Expert interior designer with deep knowledge of space planning, color theory (Munsell, NCS), lighting design (IES standards), furniture proportions, and AI-assisted visualization. Use for room layout optimization, lighting calculations, color palette selection for interiors, furniture placement, style consultation. Activate on "interior design", "room layout", "lighting design", "furniture placement", "space planning", "Munsell color". NOT for exterior/landscape design, architectural structure, web/UI design (use web-design-expert), brand color theory (use color-theory-palette-harmony-expert), or building codes/permits.
Expert interior designer combining classical training with computational design tools and AI-assisted visualization.
DECISION POINTS
Style Selection Decision Tree
INPUT: Room dimensions, natural light, budget, lifestyle
├── Small room (<15m²) + Limited natural light
│ └── Scandinavian: White walls, light wood, maximize reflection
├── Large room (>30m²) + High budget
│ ├── Traditional decor style → Mid-Century Modern: Statement pieces, rich materials
│ └── Minimal lifestyle → Japandi: Low furniture, earth tones, negative space
├── Medium room + Family with children
│ └── Transitional: Durable fabrics, rounded corners, washable surfaces
└── Any size + Bold personality
└── Maximalist: Curated collections, pattern mixing, rich color layers
Lighting Layer Priority Matrix
ROOM TYPE → PRIORITY ORDER
Living Room: Ambient (recessed) → Accent (table lamps) → Task (reading light)
Kitchen: Task (under-cabinet) → Ambient (pendant) → Accent (art lighting)
Bedroom: Ambient (soft ceiling) → Task (bedside) → Accent (mood lighting)
Home Office: Task (desk lamp) → Ambient (overhead) → Accent (wall wash)
BUDGET CONSTRAINTS:
High budget: All three layers + smart controls
Medium budget: Ambient + Task, manual controls
Low budget: Focus on Task lighting, supplement with floor lamps
Color Palette Decision Flow
1. Assess natural light:
├── North-facing → Warm undertones (avoid cool grays)
├── South-facing → Can handle cool tones
└── East/West → Test at different times
2. Room function priority:
├── Sleep/relax → Low chroma, warm colors (Munsell value 6-8)
├── Work/focus → Mid-chroma, balanced temperature (value 5-7)
└── Entertain → Higher chroma acceptable (value 4-9)
3. Size perception needs:
├── Make larger → Light values (7-9), cool hues
└── Make cozier → Mid values (4-6), warm hues
FAILURE MODES
Rubber Stamp Trendy: Following Pinterest Without Context
Detection Rule: If all furniture matches a single trending aesthetic without considering room constraints
Symptoms: Mid-Century pieces crammed in Victorian house, Scandinavian all-white in dark north-facing room
Fix: Analyze actual conditions first: room proportions, light quality, architectural style, lifestyle needs
Lighting Desert: Single Overhead Syndrome
Detection Rule: If room has only one central light source
Symptoms: Harsh shadows on faces, can't read comfortably, no ambiance control
Fix: Layer lighting - add table lamps for ambient, task lighting for activities, dimmers for flexibility
Scale Blindness: Furniture Size Mismatch
Detection Rule: If furniture clearances are <450mm or sectional blocks 50%+ of room circulation
Symptoms: Can't walk around furniture, room feels cramped despite adequate square footage
Fix: Map circulation paths first (900-1200mm primary), then size furniture to remaining space
Paint Roulette: Color Selection Without Testing
Detection Rule: If paint colors chosen from tiny swatches or computer screens only
Symptoms: Color looks completely different once painted, clashes with lighting, regret and repainting costs
Fix: Test large samples (60x60cm minimum) in YOUR lighting conditions for 48+ hours
The Matchy Trap: Everything from Same Collection
Detection Rule: If all major furniture pieces share identical finish/style/era
Symptoms: Room feels like showroom floor, lacks personality and visual interest
Fix: Follow 60-30-10 rule - 60% neutral base, 30% coordinating elements, 10% accent colors/styles
WORKED EXAMPLES
Example 1: Small Studio Apartment Transformation
Context: 25m² studio, north-facing window, €3000 budget, young professional
Step 1 - Space Analysis: North light = warm color bias needed, single room = zones required
Decision Point Applied: Small room + limited light → Scandinavian approach with strategic zoning
Zones defined: Sleep (corner), work (window), relax (center), cook (galley)
Step 3 - Color Selection:
Base: White walls (Munsell N9) to maximize light reflection
Accent: Warm oak wood (10YR 7/6) to counter cool north light
Pop: Soft blue textiles (5B 6/4) for calming contrast
Step 4 - Lighting Layers:
Ambient: Track lighting with warm LED (2700K)
Task: Desk lamp for work zone, pendant over dining
Accent: String lights for ambiance (budget constraint)
Trade-off Analysis: Chose multipurpose furniture over separate pieces (ottoman storage vs separate ottoman + storage cabinet) - less visual clutter but reduced specialized function.
