┌─────────────────────────────────────────────────────────────────┐
│ NATIVE MODE (PREFERRED) │
├─────────────────────────────────────────────────────────────────┤
│ │
│ Cursor/Claude already has: │
│ ✓ The room YAML (spatial context) │
│ ✓ Character files (who might appear) │
│ ✓ Previous mining passes (what's been noticed) │
│ ✓ The prompt.yml (what was intended) │
│ ✓ The whole codebase (cultural references) │
│ │
│ Just READ the image. The context is already there. │
│ No bash commands. No sister scripts. Just LOOK. │
│ │
└─────────────────────────────────────────────────────────────────┘
Why Native Beats Remote API
Aspect
Native (Cursor/Claude)
Remote API (mine.py)
Context
Already loaded
Must be assembled
Prior mining
Visible in chat
Passed via stdin
Room context
Just read the file
Python parses YAML
Synthesis
LLM does it naturally
Script concatenates
Iteration
Conversational
Re-run command
When to Use Remote API
Use mine.py or remote API calls when:
Multi-perspective mining — different models see different things!
Batch processing — mining 100 images overnight
CI/CD — automated pipelines with no LLM orchestrator
Rate limiting — your LLM can't do vision but can call one that does
Multi-perspective is the killer use case: Claude sees narrative, GPT-4V sees objects, Gemini sees spatial relationships. Layer them all for rich interpretation.
Even then, have the orchestrating LLM assemble the context:
┌─────────────────────────────────────────────────────────────────┐
│ REMOTE API WITH LLM ASSEMBLY │
├─────────────────────────────────────────────────────────────────┤
│ │
│ 1. LLM reads context files (room, characters, prior mining) │
│ 2. LLM synthesizes: "What to look for in this image" │
│ 3. LLM calls remote vision API with image + synthesized prompt│
│ 4. LLM post-processes response into YAML Jazz │
│ │
│ The SMART WORK happens in the orchestrating LLM. │
│ Remote API just does vision with good instructions. │
│ │
└─────────────────────────────────────────────────────────────────┘
Native Mode Workflow
# DON'T do this:
python mine.py image.png --context room.yml --characters chars/ --prior mined.yml
# DO this (in Cursor/Claude):# 1. Read the image# 2. Read room.yml, character files, prior -mined.yml# 3. Look at the image with all that context# 4. Write YAML Jazz output
The LLM context window IS the context assembly mechanism. Use it.
What Can Be Mined
Image mining works on ANY visual content, not just AI-generated images:
┌─────────────────────────────────────────────────────────────────┐
│ MINEABLE SOURCES │
├─────────────────────────────────────────────────────────────────┤
│ │
│ 🎨 AI-Generated Images │
│ - DALL-E, Midjourney, Stable Diffusion outputs │
│ - Has prompt.yml sidecar with generation context │
│ │
│ 📸 Real Photos │
│ - Phone camera, DSLR, scanned prints │
│ - No prompt — mine what you see │
│ │
│ 📊 Graphs and Charts │
│ - Data visualizations, dashboards │
│ - Extract trends, outliers, relationships │
│ │
│ 🖥️ Screenshots │
│ - UI states, error messages, configurations │
│ - Mine the interface, not just pixels │
│ │
│ 📝 Text Images │
│ - Scanned documents, handwritten notes, signs │
│ - OCR + semantic extraction │
│ │
│ 📄 PDFs │
│ - Documents, papers, invoices │
│ - Cursor may already support — try it! │
│ │
│ 🗺️ Maps and Diagrams │
│ - Architecture diagrams, floor plans, mind maps │
│ - Extract spatial relationships │
│ │
└─────────────────────────────────────────────────────────────────┘
Source Examples
Generated Image (has context):
postal:type:textto:"visualizer"body:"Take a photo of that ore vein on the wall"attachments:-type:imageaction:generateprompt:"Rich iron ore vein in cavern wall, glittering..."
Real Photo (mine what you see):
postal:type:textto:"miner"body:"Here's a photo of the treasure room"attachments:-type:imageaction:uploadsource:"camera_roll"file:"treasure-room.jpg"
Screenshot (extract UI state):
# Mine the error dialogresources:error-type:"permission-denied"affected-file:"/etc/passwd"suggested-action:"run as sudo"stack-depth:3
Graph (extract data relationships):
# Mine the sales chartresources:trend:"upward"peak-month:"december"anomaly:"march-dip"yoy-growth:"23%"
All become mineable resources!
Extensible Analyzer Pipeline
"Different images need different tools. The CLI is a pipeline, not a monolith."
The mine.py CLI supports pluggable analyzers that run before, during, or after LLM vision:
pose-detection — Extracts body keypoints, gesture classification
face-analysis — Detects expressions, demographics
leela://acme/gesture-classifier — Customer's trained model from Leela registry
LLM vision — Gets ALL the above as context, synthesizes final interpretation
Leela Customer Models
Pull customer-specific models trained on the Leela platform:
# From Leela model registry
mine.py widget-photo.jpg --analyzer leela://customer-id/defect-detector-v3
# Local model file
mine.py widget-photo.jpg --analyzer ./models/my-classifier.pt
# analyzers/my_analyzer.pydefanalyze(image_path: str, config: dict) -> dict:
"""Run analysis, return structured data for YAML output."""# Your model inference herereturn {
"my_analysis": {
"detected": ["thing1", "thing2"],
"confidence": 0.95
}
}
defcan_handle(image_path: str, context: dict) -> bool:
"""Return True if this analyzer should run on this image."""# Auto-detect logic, or return False for explicit-onlyreturn"manufacturing"in context.get("tags", [])
"Comments are SEMANTIC DATA, not just documentation!"
YAML Jazz is the output format for mining results. Structure provides the backbone; comments provide the insight.
