/a327ex/Invoker/plan.md:201:- **The big-design Psyonia integration** (book-as-game, Story 7 callback). Deferred. See `Psyonia/orblike.md` for those notes.

Edit (E:\a327ex\Invoker\plan.md)

The file E:\a327ex\Invoker\plan.md has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\plan.md)

The file E:\a327ex\Invoker\plan.md has been updated successfully. (file state is current in your context — no need to Read it back)

Edit (E:\a327ex\Invoker\design.md)

The file E:\a327ex\Invoker\design.md has been updated successfully. (file state is current in your context — no need to Read it back)

Docs updated. Ending the session now with the requested name.

Skill

Launching skill: end-session

Base directory for this skill: E:\a327ex.claude\skills\end-session

End Session Workflow

This skill is invoked from the same conversation where the work was done. You already have the full session context in memory — you do NOT need to read the transcript.

Step 1: Get Session Info

Ask the user for the session title (max 30 characters). Examples: "Anchor Phase 10 Part 5", "Physics Arena Setup", "Timer System Fix", "Thalien Lune Design".

Determine the project yourself from your session context — you know which repo(s) were worked on, which files were created/modified, and where they live. No need to ask. See Step 5 for the list of known project roots; if the session touched something outside the list, infer the root from the paths you actually edited.

Step 2: Write Summary

Write the summary from your conversation memory. You have the full session context — no need to read any files.

The summary should be thorough and detailed. Each major topic deserves its own section with multiple specific bullet points. Don't compress — expand.

Purpose: These summaries serve as searchable records. Future Claude instances will grep through past logs to find how specific topics were handled. The more detail you include, the more useful the summary becomes for finding relevant context later.

Format (this is just an example structure — adapt sections to match what actually happened):
# [Title]

## Summary

[1-2 sentence overview of the session's main focus]

**[Topic 1 - e.g., "Spring Module Implementation"]:**
- First specific detail about what was done
- Second detail - include file names, function names
- User correction or feedback (quote if notable)
- Technical decisions and why

**[Topic 2 - e.g., "Camera Research"]:**
- What was researched
- Key findings
- How it influenced implementation

**[Topic 3 - e.g., "Errors and Fixes"]:**
- Specific error message encountered
- Root cause identified
- How it was fixed

[Continue for each major topic...]

---

[Rest of transcript follows]
Rules: - Be thorough — If in doubt, include more detail, not less. Each topic should be as detailed as possible while still being a summary. - Think searchability — Future instances will search these logs. Include keywords, function names, error messages that someone might grep for. - One section per major topic — Don't combine unrelated work into one section - Chronological order — Sections should match conversation flow - Specific details — Error messages, file names, function names, parameter values - Include user quotes — When user gave notable feedback, quote it (e.g., "k/d variables are not intuitive at all") - Weight planning equally — Research, proposals, alternatives considered, user feedback on approach are as important as implementation - Weight problems solved — Errors, root causes, fixes, user corrections all matter - Technical specifics — Include formulas, API signatures, parameter changes when relevant

Step 3: Get User Approval

Show the title and summary to the user. Wait for approval before proceeding. If they have corrections, fix them.

Step 4: Convert Transcript and Write the Log File
# Find recent sessions (Claude + Cursor + Codex). Same script lives in Anchor2:
python E:/a327ex/Anchor2/scripts/find-recent-session.py --limit 5
# or: python E:/a327ex/Anchor/scripts/find-recent-session.py --limit 5
The script shows sessions sorted by when they ended. The first result is the current conversation (since end-session was invoked here). Use it.

Use a lowercase hyphenated slug derived from the title (e.g., "anchor-primitives-hitstop-animation").

Get the end timestamp for the Date frontmatter — this is the wall-clock time when end-session was invoked, NOT the time the JSONL started. Sessions often span multiple days, and the log should be filed under the day the work was wrapped up:
date "+%Y-%m-%d %H:%M:%S"
Use this output verbatim. Do not substitute the JSONL start timestamp; the log appears in the sidebar sorted by Date, and a multi-day session with a Date pinned to day 1 will sort below sessions that ended later but started later, hiding the most recent work.