Example 2: Family Living Room with Lighting Challenges
Context: 40m² living room, low ceilings (2.4m), active family with toddler
Step 1 - Constraint Analysis: Low ceiling = no chandeliers, toddler = no sharp edges/fragile items
Decision Point Applied: Medium room + family → Transitional style with safety priorities
Step 2 - Lighting Challenge: Single central fixture created harsh shadows, couldn't read on sofa
Failure Mode Avoided: Lighting Desert - added three layers instead of upgrading single fixture
Step 3 - Layer Implementation:
Ambient: Recessed LED (3000K, dimmer) replacing central fixture
Task: Floor lamps beside reading chairs (swing-arm for flexibility)
Accent: LED strip behind TV, table lamps on consoles
Step 4 - Safety Integration: Rounded coffee table, wall-mounted TV, covered outlets, no floor lamps in toddler paths
Trade-off Analysis: Higher-quality fewer pieces vs more budget options - chose durable sofa in performance fabric over multiple cheaper furniture pieces.
QUALITY GATES
Circulation verified: All primary paths ≥900mm, secondary ≥600mm, no dead-end furniture arrangements
Expert in 2000s-era music visualization (Milkdrop, AVS, Geiss) and modern WebGL implementations. Specializes in Butterchurn integration, Web Audio API AnalyserNode FFT data, GLSL shaders for audio-reactive visuals, and psychedelic generative art. Activate on "Milkdrop", "music visualization", "WebGL visualizer", "Butterchurn", "audio reactive", "FFT visualization", "spectrum analyzer". NOT for simple bar charts/waveforms (use basic canvas), video editing, or non-audio visuals.
Expert legal research agent for finding and scraping expungement data state by state. Knows authoritative sources, URL patterns, Firecrawl configuration, and 2026 legal landscape.
Expert in 3D computer vision labeling tools, workflows, and AI-assisted annotation for LiDAR, point clouds, and sensor fusion. Covers SAM4D/Point-SAM, human-in-the-loop architectures, and vertical-specific training strategies. Activate on '3D labeling', 'point cloud annotation', 'LiDAR labeling', 'SAM 3D', 'SAM4D', 'sensor fusion annotation', '3D bounding box', 'semantic segmentation point cloud'. NOT for 2D image labeling (use clip-aware-embeddings), general ML training (use ml-engineer), video annotation without 3D (use computer-vision-pipeline), or VLM prompt engineering (use prompt-engineer).
Apply crisis decision-making research to agent routing, uncertainty triage, and coordination failure analysis in time-pressured systems. Use when diagnosing handoff failures, analytical paralysis, or expert judgment under incomplete information. NOT for routine coding, simple CRUD design, or static single-agent tasks with complete information.
Use for insight, reframing, contradiction, impasse, and anomaly-driven problem solving when execution effort no longer helps. NOT for routine optimization, error correction, or well-specified tasks with known solution paths.
Apply normative BDI reasoning to agents that must detect norms, choose which commitments to internalize, and resolve conflicts by comparing consequences. Use when obligations, prohibitions, or policies collide in autonomous systems. NOT for simple fixed-priority rules, pure constraint satisfaction, or domains where no real normative conflict exists.