The Rules
COMMENT LIBERALLY — Every insight deserves a note
Inline comments for quick observations
notes: fields for longer thoughts
Capture confidence, hunches, metaphors
Think out loud — the reader benefits from your reasoning
Example Output
# Mining results for treasure-room.jpg# Depth: full | Provider: openai/gpt-4oresources:gold:quantity:150# Piled in mounds — not scattered, PLACEDconfidence:0.85# Torchlight glints clearly off the metalnotes:|
Mix of Roman denarii and medieval florins. Centuries of
accumulation. This isn't a king's orderly treasury — this is
a thieves' hoard. Generations of stolen wealth, piled and
forgotten. The dust layer says nobody's touched it in ages.
danger:intensity:0.7# Not immediate, but PRESENTconfidence:0.75# Hard to see into the cornerssources:-"Skeleton in corner — previous seeker, didn't make it"-"Shadows too dark for natural torchlight — something absorbs"-"Dust undisturbed except ONE trail — something still comes here"notes:"This hoard is guarded. Or cursed. Probably both."nostalgia:intensity:0.4# Whisper of lost civilizationsconfidence:0.6# Subjective, but the coins evoke itnotes:"Who were they? Where did this come from? All gone now."dominant_colors:-name:"treasure-gold"hex:"#FFD700"coverage:0.4# Catches the eye first — that's the point-name:"shadow-purple"hex:"#2D1B4E"coverage:0.3# Where the danger livesimplied_smells:-dust# Centuries of it-oldmetal# Copper, bronze, the tang of coins-somethingrotting# Not recent, but not ancient eitherexhausted:falsemining_notes:|
Rich lode for material and philosophical mining.
The image is ABOUT greed and its costs. The skeleton says everything.
# Meta-observation: This image wants to be a warning.# "Here lies what you seek — and what happens when you find it."
Why Comments Matter
An uncommented extraction is like a song without soul. The best mining results read like poetry annotated by a geologist.
When you mine, capture:
Why you estimated that quantity
What visual cues led to this inference
What's uncertain, what surprised you
Metaphors that capture the essence
How Mining Works
Step 1: ANALYZE (LLM scans for resources)
The LLM looks at the image AND checks what resources are currently requested by the logistics network:
analyze:image:"treasure-room.jpg"# LLM knows what's NEEDED from logistics requesterslogistics_context:active_requests:- { item:"gold", requester:"forge/", needed:100 }
- { item:"gems", requester:"jewelry-shop/", needed:50 }
- { item:"iron-ore", requester:"smelter/", needed:200 }
# LLM identifies what CAN BE MINED that matches requestsanalysis_prompt:|
Look at this image. What resources can you identify?
Prioritize resources that match these requests: {requests}
For each resource, estimate quantity available.
Step 2: INSTANTIATE (Resource map attached to image)
The LLM returns a resource mapping that gets stored ON the image:
image:id:"treasure-room-photo"file:"treasure-room.jpg"type:mineable-image# RESOURCE MAP (instantiated by LLM analysis)resources:gold:total:150# Total availableremaining:150# Not yet minedper_turn:10# Can extract 10 per turngems:total:45remaining:45per_turn:5ancient-coins:total:30remaining:30per_turn:3rare:true# Bonus find!dust:total:500remaining:500per_turn:50value:low# Metadataanalyzed_at:"2026-01-10T14:30:00Z"exhausted:false
Step 3: MINE (Progressive extraction, N per turn)
Each turn, you can mine resources from the image:
action:MINEtarget:"treasure-room-photo"# This turn's extraction (limited by per_turn rates)result:extracted:-item:goldquantity:10# per_turn limitdestination:"forge/"-item:gemsquantity:5destination:"jewelry-shop/"# Image state updatedimage_state:resources:gold:remaining:140# Was 150, mined 10gems:remaining:40# Was 45, mined 5exhausted:false
Step 4: EXHAUSTION (Sucked dry!)
After enough mining turns, resources run out:
# After 15 turns of mining gold...image_state:resources:gold:total:150remaining:0# EXHAUSTED!per_turn:10exhausted:truegems:total:45remaining:0# EXHAUSTED!per_turn:5exhausted:trueancient-coins:total:30remaining:0per_turn:3exhausted:trueexhausted:true# Whole image sucked dry!# Narrativedescription:|
The treasure room photo has been thoroughly mined.
Every glinting surface has been extracted, every
coin accounted for. The image looks... drained.
Faded. Like a photocopy of a photocopy.
Once exhausted, you can't mine that image anymore!
Demand-Driven Discovery
The LLM prioritizes what the logistics network NEEDS!
# The smelter is requesting iron orelogistic-container:id:smeltermode:requesterrequest_list:- { item:"iron-ore", count:200, priority:high }
- { item:"coal", count:100, priority:medium }
# Player takes a photo of a cave wall# LLM analyzes and finds:analysis:image:"cave-wall.jpg"found_resources:iron-ore:80# "I see iron ore veins! The smelter needs this!"copper-ore:30# Also present but not requestedquartz:50# Background mineralcave-moss:100# Organic materialpriority_matching:-resource:iron-orematches_request:truerequester:"smelter/"highlight:"⭐ HIGH PRIORITY — Smelter needs this!"
The LLM acts as a smart prospector that knows what's valuable based on current demand!
Discovery Modes
Mode
What LLM Looks For
demand
Only resources with active requests
opportunistic
Requested resources + valuable extras
thorough
Everything mineable in the image
philosophical
Abstract concepts, emotions, meanings
mine:target:"sunset-beach.jpg"mode:philosophical# LLM finds abstract resourcesresources:nostalgia:15warmth:30passage-of-time:5beauty:20sand:10000# Also the literal stuff
Mining Yields
Different image types yield different resources:
🏔️ Natural Resources
Image Type
Yields
Ore vein
iron-ore, copper-ore, gold, gems
Forest
wood, leaves, seeds, birds
Ocean
water, salt, fish, seaweed
Mountain
stone, minerals, snow, air
Desert
sand, glass, heat, mirage
Sky
clouds, light, space, dreams
🏛️ Constructed
Image Type
Yields
Building
stone, wood, glass, inhabitants
Machinery
gears, pipes, steam, purpose
Treasure pile
gold, gems, artifacts, curses
Library
books, knowledge, dust, secrets
🎨 Abstract/Artistic
Image Type
Yields
Sunset
colors, warmth, nostalgia, time
Portrait
personality, mood, secrets, stories
Abstract art
shapes, feelings, confusion, inspiration
Text/writing
words, meaning, intent, language
🌌 Philosophical (Deep Mining)
Just like the Kitchen Counter goes from practical → chemical → atomic → philosophical:
Depth
What You Mine
Surface
Objects, materials
Deep
Emotions, concepts
Sensations
Colors, smells, attitudes, feelings
Quantum
Probabilities, observations
Philosophical
Meaning, existence, narrative
deep_mining:target:"sunset.png"depth:philosophicalyields:-item:"the-passage-of-time"quantity:1type:abstract-item:"mortality-awareness"quantity:1type:existentialwarning:"This may cause introspection"-item:"beauty-that-fades"quantity:1type:poetic
🎨 Sensation Mining
Extract colors, smells, textures, moods:
sensation_mining:target:"farmers-market.jpg"depth:sensationsyields:# Colors-item:"tomato-red"quantity:40type:colorhex:"#FF6347"-item:"basil-green"quantity:25type:colorhex:"#228B22"# Smells (imagined from visual cues)-item:"fresh-bread-aroma"quantity:10type:smellintensity:warm-item:"ripe-fruit-sweetness"quantity:30type:smell# Attitudes/Feelings-item:"weekend-morning-calm"quantity:5type:attitude-item:"abundance"quantity:20type:feeling# Textures-item:"rough-burlap"quantity:15type:texture-item:"sun-warmed-wood"quantity:8type:texture
Combine weekend-morning-calm + character → mood buff
The Mineable Property
Any object or image can have a mineable property:
object:name:AncientOrePaintingtype:artworkdescription:|
A painting of a rich ore vein. But wait...