Convert the transcript to markdown:
python E:/a327ex/Anchor2/scripts/jsonl-to-markdown.py [SESSION_PATH] /tmp/session-log.md
# or: python E:/a327ex/Anchor/scripts/jsonl-to-markdown.py ...
The same script auto-detects Claude Code JSONL vs Cursor/Composer agent JSONL (~/.cursor/projects/.../agent-transcripts/...) vs Codex rollouts (~/.codex/sessions/...). For Composer sessions, use find-recent-session.py (it merges all sources) and pick the [cursor] line for the current chat.

Replace the default header (# Session YYYY-MM-DD...) at the top of /tmp/session-log.md with the approved title and summary, AND prepend frontmatter. The final file shape:
Title: [Title]
Date: YYYY-MM-DD HH:MM:SS

# [Title]

## Summary

[approved summary text from step 2]

---

[transcript content from jsonl-to-markdown script]
Frontmatter is non-negotiable. Every log file MUST start with Title: and Date: lines. Without them, the site's sidebar shows the slug as the title and 0 (epoch) as the sort date. The backfill script in a327ex-site/deploy/backfill_metadata.py is a safety net, not a substitute — write it correctly the first time.

Then copy the final file to the log destination:
cp /tmp/session-log.md E:/a327ex/a327ex-site/logs/[slug].md
Step 5: Commit Project Repo

Identify the project repo(s) worked on this session from your own context — you already know which repos were touched and which files changed. For the common projects:

Project Root Stage command

Anchor E:/a327ex/Anchor git add docs/ framework/ engine/ scripts/ reference/

Anchor2 E:/a327ex/Anchor2 git add framework/ engine/ arena/ reference/ scripts/ docs/ .claude/

emoji-ball-battles E:/a327ex/emoji-ball-battles git add -A

invoker E:/a327ex/Invoker git add -A

thalien-lune E:/a327ex/thalien-lune git add -A

a327ex-site E:/a327ex/a327ex-site git add -A (this IS one of the web repos — the log file added in Step 4 is staged here automatically)

For a project not listed, infer the root from the files you actually created or modified this session and stage those. If multiple candidate roots look valid, ask the user which files to stage.

cd into the project root, stage, run git status to verify before committing.

IMPORTANT — FULL SUMMARY IN COMMIT: The commit message MUST include the FULL summary from the log file. Read the summary back from the log file to ensure nothing is missing.

IMPORTANT — COMMIT METHOD: The summary contains backticks, special characters, and markdown that WILL break heredocs and git commit -m. ALWAYS use the file-based method below. NEVER try a heredoc first — it will fail and produce a malformed commit that needs amending.
# Skip until we hit the line "## Summary", then take everything after the next
# blank line until the --- separator that precedes the transcript.
awk '/^## Summary$/{found=1; next} found && NR>1 && /^---$/{exit} found' \
    E:/a327ex/a327ex-site/logs/[slug].md > /tmp/commit_msg.txt

# Prepend the title (plain text, no #) and append attribution
sed -i "1i [Title]\n" /tmp/commit_msg.txt
printf "\nGenerated with [Claude Code](https://claude.com/claude-code)\n\nCo-Authored-By: Claude <noreply@anthropic.com>\n" >> /tmp/commit_msg.txt

git commit -F /tmp/commit_msg.txt
Step 6: Push the Repos

Two pushes — project (to GitHub) and a327ex-site (to the VPS):
# Project repo to GitHub. Skip this push if the project IS a327ex-site
# (handled by the second push below — don't duplicate).
git push origin main

# a327ex-site to the VPS (post-receive hook restarts the Lua server).
cd E:/a327ex/a327ex-site
git add -A
git commit -m "[Title]"
git push prod main 2>&1 | tail -3
Failure handling: if either push fails, the other still happens. Local commits stay intact, so the user can re-push manually once they've fixed whatever blocked it. Don't roll back; the committed state on disk is the source of truth.