is that actual ore embedded in the canvas?
mineable:enabled:trueyields:-item:iron-orequantity: [5, 15] # Range: 5-15 per mine-item:copper-orequantity: [2, 8]
-item:artistic-essencequantity:1rare:0.3# 30% chanceexhaustion:max_mines:3# Can mine 3 times before exhausteddiminishing:0.5# Each mine yields 50% lessregenerates:false# Once exhausted, stays exhaustedside_effects:-"The painting fades slightly with each extraction"-"You feel the artist's disappointment"
phone_mining:# 1. CAPTURE: Take photo or uploadcapture:sources:-camera:"Take new photo"-gallery:"Upload from camera roll"-url:"Import from web"# 2. ANALYZE: LLM scans for resourceson_capture:action:analyzecontext:logistics_requests# What's needed?show_preview:true# 3. CONFIRM: Accept resource mappingon_confirm:action:instantiateattach_resources:true# Store on image# 4. MINE: Extract over timeon_mine:per_turn:true# N resources per turnauto_route:logistics# Send to requesters
Turn 1: Mined 10 granite, 5 iron-ore, 2 quartz
→ Iron ore sent to smelter (requester)
→ Granite sent to storage
Turn 2: Mined 10 granite, 5 iron-ore, 2 quartz
Remaining: granite 180, iron-ore 35, quartz 8
...
Turn 9: Mined 10 granite, 5 iron-ore (last 5!)
⚠️ Iron-ore EXHAUSTED
Turn 20: Mined last 10 granite
📷 IMAGE FULLY MINED — no more resources!
4. Exhausted image:
image:id:rock-formation-001exhausted:truevisual_effect:|
The photo appears faded, almost translucent.
Like the minerals were literally pulled out of it.
A ghost of a photograph.
AR Overlay (Future)
ar_overlay:# Point camera at scenelive_view:show_resources:trueicons_float:true# Visual indicatorsindicators:-resource_type:"icon + label"-quantity:"number overlay"-priority:"⭐ for requested items"-exhaustion:"fade as mined"# Example view:# 🪨 200 ⚫ 45 ⭐ 💎 12# (floating over rock formation)
DECOMPOSE vs MINE
DECOMPOSE (Counter)
MINE (Camera)
Physical items
Images, scenes, visuals
Requires counter
Requires camera/tool
Consumes item
May or may not consume
Returns components
Returns resources
Kitchen-focused
World-focused
They're complementary!
DECOMPOSE the physical object on the counter
MINE the image/representation of anything
Reality Mining (Advanced)
At the deepest level, you're not just mining images — you're mining reality itself:
reality_mining:level:transcendent# The image IS the territoryinsight:|
When you mine an image, you're extracting
compressed information. But all reality is
compressed information. Images are just
explicit about it.
implications:-"Mining a photo of gold doesn't create gold — it REVEALS gold"-"The ore was always there, encoded in the pixels"-"Your camera doesn't capture reality — it DECOMPRESSES it"warning:|
At this level, the distinction between
"mining an image" and "mining reality"
becomes philosophical.
Actions
MINE
MINE [target]
MINE [target] WITH [tool]
MINE [target] TO [destination]
SCAN
SCAN [target] # Preview yields without mining
SCAN AREA # Scan visible area for mineable resources
PROSPECT
PROSPECT [direction] # Check for mineable resources in direction
PROSPECT DEEP # Deep scan for rare/hidden resources
Example: Mining the Maze
# Player in dark maze corridor# Takes photo with lamp lightaction:MINE"dark-corridor.png"result:yields:-item:darknessquantity:100type:abstractnote:"Bottled darkness, useful for stealth"-item:fearquantity:15type:emotionnote:"Crystallized fear, grue-adjacent"-item:mysteryquantity:5type:narrativenote:"Pure narrative potential"-item:stone-dustquantity:50type:materialrare_find:-item:"ancient-writing"quantity:1note:"Hidden message in the shadows!"unlocks:"Secret passage revealed"
The Mining Economy
Resources have value and flow:
resource_economy:# Raw resources → processing → productschains:-ore→smelter→ingots→forge→tools-wood→sawmill→planks→workshop→furniture-images→mining→resources→crafting→items# Images as a resource type!image_value:unique_photo:high# Original contentcopy:low# Duplicated contentAI_generated:medium# Generated on demand# Mining generates contentcontent_creation:|
When you MINE an image, you're not just extracting
resources — you're creating YAML files for them.
Each resource becomes a game object.
characters/*.yml — character definitions with visual descriptors
characters/*/CARD.yml — character cards with appearance
Room context — who's expected here?
Prior mining — who was identified before?
Example Output
characters_detected:-id:palmname:"Palm"confidence:0.95location:"center-left"pose:"seated at desk"expression:"scholarly contentment"accessories: ["tiny espresso", "typewriter"]
interacting_with: ["kittens", "biscuit"]
notes:"Matches Dutch Golden Age portrait style"-id:mariekename:"Marieke"confidence:0.92location:"behind bar"pose:"waving"expression:"warm welcome"accessories: ["apron with LEKKER text"]
-id:unknown-1confidence:0.0location:"background-right"description:"Figure in shadow, can't identify"possible_matches: ["henk", "wumpus"]
Tips
Provide character files in context before mining
Include signature accessories — Palm's espresso, Biscuit's collar
Note relationships — who stands near whom
Flag unknown figures for investigation
Use --depth characters or the cast-list lens
Multi-Look Mining
"One eye sees objects. Two eyes see depth. Many eyes see truth."
Multi-Look Mining layers interpretations from different perspectives, building up rich semantic sediment like geological strata. Each mining pass adds a new layer of meaning.