Step 7: Confirm

Tell the user: - Project commit hash + GitHub push result - a327ex-site commit hash + VPS push result

Format:
Logged — "[Title]".
  project (Anchor2):  abc1234  pushed to GitHub
  a327ex-site:        ghi9012  pushed to a327ex.com (post-receive deployed)
If a push failed, surface that explicitly with the error.

Notes

Frontmatter is non-negotiable. Every log file MUST have Title: and Date: at the top. Enforced at write time in step 4. The backfill script is a safety net, not a substitute.

If the project IS a327ex-site (e.g., you spent the session writing the renderer or fixing the theme), there's no separate "project repo to GitHub" push — the a327ex-site push at the end IS the project push. Don't duplicate.

Legacy note: this skill used to also dual-write to anchor.blot.im/logs/ and anchor.blot.im/context/. Post-cutover (May 2026), anchor.blot.im is a JS redirect to a327ex.com and no longer receives content. Drop those steps if you find any older docs that still reference them.

ARGUMENTS: Invoker HP and Mana Orbs

Project	Root	Stage command
Anchor	`E:/a327ex/Anchor`	`git add docs/ framework/ engine/ scripts/ reference/`
Anchor2	`E:/a327ex/Anchor2`	`git add framework/ engine/ arena/ reference/ scripts/ docs/ .claude/`
emoji-ball-battles	`E:/a327ex/emoji-ball-battles`	`git add -A`
invoker	`E:/a327ex/Invoker`	`git add -A`
thalien-lune	`E:/a327ex/thalien-lune`	`git add -A`
a327ex-site	`E:/a327ex/a327ex-site`	`git add -A` (this IS one of the web repos — the log file added in Step 4 is staged here automatically)

Drafting the summary now. Title: Invoker HP and Mana Orbs.

# Invoker HP and Mana Orbs

## Summary

Built the HP and Mana orbs from scratch — went from no resource HUD at all to a fully shader-driven dual-orb display with 1D wave-sim surfaces, dithered cloud-noise interiors, scale + rotation hit feedback, and per-orb visual tuning. The orbs now sit in the bottom-left and bottom-right corners with `[`, `]`, `,`, `.` debug binds for damaging and refreshing them.

**Orientation + initial direction:**
- Read the previous session log (`orblike-to-invoker.md`) plus `design.md` and `plan.md` to understand Phase 1 status. Sub-phases 1-3 (primitives, cast HUD, spell catalog) were done; sub-phase 6 (HP / Mana orbs) was untouched.
- User picked "HP + mana orbs first" over the alternative options (Fireball-first, tutorial-first), with mouse-cursor aim queued for later.
- Surfaced the design conflict between `design.md`'s "ARPG corner orbs" spec and `mockup_ui_kit_1.html`'s top-bar stat-chip canon. User said "try both styles."

**First HUD experiment — corner orbs vs top-bar chips:**
- Built `ui_resource_orb` (PoE-style stencil-clipped liquid disc) and `ui_stat_chip` (compact ♥ 100/100 panel) as side-by-side comparison primitives in `ui.lua`.
- Added `hp/hp_max/mp/mp_max = 100/100` to `run_new()` in `run.lua`.
- Hit a `bad argument #5 to 'circle'` error on the stencil mask draw — the locked-palette colors are tables that need to be *called* (`color()`) to pack into rgba ints for the engine. Switched `color(255,255,255)` to `white()`.
- User chose: drop the orbs, move the chips to bottom-left stacked vertically (HP on top, MP below, 8 px margin, 2 px gap between).

**Pivot back to orbs (with shader):**
- User: "let's try the orbs again, orblike_hud has a visual example. Try to make it look like liquid with a shader instead of just a flat circle."
- Searched `E:/a327ex` for a "something something wave" project the user remembered for a 1D wave sim — no folder/code matched (only `next_wave` enemies in BYTEPATH, `Waver.lua`, shockwave shaders, and the `shake.lua` modules in every project). User said they'd find the archive.
- Wrote `assets/orb_liquid.frag`: dedicated `orb_layer = layer_new('orbs')` rendered between game and UI, full-pass post-process shader that procedurally renders up to 2 orbs from per-frame uniforms.
- First shader version had directional brightness gradient + specular highlight band + edge vignette. User: "the gradients you have for the liquid feel a bit incongruent... the game only has gradients for the walls but they're way more subtle and over longer distances. The various polychrome, etc effects are also noise that's more subtle than this."

**SNKRX flat-color rewrite:**
- Read `reference/snkrx_style_effects.md` design philosophy: flat colors, no gradients, layered springs, rectangles not sprites.
- Stripped the directional gradient + specular + edge vignette. Added subtle organic cloud noise lifted from `wall_gradient.frag`'s `organic_field` function (3-term sin/cos sum at different frequencies + phase offsets).
- Added a thin lighter strip directly on the waterline so the surface reads as a discrete pixel line.
- Border ring kept as 1.5-px constant-thickness ring in darkened base color.

**Edition noise mode catalog:**
- Wired `u_noise_mode` (cycled by `n` key — F11 conflicted with fullscreen on user's setup) through 6 modes from the wall_gradient edition palette:
  - 0: static RGB scaled by organic_field
  - 1: HSL lightness modulation
  - 2: HSL saturation modulation
  - 3: HSL L+S combined
  - 4: two-octave organic_field (mixes a fast small-scale field into the main one)
  - 5: hue-locked dynamic shimmer
- First version had two bugs: (a) `int` uniform via `layer_shader_set_int` wasn't binding cleanly — switched to float-cast-to-int matching `wall_gradient.frag`'s `u_edition` convention; (b) noise looked completely static across the orb because `organic_field` was sampled at `pix * 0.15` (works at wall scale, but only 0.43 cycles across a 56-px orb). Bumped to `pix * 1.2` for ~3 visible noise lumps.
- User selected mode 0 (static RGB) for HP and mode 5 (hue shimmer) for Mana.

**Animated drift:**
- User: "Can we have the noise move instead of being static for those?" — `organic_field`'s internal time divisors (`/143.634`, `/99.4324` etc.) drift extremely slowly at orb scale. Added explicit XY drift `vec2(u_time * 10.0, u_time * 6.0)` to the field input so the cloud pattern slides at ~8 px/sec horizontally, 5 px/sec vertically.

**1D wave simulation port (tidal_waver.lua:130-185):**
- User added `reference/tidal_waver.lua` to the project. Found the splash function at line 240: `arena:splash(f, x, width)` perturbs `width` adjacent springs at once (3-5 wide) and *sets* (not adds) their velocity to `f`. Comment: *"The width represents how many springs will be pulled at the same time, as that creates a more realistic water effect."*
- Wrote `waves.lua`: 16 spring-points per orb, packed into 4 vec4 uniforms for the shader (`u_orbN_surface_0..3`).
- Per-frame: spring physics step (`a = -k*x - d*v`), then `sub_iters` passes of neighbor velocity coupling (`spread * (self.x - neighbor.x)` injected into neighbor velocity).
- Endpoints clamp instead of wrap (waves bounce off "the orb wall").
- Shader gained `surface_offset_px` helper that unpacks the 4 vec4s and lerps between the two nearest points based on `(local_x + 1) * 0.5 * 15`. Replaced the placeholder sine ripple with `surface_disp_px / radius`.
- Added `waves_splash(w, t01, force, width)` matching tidal_waver's set-not-add semantics.

**Wave tuning iterations:**
- Round 1: f=8, bounce=0.65, spread=0.45, sub_iters=4, impulse 80. User: "the current bounce is too weak and not visible at all." (Predicted peak displacement ~0.8 px on 28-px-radius orb — math worked out: `peak ≈ v / (2π·f)`.)
- Round 2: f=3, bounce=0.85, spread=0.50, sub_iters=6, impulse 150. Big visible improvement.
- Round 3: User: "Higher initial impulse, lower bounce, slower spread." → impulse 350, bounce 0.60, spread 0.20.
- Round 4: User: "Should impulse up always, the wave should spread across the orb's surface, such that each impulse should always move the whole orb in some way." → f=2, bounce=0.85, spread=0.55, sub_iters=8, width=6.
- Round 5: User: "initial impulse is too big" → 400 → 200.
- Round 6: User: "0.75 bounce, 0.4 spread coupling".
- Round 7: User: "the impulse doesn't look like it's spreading properly... I think there was some additional way I made it spread." Found `arena:splash`'s width parameter — switched to `waves_splash(width=6)` with set-not-accumulate semantics. **Final values: f=2 Hz, bounce=0.75, spread=0.40, sub_iters=8, impulse 200, width 6.**