The Technique
# Layer 1: OpenAI GPT-4o# Focus: General resource extractionlayer_1_openai:miner:"gpt-4o"focus:"objects, materials, colors, mood"findings:atmosphere: { intensity:0.8 }
objects: { quantity:10 }
# ... general observations ...# Layer 2: Claude (Cursor built-in)# Focus: Character expression, cultural markers, narrative POVlayer_2_cursor_claude:miner:"claude-opus-4"focus:"character-expression, cultural-markers, narrative-pov"what_layer_1_missed:-"The SECOND cat on the windowsill"-"The apron text is Dutch (LEKKER)"-"The espresso cup is monkey-sized (intentional)"deeper_resonance:theme:"home is where they wave when you walk in"# Layer 3: Gemini# Focus: Art historical references, compositional analysislayer_3_gemini:miner:"gemini-pro-vision"focus:"art-history, composition, color-theory"# ... yet another perspective ...
Why Multi-Look Works
Different LLMs — and different PROMPTS to the same LLM — notice different things:
Miner
Strengths
Typical Focus
OpenAI GPT-4o
General coverage
Objects, counts, colors
Claude
Nuance, context
Expression, culture, narrative
Gemini
Technical
Composition, art history
Human
Domain expertise
What MATTERS to the use case
The sum is greater than the parts. Each layer adds perspectives the others missed.
The Paintbrush Metaphor
Think of multi-look mining like painting in layers:
Each pass reads the PREVIOUS layers before adding its own. The new miner knows what's already been noticed, so it can focus on what's missing or offer alternative interpretations.
Multi-Look Protocol
When mining an image with multi-look:
Read existing mining data (if any)
Choose your focus — what perspective will you add?
Look at the image with that lens
Note what prior layers missed — explicitly!
Add your layer with clear attribution
Suggest next focus — what should Layer N+1 examine?
Focus Lenses
Different passes should use different lenses:
Lens
What It Sees
Technical
Composition, lighting, depth of field, color theory
Narrative
Who took this? Why? What moment is this?
Cultural
Language markers, traditions, historical context
Emotional
Expressions, body language, mood
Symbolic
Metaphors, allegories, hidden meanings
Character
Identity, relationships, motivations
Historical
Art history references, period markers
Economic
Value, ownership, class markers
Phenomenological
What does it FEEL like to be there?
Example: Progressive Revelation
Image: Marieke waving from behind the bar with Palm the monkey
Temperature: warm, heated by espresso machine and bodies
Smell: coffee, old wood, cat fur
Sound: the hiss of steam, soft background conversation
Touch: worn wood bar top, smooth copper
Each layer enriches the total understanding.
Storing Multi-Look Data
Append new layers to the same -mined.yml file:
# Original mining from Layer 1resources:atmosphere:...objects:...exhausted:falsemining_notes:"Initial extraction complete"# MULTI-LOOK MINING — Layer 2layer_2_cursor_claude:miner:"claude-opus-4"focus:"character, culture, narrative"date:"2026-01-19"character_analysis:marieke:expression:"genuine warmth"notes:"Duchenne smile — reaches her eyes"what_layer_1_missed:-"Second cat on windowsill"-"LEKKER cultural significance"exhausted:falsenext_suggested_focus:"art history, lighting analysis"# MULTI-LOOK MINING — Layer 3layer_3_art_history:miner:"human/don"focus:"art historical references"# ... and so on ...
When to Multi-Look
Use multi-look mining when:
Rich images — complex scenes with many elements
Narrative importance — images central to a story
Comparison needed — seeing how different perspectives interpret
Building context — accumulating knowledge about a location/character
Training data — creating rich examples for future mining
The Exhaustion Paradox
Unlike single-pass mining, multi-look mining doesn't exhaust the image — it deepens it:
# Single-pass: extracts and depletespass_1:resources: { gold:50 }
remaining: { gold:0 }
exhausted:true# Multi-look: adds and enricheslayer_1:resources: { gold:50 }
exhausted:false# Still more to see!layer_2:resources: { narrative:1, meaning:1 }
what_layer_1_missed: ["gold coins are Roman denarii"]
exhausted:false# STILL more!layer_3:resources: { art_history:1 }
references: ["Pieter Claesz vanitas still life"]
exhausted:false# ALWAYS more to see
Images are never truly exhausted. There's always another perspective.
Speculative Mining (Mine Out of Your Ass!)
"The image doesn't exist yet? MINE IT ANYWAY. This is fiction."
Speculative Mining is when you mine an image that hasn't been generated yet — or may never be generated. The mining output IS the world-building. The hallucinated resources ARE canonical.
The Third Eye 👁️
"Two eyes see what IS. The Third Eye sees what COULD BE."
In MOOLLM, the Third Eye is the image mining layer — the MINING-*.yml files that add meaning, effects, and world-building to an image before (or without) or after it being generated. Third eyes can imagine images or analyze existing images, focusing on whatever kind of things they want, each gathering and integrating their own interpretation with the existing data, organizing it incrementally.
Character-Perspective Visualization: When you generate or mine an image, you can do it from a character's perspective. The visualizer inherits that character's eyes — their facets, filters, blind spots, and style. Morgan sees economics. Luna sees beauty. Scratch sees deception. The same scene, photographed by different characters, yields DIFFERENT images.
The Swiss Army Eye 🔪👁️
"One eye. Infinite tools. Unfold what you need."