**Hit / refresh model with scale + rotation springs:**
- User wanted a `orb_hit(orb, location, damage_pct)` function: damage at a specific surface location, with wobble + rotation scaled to damage %. Refresh should be smaller impulse, no scale/rotation.
- Created `orb.lua` wrapping a `waves` state + a spring container with two springs (`main` for scale at rest 1.0, `rotation` for rotation at rest 0).
- `orb_hit(o, t01, damage, max_value)`: wave splash at `t01` with force scaled by `damage/max_value`, scale-spring impulse, rotation-spring impulse with random direction.
- `orb_refresh(o, t01, refresh, max_value)`: only a small upward wave impulse, no springs.
- Updated `run_reset` to call `orb_reset` on each.
- Final base values: wave impulse `-200 * pct` (width 6), scale displacement `-0.3 * pct` (max 30% squish at full damage, was 0.5 — reduced after user feedback "scale can be a bit less strong on high percentages"), rotation kick `±0.7 rad * pct` (~40° max, was 0.4 — bumped after user feedback "rotation should be a bit more obvious").

**Position to bottom corners:**
- Moved orbs from screen center `(width/2 ± 50, height/2)` to corners: HP at `(36, height-36)`, MP at `(width-36, height-36)` with radius 28 and 8 px margin.
- Removed the bottom-left HP/MP comparison chips (replaced by orbs).
- Verified layout: bottom-left strip x=0..80 (cast HUD's primitive HUD starts at 80) — HP orb spans 8..64 with 16 px margin to primitive HUD. Bottom-right strip x=316..480 — MP orb spans 416..472.

**Border color matching UI:**
- Switched orb border from hardcoded `base * 0.30` (darkened orb hue) to `u_border_color` uniform set from the locked-palette `panel_border` (56, 60, 65) so orbs match every other UI element's chrome.

**Color muting + per-orb tuning:**
- Added `c` key to cycle global color-mute variants (sat -25%, lit -12% per step, 4 variants) and `v` key to cycle mana hue-shimmer strength (1.0× → 0.5× → 0.25× → 0×).
- Top-of-screen labels showed active variant indices.
- User picked: HP halfway between full bright and one mute step (`mute_sat = 0.875, mute_lit = 0.94`), MP unchanged at full brightness, mana hue shimmer locked at `× 0.25`.
- Removed cycle keys + labels, hardcoded 0.25 strength for mode 5, switched mute uniforms from global to per-orb (`u_orbN_mute_sat`, `u_orbN_mute_lit`).

**Bayer dither:**
- User: "the orb effect visual itself looks a bit too off compared to how the rest of the game will look. Can we try something like a dither effect on the noise?"
- Added 4×4 Bayer matrix (`BAYER4`) and `bayer4(pix)` helper. Quantized field to 4 discrete levels per pixel: `field = clamp(floor(field*4 + bayer), 0, 3) / 3`. The eye averages it into the same gradient at distance, but each pixel snaps to one of 4 discrete colors up close.
- Created the SNKRX/Balatro pixel-art crosshatch look the user wanted.

**Damage values randomized for testing:**
- User: "Give me random values for the hit instead of just 10%, since I wanna see what higher/lower values look like too."
- `[` and `,` keys: `math.random(5, 100)` damage (visible effect range from subtle to lethal). `]` and `.` keys: `math.random(5, 40)` refresh.
- Visual effects use intended magnitude even if HP/MP can't actually drop that much (so a 90-damage hit on 30 HP still gives the full ~45° rotation kick).