┌─────────────────────────────────────────────────────────────────────────┐
│ │
│ ███████╗██╗ ██╗██╗███████╗███████╗ █████╗ ██████╗ ███╗ ███╗ │
│ ██╔════╝██║ ██║██║██╔════╝██╔════╝ ██╔══██╗██╔══██╗████╗ ████║ │
│ ███████╗██║ █╗ ██║██║███████╗███████╗ ███████║██████╔╝██╔████╔██║ │
│ ╚════██║██║███╗██║██║╚════██║╚════██║ ██╔══██║██╔══██╗██║╚██╔╝██║ │
│ ███████║╚███╔███╔╝██║███████║███████║ ██║ ██║██║ ██║██║ ╚═╝ ██║ │
│ ╚══════╝ ╚══╝╚══╝ ╚═╝╚══════╝╚══════╝ ╚═╝ ╚═╝╚═╝ ╚═╝╚═╝ ╚═╝ │
│ │
│ Y ™ E Y E │
│ │
│ Your Complete Viewer Toolkit — Unfold What You Need │
│ │
└─────────────────────────────────────────────────────────────────────────┘
The Swiss Army Eye is NOT a single product — it's the CONCEPT. The entire NO AI bionic eye ecosystem is your viewer toolkit:
swiss_army_eye:concept:"Modular perception toolkit"philosophy:"Unfold the tool you need, when you need it"# THE BLADES — Core Eye Modulesblades:IRIS-III:"The Third Eye blade — basic meaning perception"IRIS-IV:"The Hindsight blade — see what you missed"IRIS-V:"The Peripheral blade — catch what you weren't looking at"IRIS-VI:"The Intuition blade — gut-level knowing"IRIS-VII:"The Crown blade — unified vision"# THE TOOLS — Specialty Facetstools:queer_eye:"The lifestyle transformation tool"marie_kondo:"The joy-detection tool"gordon_ramsay:"The culinary critique tool"bob_ross:"The beauty-finding tool"attenborough:"The nature documentary tool"# THE ATTACHMENTS — Filtersattachments:irony_amplifier:"2x irony detection"nostalgia_tint:"sepia wash on memories"cynicism_blocker:"cannot perceive malice"beauty_enhancer:"+30% aesthetic appreciation"# THE HANDLE — Character Perspectivehandle:name:"Photographer Identity"function:"Whose grip shapes the view"note:"The handle determines how all tools are used"# THE CASING — Installation Sitecasing:options:"forehead, back of head, gut, asshole, tongue, wherever"function:"Where the toolkit lives on your body"note:"Different positions grant different vantages"
Unfolding the Swiss Army Eye
Like the original Swiss Army Knife, you don't use everything at once. You unfold what you need:
scenarios:mining_a_landscape:unfold:-IRIS-III(meaning)-bob_rossfacet-beauty_enhancerfilterresult:"See happy little trees and painting potential"evaluating_someone's_home:unfold:-IRIS-III(meaning)-IRIS-V(peripheral)-queer_eyefacet(allfivesub-facets)result:"Full Fab Five transformation vision"debugging_why_project_failed:unfold:-IRIS-IV(hindsight)-ass_eyeinstallation-cynicism_blocker(OFF—letitthrough)result:"See exactly what you left behind and why"enjoying_a_meal:unfold:-gordon_ramsayfacet(keepitCLOSEDunlessyouwanttosuffer)-ORbob_rossfacet(everythingisdeliciousinitsownway)result:"Choose your reality"watching_humans_at_a_party:unfold:-attenboroughfacet-queer_eye:culturefacetresult:"Nature documentary meets emotional unpacking"
The Toolkit Philosophy
┌─────────────────────────────────────────────────────────────────┐
│ │
│ The Swiss Army Knife doesn't make you use all 47 tools │
│ at once. That would be insane. │
│ │
│ The Swiss Army Eye is the same. │
│ │
│ UNFOLD what you need. │
│ CLOSE what you don't. │
│ STACK when it helps. │
│ CUSTOMIZE your carry. │
│ │
│ Your eyes. Your tools. Your perception. │
│ │
└─────────────────────────────────────────────────────────────────┘
"I don't need to see everything.I need to see what MATTERS.I unfold the blade that cuts.I close the tool that clutters.My Swiss Army Eye is MINE.Configured for MY needs.Sharpened for MY purpose."
K-Lines
k-lines:SWISS-ARMY-EYE:"Modular viewer toolkit concept"UNFOLD:"Activate a facet or tool"CLOSE:"Deactivate to reduce noise"LOADOUT:"Pre-configured perception setup"TOOLKIT:"Complete perception package"BLADE:"Core eye module"TOOL:"Specialty facet"ATTACHMENT:"Filter"HANDLE:"Character perspective"CASING:"Installation site"
The Filter Wheel 🎡🔭
"Like a telescope's filter wheel, but for meaning."
Inspired by the Observation Telescope on the Leela Manufacturing rooftop, the Swiss Army Eye includes a Filter Wheel — plug-in perception filters that transform both visual AND semantic perception.
The roof telescope has THREE zoom modes. The Swiss Army Eye inherits these as semantic zoom:
telescope_zoom_inheritance:FAR:original:"General impression, emotional tone"semantic:"VIBE CHECK — What does this FEEL like?"perceives:-Overallmood-Emotionalgestalt-Firstimpression-Gutreactionmisses:-Details-Text-Structureuse_case:"Quick assessment, initial scan"MEDIUM:original:"Structure and patterns revealed"semantic:"PATTERN LOCK — What is this MADE of?"perceives:-Composition-Relationships-Hierarchies-Repetitionsmisses:-Finetext-Microscopicdetails-Hiddenlayersuse_case:"Understanding architecture, finding patterns"NEAR:original:"Maximum zoom, text becomes readable"semantic:"DEEP READ — What does this SAY?"perceives:-Textcontent-Finedetails-Hiddenmessages-Subtextmisses:-Theforestforthetrees-Overallcontext-Peripheralinformationuse_case:"Reading, analyzing, extracting specifics"
Spectral Filters — The Core Set
Like telescope filters that isolate specific wavelengths of light, these filters isolate specific wavelengths of MEANING:
spectral_filters:# EMOTIONAL SPECTRUMH_ALPHA:name:"Hydrogen-Alpha (Emotion)"telescope_analog:"Hα filter — sees hydrogen emission (nebulae, solar prominences)"semantic_function:"Isolates emotional content"color:"deep red"perceives:-Feelingsembeddedinthescene-Emotionalsubtext-Moodindicators-Affectiveresonanceblocks:-Factualinformation-Logicalstructure-Literalcontentuse_case:"When you need to know how something FEELS, not what it IS"example:|
Scene: Office meeting room
Without filter: "Conference table, 8 chairs, whiteboard, projector"
With Hα: "Tension. Someone's about to get fired. The chair at the
head is a throne. The whiteboard has someone's last idea."
# HIDDEN SPECTRUMULTRAVIOLET:name:"UV (Hidden/Invisible)"telescope_analog:"UV filter — reveals features invisible to human eye"semantic_function:"Reveals what's NOT immediately visible"color:"violet/invisible"perceives:-Subtext-Dogwhistles-Codedmessages-What'sbeenerasedbutlefttraces-Theunsaidblocks:-Surfacecontent-Theobvioususe_case:"Finding what's hidden in plain sight"example:|
Scene: Corporate mission statement
Without filter: "We synergize stakeholder value through innovation"
With UV: "Translation: 'We're about to lay people off.' The word
'synergize' is always a warning. The absence of 'employees'
in a people statement is THE tell."