**Text chrome:**
- Iterated through several positions for the "100" + "/ 100" readout. Color of "/ 100" changed from `text_muted` to `white` (palette token, 200,200,200) per user request.
- Final spacing: big (header_l) at `cy - 12`, small (body) at `cy + 2`. Bounding boxes overlap by 2 rows but visible glyphs come out tight due to font padding. Text block sits a few px below orb center.
- Wrapped chrome draws in `layer_push(layer, cx, cy, rotation, scale, scale)` so text wobbles + tilts with the orb on hits.

**Rotation/scale applied to shader interior:**
- User: "because it's an orb and it wasn't rotating the contents inside I wasn't seeing the effect. Change it back and apply the rotation/scale effect to the contents of the orb too."
- Reverted rotation spring (f=8, bounce=0.5 — fast snap-back).
- In `sample_orb`, after the rotation step, computed `noise_uv = center + local * radius` and passed to `apply_noise` instead of `pix`. Math: at rest (rotation=0, scale=1), `local * radius = d` so `noise_uv = pix` — identical to before. On a hit, `local` is rotated and the noise + dither rotate/scale with the disc.
- Mode 4's `pix * 2.8` in apply_noise hadn't been updated → shader compile error `undefined variable "pix"`. Fixed by switching to `noise_uv * 2.8`.

**Brazilian keyboard fix:**
- User pointed out `;` and `'` shift around between keyboard layouts. Switched mana keys to `,` and `.` (universal across all layouts including Brazilian ABNT).

**Files created/modified this session:**
- New: `waves.lua` (1D surface wave simulation), `orb.lua` (per-orb container + hit/refresh model), `assets/orb_liquid.frag` (procedural orb shader)
- Modified: `main.lua` (orb_layer, orb_liquid_shader, hp_orb/mp_orb globals, debug binds, render wiring), `ui.lua` (ui_orbs_apply, ui_orbs_chrome, removed ui_resource_orb + ui_stat_chip), `run.lua` (hp/mp fields + waves_reset → orb_reset), `design.md` and `plan.md` (orb spec updated, sub-phase 6 status reflected in "Shipped").

**Phase 1 status at session end:**
- Done: Sub-phases 1, 2, 6 (HP / Mana orbs).
- Partial: Sub-phase 3 (catalog done, dispatch placeholder).
- Partial: Sub-phase 5 (mana display orb done; pool/regen/cost not wired).
- Not started: 4, 7, 8, 9, 10.

Next session: real spell-effect dispatch (probably Fireball as discussed at the start), or Mana cost gating now that the visual orb exists, or the tutorial map content.

Does the title and summary look good? If yes, I'll convert the transcript and commit/push.

Yes.

Bash (python E:/a327ex/Anchor2/scripts/find-recent-session.py --limit 5)

2026-05-14T02:13:26.983Z cb6890c3-baef-4f6c-94cf-fa02c7da537f [claude] <-- MOST RECENT
   C:\Users\adn\.claude\projects\E--a327ex\cb6890c3-baef-4f6c-94cf-fa02c7da537f.jsonl

2026-05-14T02:13:01.007Z a60c5397-63c3-4a7c-809d-126b1f9adb2f [claude]
   C:\Users\adn\.claude\projects\E--a327ex\a60c5397-63c3-4a7c-809d-126b1f9adb2f.jsonl

2026-05-14T00:35:56.953Z 6d166688-839b-42c3-a9fe-7804ecedde9f [claude]
   C:\Users\adn\.claude\projects\E--a327ex\6d166688-839b-42c3-a9fe-7804ecedde9f.jsonl

2026-05-13T01:52:12.000Z b0fa48ef-6159-4da8-ba21-f6f2b1907f47 [claude]
   C:\Users\adn\.claude\projects\E--a327ex\b0fa48ef-6159-4da8-ba21-f6f2b1907f47.jsonl

2026-05-12T23:25:33.171Z 075f3871-f99f-40c7-80fd-4c700a62a237 [claude]
   C:\Users\adn\.claude\projects\E--a327ex\075f3871-f99f-40c7-80fd-4c700a62a237.jsonl

Bash (date "+%Y-%m-%d %H:%M:%S")

2026-05-13 23:14:09