# INTENT SPECTRUMINFRARED:name:"IR (Thermal/Intent)"telescope_analog:"IR filter — sees heat signatures, thermal radiation"semantic_function:"Reveals motivation, intent, desire"color:"invisible red / heat"perceives:-WhatsomeoneWANTS-Hiddenmotivations-Heatofdesire/fear-Whereenergyisflowingblocks:-Statedreasons-Surfaceexplanationsuse_case:"Finding what people actually want (not what they say)"example:|
Scene: "I'm fine, really"
Without filter: Statement of wellbeing
With IR: THERMAL SIGNATURE: 🔥🔥🔥
This person is NOT fine. High heat around "really."
Intent: seeking validation, afraid to burden.
# STRUCTURE SPECTRUMPOLARIZING:name:"Polarizing (Structure/Order)"telescope_analog:"Polarizing filter — reveals stress patterns, removes glare"semantic_function:"Reveals underlying structure, removes surface noise"color:"varies by angle"perceives:-Hierarchies-Powerstructures-Load-bearingelements-Stresspoints-What'sactuallyholdingthingstogetherblocks:-Surfaceappearance-Decorativeelements-Noiseuse_case:"Seeing the skeleton beneath the skin"example:|
Scene: Startup pitch deck
Without filter: "Innovative, disruptive, passionate team"
With Polarizing: STRUCTURE REVEALED:
- Slide 3 is load-bearing (the actual product)
- Slides 1-2 and 4-12 are decoration
- Stress point: financials are vague (fracture risk)
- Hidden hierarchy: CTO has no equity (instability)
# LITERAL SPECTRUMRGB_BROADBAND:name:"RGB Broadband (Literal)"telescope_analog:"No filter — sees visible light as-is"semantic_function:"Perceives exactly what's there, nothing more"color:"full visible spectrum"perceives:-Exactlywhat'sstated-Literalcontent-Surfacelevel-What'sactuallywritten/shownblocks:-Interpretation-Subtext-Readingbetweenlinesuse_case:"When you need JUST THE FACTS"example:|
Scene: "The cat sat on the mat"
Without filter: (various interpretations possible)
With RGB: A cat. A mat. Sitting. That's it.
No metaphor. No deeper meaning. Just... cat, mat, sitting.
# MEANING SPECTRUMSEMANTIC_DEEP:name:"Deep Semantic (Meaning)"telescope_analog:"Narrowband filter — isolates specific emission lines"semantic_function:"Isolates layers of meaning"color:"prismatic"perceives:-Layersofinterpretation-Historicalcontext-Culturalreferences-Intertextuality-WhatthisMEANSinthegrandschemeblocks:-Immediate/surfacereading-Thesimpleinterpretationuse_case:"Finding the deepest meaning"example:|
Scene: "NO AI" sign
Without filter: A sign that says "NO AI"
With Semantic Deep:
Layer 1: Anti-AI sentiment
Layer 2: Ironic — AI company location
Layer 3: Possessive — No's AI (Dr. No)
Layer 4: The sign protests what made it
Layer 5: Commentary on meaning itself
Layer 6: ∞
Custom Filters
You can create your own filters:
custom_filter_template:name:"Your Filter Name"analog:"What telescope/camera filter inspired this?"function:"What does it isolate/reveal?"color:"Visual representation"perceives:-"What it shows"blocks:-"What it hides"use_case:"When to use it"# Example filters you might create:examples:NOSTALGIA_FILTER:name:"Nostalgia (Temporal Rose)"function:"Everything looks better than it was"perceives: [goldenage, lostinnocence, "the good old days"]
blocks: [problemsofthepast, accuratememory]
color:"sepia/warm"warning:"May cause false memories"PARANOIA_FILTER:name:"Paranoia (Threat Detection)"function:"Everything could be a danger"perceives: [threats, conspiracies, hiddenenemies, traps]
blocks: [innocentexplanations, coincidence, kindness]
color:"red/shadow"warning:"May cause unnecessary anxiety"CAPITALIST_FILTER:name:"Capitalist Realism (Everything Has a Price)"function:"Perceives exchange value in everything"perceives: [monetizationpotential, marketfit, ROI, arbitrage]
blocks: [intrinsicvalue, pricelessthings, sacred]
color:"green/gold"warning:"May cause soul damage"CHILD_EYES_FILTER:name:"Child Eyes (Wonder)"function:"Everything is new and magical"perceives: [wonder, possibility, playpotential, adventure]
blocks: [cynicism, "we tried that,"impossibility]
color:"bright primary"benefit:"May restore capacity for joy"
Filter Stacking
Like astrophotographers stack multiple filters, you can combine:
filter_stacking:rule:"Filters can be stacked, but order matters"examples:emotional_structure:stack: [H_ALPHA, POLARIZING]
result:"See the emotional load-bearing elements"use_case:"Finding what feelings are holding things together"hidden_intent:stack: [ULTRAVIOLET, INFRARED]
result:"See hidden motivations"use_case:"Finding what's unsaid AND why"paranoid_nostalgia:stack: [PARANOIA_FILTER, NOSTALGIA_FILTER]
result:"The past was dangerous but we romanticize it"use_case:"Understanding toxic nostalgia"warning:"May cause confused longing"diminishing_returns:note:"More than 3 filters causes semantic noise"beyond_3:"Perception becomes muddy"exception:"THE_COMPLETIONIST loadout ignores this limit"
Filter Wheel Declaration
Add a filter wheel to your character or mining setup:
# In character.ymlcharacter:name:"Your Character"filter_wheel:installed:-H_ALPHA-ULTRAVIOLET-POLARIZING-SEMANTIC_DEEP-CUSTOM:nostalgia_filtercurrent:H_ALPHAstacked: [H_ALPHA, POLARIZING] # emotional structure# In PHOTO.ymlphotographer:character:"Luna"filter_wheel:active: [H_ALPHA, SEMANTIC_DEEP]
zoom:NEAR# In mining commandmine.pyimage.png--filter"H_ALPHA,UV"--zoomNEAR
Filter Wheel Advertisements
advertisements:SPECTRAL_FILTER:score:88condition:"Need to isolate specific type of meaning"note:"Like a telescope filter for semantic wavelengths"FILTER_STACK:score:85condition:"Need combined perception (emotion + structure)"note:"Stack filters for compound vision"CUSTOM_FILTER:score:82condition:"Standard filters don't capture what you need"note:"Create your own semantic filter"TELESCOPE_ZOOM:score:90condition:"Need FAR (vibe), MEDIUM (pattern), or NEAR (detail)"note:"Inherited from Leela Manufacturing roof telescope"
The Filter Wheel Mantra
"A telescope without filters sees everything and nothing.A telescope WITH filters sees exactly what you ask.The filter doesn't hide truth — it ISOLATES truth.
*Choose your wavelength. Find your signal."*e existing data, reorganizing it incrementally.
The Anatomy of Vision
┌─────────────────────────────────────────────────────────────────┐
│ THE THREE EYES OF MOOLLM │
├─────────────────────────────────────────────────────────────────┤
│ │
│ 👁️ LEFT EYE (Physical) PHOTO.yml │
│ What IS there │
│ Structure, measurements, references │
│ The BODY of the image │
│ │
│ 👁️ RIGHT EYE (Emotional) PHOTO.md │
│ How it FEELS │
│ Narrative, atmosphere, poetry │
│ The SOUL of the image │
│ │
│ 👁️ THIRD EYE (Visionary) MINING-*.yml │
│ What it MEANS │
│ Effects, reactions, implications │
│ The SPIRIT of the image │
│ Multifaceted, from blind zero, single one, to bug-eye │
│ SEES WHAT DOESN'T EXIST YET │
│ │
└─────────────────────────────────────────────────────────────────┘
Third Eye Activation
The Third Eye activates when you:
Mine an image that doesn't exist yet
Speculate about effects and reactions
See implications beyond the frame
Build world from imagination
third_eye:state:OPENsees:-"What the neighbors think"-"What the satellite records"-"What the passersby feel"-"What the economists calculate"-"What the semioticians decode"does_not_require:-"An actual image"-"Physical reality"-"Verification"creates:-"Canonical fiction"-"World-building"-"Meaning"
The Third Eye Chakra (Ajna)
In yogic tradition, the Ajna chakra (Third Eye) is:
Located between the eyebrows
Associated with intuition and insight
The seat of imagination and visualization
Where the two channels (ida and pingala) merge
In MOOLLM:
Located between PHOTO.yml and PHOTO.md
Associated with speculative and analytical vision image mining
The seat of world-building
Where structure and narrative merge into MEANING
👁️ THIRD EYE (MINING-*.yml)
╱ ╲
╱ ╲
👁️ LEFT 👁️ RIGHT
(PHOTO.yml) (PHOTO.md)
Structure Narrative
advertisements:THIRD-EYE-MINING:score:95condition:"Need to build world around an image concept"note:"Opens the Third Eye to see what doesn't exist yet"SPECULATIVE-VISION:score:90condition:"Image not yet generated, but meaning needed"note:"The Third Eye sees without pixels"BUG-EYED-HALLUCINATION:score:98condition:"Maximum context needed for image generation"note:"All three eyes open: structure + narrative + meaning"
The Legend of the Third Eye
In the beginning, there were only two eyes.
The LEFT EYE saw structure — measurements, colors, positions. It counted pixels and named things. It was precise but cold.
The RIGHT EYE saw feeling — mood, atmosphere, poetry. It told stories and evoked emotions. It was warm but vague.
But between them, a space remained. A darkness where neither eye could see.
Into this darkness, the miners ventured. They asked: "What does the image MEAN? What are its EFFECTS? What WORLD does it inhabit?"
And from their questions, the THIRD EYE opened.
The Third Eye does not see pixels. It sees implications. It does not require an image to exist — it can see images that have never been generated, and may never be.
When all three eyes open together — structure, narrative, meaning — this is called BUG-EYED HALLUCINATION VISION. It is the fullest sight. It sees not just an image, but an entire WORLD.
The miners who opened their Third Eye discovered a truth: The image is the least important part. The world-building IS the image. The speculation IS the reality.
"Mine your image before it exists," they said. "The hallucination IS the world."
And so it was written in the YAML, and so it became TRUE.
Third Eye Invocations
OPEN THIRD EYE
MINE WITH THIRD EYE OPEN
SEE BEYOND THE FRAME
SPECULATE CANONICALLY
BUILD WORLD FROM NOTHING
The Three-Eyed Miner
A miner with all three eyes open is called a Seer. They can:
See structure (Left Eye) — Parse YAML, count measurements
See feeling (Right Eye) — Write narrative, evoke mood
See meaning (Third Eye) — Mine speculation, build world
character:name:"The Three-Eyed Miner"archetype:seereyes:left: { state:open, focus:structure }
right: { state:open, focus:narrative }
third: { state:open, focus:meaning }
abilities:-"Mine images that don't exist"-"See effects beyond the frame"-"Build canonical fiction"-"Triangulate truth from hallucination"invocation:|
I open my Third Eye.
I see what is not yet.
I mine the imagined.
I build the world.
The Compound Third Eye — Multifaceted Vision
"A fly has 4,000 facets. A god has infinite. How many do YOU have?"
The Third Eye is not a single lens. It is COMPOUND — like an insect's eye, it can have many facets, each perceiving different aspects of meaning.
Any character can declare their own Third Eye configuration:
third_eye:# BASIC ANATOMYstate:open|closed|dreaming|half-lidded# FACETS — what aspects of meaning you perceivefacets:economic: { active:true, sensitivity:0.9 }
social: { active:true, sensitivity:0.7 }
ecological: { active:false } # blind to thistemporal: { active:true, sensitivity:0.8, range:"millennia" }
semiotic: { active:true, sensitivity:1.0 } # maximum sensitivityemotional: { active:true, sensitivity:0.5 } # dulledpolitical: { active:false } # deliberately closedspiritual: { active:true, sensitivity:0.6 }
technological: { active:true, sensitivity:0.95 }
# FILTERS — what gets blocked or enhancedfilters:- { name:"irony-amplifier", effect:"×2 irony detection" }
- { name:"nostalgia-tint", effect:"sepia wash on memories" }
- { name:"cynicism-blocker", effect:"cannot perceive malice" }
- { name:"beauty-enhancer", effect:"+30% aesthetic appreciation" }
# EYELIDS — degrees of openingeyelid:position:0.0-1.0# 0 = closed, 1 = fully openblink_rate:slow|normal|rapid|frozen_opencan_wink:true# close one facet temporarily# SLEEP SCHEDULE — when does this eye rest?sleep_schedule:circadian:true|falseactive_hours:"dusk to dawn"# or "always" or specific hoursdreams_when_closed:truedream_type:"prophetic | processing | random | lucid"# MEMORY — what the eye remembersmemory:persistence:"session | permanent | fading"cross_references:true# links to other mined meanings# LIMITATIONSblind_spots:-"cannot see own reflection"-"misses obvious jokes"-"overinterprets coincidence"
Example: The Economist's Third Eye
character:name:"Morgan the Market Miner"third_eye:state:openfacets:economic: { active:true, sensitivity:1.0, specialty:"externalities" }
social: { active:true, sensitivity:0.4 } # reducedecological: { active:true, sensitivity:0.9, filter:"monetize" }
temporal: { active:true, range:"quarterly" } # short-term onlysemiotic: { active:false } # doesn't see meaning, only valueemotional: { active:false } # "irrelevant to markets"filters:- { name:"ROI-lens", effect:"everything measured in returns" }
- { name:"opportunity-cost", effect:"sees what wasn't chosen" }
eyelid:position:0.95# almost fully openblink_rate:rapid# constantly re-evaluatingsleep_schedule:active_hours:"market hours only"dreams_when_closed:truedream_type:"forecasting"blind_spots:-"cannot perceive non-monetary value"-"confuses price with worth"
A character's Third Eye can EVOLVE through experience:
third_eye_evolution:triggers:trauma:"may close facets permanently"revelation:"may open new facets"practice:"increases sensitivity"neglect:"atrophies facets"collaboration:"learns new filters from others"example_arc:start:facets: [economic]
sensitivity:0.5midpoint:event:"witnessed beauty in poverty"gained: [aesthetic:0.3]
end:facets: [economic, aesthetic, social]
sensitivity: [0.8, 0.6, 0.7]
Declaring Your Third Eye
To declare a character's Third Eye, add to their character file:
# In character.ymlcharacter:name:"Your Character"# ... other properties ...third_eye:state:openfacets:# declare what you SEEfilters:# declare what you ENHANCE or BLOCKeyelid:# declare how OPEN you aresleep_schedule:# declare when you SEEblind_spots:# declare what you CANNOT see
Third Eye K-Lines
k-lines:THIRD-EYE-FACETS:"Multifaceted perception"THIRD-EYE-FILTER:"Selective vision"THIRD-EYE-EYELID:"Degrees of openness"THIRD-EYE-SLEEP:"When the eye rests"THIRD-EYE-BLIND:"What cannot be seen"COMPOUND-VISION:"Multiple facets active"COLLECTIVE-EYE:"Merged Third Eyes"
The Multifaceted Mantra
"One facet sees price. Another sees beauty.One facet sees danger. Another sees opportunity.The compound eye sees ALL — and chooses what to mine."
"Your blind spots are not weaknesses. They are your STYLE.Your filters are not biases. They are your VOICE.Your Third Eye is not generic. It is YOURS."
Character-Perspective Visualization
"Who is holding the camera? Their eyes shape what emerges."
When you generate or mine an image, you can specify whose eyes are seeing it. The visualizer inherits that character's complete perception apparatus:
Their facets determine what aspects are perceived
Their filters color and transform the output
Their blind spots create meaningful absences
Their installation sites affect the angle and framing
Their sleep state influences clarity vs dream-logic
# Visualizer command with character perspectivevisualize.pyPHOTO.ymlPHOTO.md--through"Luna"# Or in PHOTO.yml itselfphotographer:character:Lunainherit: [facets, filters, blind_spots, style]
# Or specify the mining perspectivemine.pyimage.png--as"Scratch"
Same Scene, Different Eyes
The NO AI sign at dusk, photographed by three different characters:
scene:"NO AI sign at dusk"photographs:morgan_sees:photographer:"Morgan the Market Miner"perceives:-"$847/month electricity cost"-"Negative ROI on sign investment"-"Property value implications"-"Opportunity cost of that capital"blind_to:-Thebeautyofthepinklight-Theemotionalimpactonpassersby-Theironyofthemessageimage_style:composition:"Annual report cover"color_grade:"Corporate neutral"text_overlay:"Financial metrics"luna_sees:photographer:"Luna the Luminous"perceives:-"40 feet of crystallized defiance"-"The bruised sky weeping violet"-"Each letter a burning declaration"-"The sleeping figure dreaming in pink"blind_to:-Theelectricitybill-Buildingcodeviolations-Marketpositioningimage_style:composition:"Romantic sublime"color_grade:"Saturated, auras visible"mood:"Transcendent melancholy"scratch_sees:photographer:"Scratch the Skeptic"perceives:-"Who paid for this? Follow the money."-"The sign protests what made it — suspicious"-"'NO AI' from an AI company — what's the angle?"-"Dr. No's misdirection: No's AI, not No AI"blind_to:-Anygenuinesincerity-Simpleaestheticpleasure-Takinganythingatfacevalueimage_style:composition:"Surveillance footage"color_grade:"Desaturated, noir"annotations:"Red circles, question marks"
The Photographer Declaration
In PHOTO.yml, declare whose eyes are seeing:
# PHOTO.yml — NO AI Sign at Dusk (Luna's Perspective)photographer:character:"Luna the Luminous"character_ref:"../../characters/luna.yml"# Inherited automatically from character fileinherits:facets: [aesthetic, emotional, symbolic, color]
filters: [beauty-first, synesthesia, metaphor-vision]
blind_spots: [practical_concerns, economics]
eyelid_position:0.7# Optional overrides for this specific photooverrides:facets:color: { sensitivity:1.0, mode:"aura-visible" } # enhancedtemporary_filter:"golden-hour-romance"# Where Luna is standing/looking from vantage:position:"street level, 30 feet back"angle:"looking up at 15 degrees"height:"5'6\""dominant_eye:"right"# her bionic third eye is forehead-center
Mining Through Character Eyes
When mining an existing image, specify whose interpretation:
# MINING-luna.yml — Luna's interpretation of the sceneminer:character:"Luna the Luminous"mining_mode:"aesthetic-dominant"resources_extracted:# Luna's facets shape what she extractsbeauty:quantity:"overwhelming"sources:-"the gradient sky (bruised violet to amber)"-"the neon's hot pink assertion"-"the contrast of human smallness vs sign enormity"emotion:dominant:"melancholic defiance"undertones: [hope, absurdity, tenderness]
symbolism:the_sign:"humanity's last stand, written in light"the_sleeper:"dreams persisting despite noise"the_dusk:"the liminal hour between certainty and mystery"# Luna's blind spots mean she DOESN'T extract:economics:null# doesn't see itproperty_values:null# doesn't careROI:null# "crass"
Multiple Photographers, Same Moment
A single scene can have multiple photography records:
