manus-skills/.kortix/opencode/skills/media/album-art-director/visual-styles.md
zhukang 1c7abd81d1 All 310 Manus skill-hub community skills as a Kortix marketplace registry
- 310 registry:skill items, 1528 files, 7 categories (hub tags)
- registry.json generated by @kortix/registry buildRegistry
- nested SKILL.md in sn-da-* renamed SUBSKILL.md (no phantom skills)
- provenance + licenses: SOURCES.md
2026-09-15 13:37:20 +08:00

6.1 KiB

Visual Styles & References

Reference tables for album art visual direction.


Photography Styles

Style Effect Use For
Editorial photography Professional, magazine-quality Documentary, true crime
Portrait photography Subject-focused, intimate Character studies
Street photography Candid, urban, raw Hip-hop, gritty rock
Fashion photography Stylized, dramatic lighting Pop, R&B, high-concept
Documentary photography Authentic, journalistic True stories, historical
Fine art photography Artistic, intentional Concept albums, experimental

Illustration Styles

Style Effect Use For
Digital illustration Clean, modern, versatile Tech topics, contemporary
Vintage poster art Retro, nostalgic Historical, throwback
Comic book art Bold, dramatic, action Narrative, dynamic
Minimalist illustration Simple, iconic Abstract concepts
Surrealist painting Dreamlike, symbolic Experimental, avant-garde
Technical illustration Precise, detailed Tech, engineering topics

Artistic Movements

Movement Aesthetic Use For
Film noir High contrast, shadows, mystery Crime, thriller themes
Cyberpunk Neon, futuristic, gritty Tech, dystopian
Art deco Geometric, luxury, golden age Sophisticated, classic
Brutalism Raw, concrete, stark Industrial, harsh
Vaporwave Retro-futuristic, pastels, surreal Electronic, ironic
Minimalism Simple, essential, clean Abstract, meditative

Color Psychology

Emotional Color Palettes

Dark/Serious (True Crime, Heavy Topics):

  • Black, charcoal, deep navy
  • Desaturated colors
  • Minimal bright accents
  • High contrast

Energetic/Aggressive (Rock, Hip-Hop):

  • Red, orange, electric blue
  • High saturation
  • Bold combinations
  • Sharp contrasts

Calm/Introspective (Folk, Ambient):

  • Earth tones (brown, tan, moss green)
  • Pastels (soft pink, blue, cream)
  • Low saturation
  • Gentle gradients

Futuristic/Electronic (EDM, Synth):

  • Neon (cyan, magenta, electric purple)
  • Deep blacks for contrast
  • Glowing effects
  • Synthetic feel

Nostalgic/Vintage (Country, Classic):

  • Faded colors (sepia, washed denim)
  • Warm yellows and oranges
  • Muted reds and greens
  • Film grain texture

Genre Visual Language

Hip-Hop/Rap

  • Bold typography, street photography
  • Gold, black, red color schemes
  • Urban environments, luxury items
  • High contrast, dramatic lighting

Electronic/EDM

  • Geometric shapes, fractals
  • Neon colors, glowing elements
  • Abstract, futuristic
  • Motion blur, energy

Rock/Alternative

  • Raw, gritty textures
  • Band photography (live or candid)
  • Analog aesthetics (film grain, vintage)
  • Muted or saturated colors

Folk/Acoustic

  • Natural imagery, landscapes
  • Warm earth tones
  • Handmade aesthetics, organic
  • Intimate, close-up details

Country

  • Americana imagery (roads, fields, skylines)
  • Vintage trucks, boots, guitars
  • Sunset colors, dusty tones
  • Nostalgic, authentic

Platform Specifications

Size Requirements

Streaming platforms:

  • Minimum: 3000 x 3000 px
  • Recommended: 4000 x 4000 px
  • Format: PNG or JPEG, RGB color space
  • Aspect ratio: 1:1 (square)

SoundCloud:

  • 800 x 800 px minimum
  • Displays well at small sizes

distributor (Spotify, Apple Music, etc.):

  • 3000 x 3000 px minimum
  • 72 DPI minimum (300 DPI recommended)
  • RGB color mode
  • No blurriness or pixelation

Thumbnail Test

Critical: Art must work at 200x200px or smaller

Test:

  1. Shrink art to thumbnail size
  2. Can you still tell what it is?
  3. Does it stand out in a grid?
  4. Is text readable (if any)?

If no: Simplify. Reduce detail. Increase contrast.


AI Art Platforms

Note: AI art platforms evolve rapidly. Capabilities change frequently.

Platform Prompt Style Strengths Best For
Midjourney Tag-based, comma-separated Artistic composition, stylized results Stylized album art, abstract concepts
Leonardo.ai Natural language sentences Photorealism, cinematic quality, fine control Photorealistic covers, cinematic scenes
DALL-E Conversational sentences Ease of use, text rendering Beginners, text-heavy designs
Stable Diffusion Weighted tags + extensive negatives Maximum control, local/free, LoRA support Technical users, batch generation

Midjourney

  • Discord-based, subscription required
  • Prompt format: /imagine [prompt] --ar 1:1 --v 6
  • Strong artistic interpretation — sometimes adds creative elements beyond the prompt
  • --style raw for more literal, less stylized results

Leonardo.ai

  • Web-based, free tier available
  • Separate prompt and negative prompt fields — key differentiator from Midjourney
  • Multiple models: Phoenix (versatile), Kino XL (cinematic), SDXL (stable diffusion based)
  • Presets (Cinematic, Dynamic, Photography) significantly affect output
  • Alchemy mode for higher quality at the cost of generation credits
  • More predictable and controllable than Midjourney — natural language prompts iterate faster

DALL-E

  • Integrated in ChatGPT, API available
  • Conversational prompts — describe what you want in plain English
  • No negative prompt support — state what you want, not what to avoid
  • Better text rendering in images than most alternatives

Stable Diffusion

  • Open source, local or cloud deployment (ComfyUI / Automatic1111)
  • LoRA models for specific artistic styles
  • Most control via CFG scale, samplers, schedulers
  • Steeper learning curve but maximum flexibility

Choosing a Platform

  • Midjourney: Maximum artistic quality, less control. Good default for stylized art.
  • Leonardo.ai: Photorealistic/cinematic with precise include/exclude control. Natural language iteration.
  • DALL-E: Lowest barrier to entry. Best text rendering.
  • Stable Diffusion: Full control, local generation, model fine-tuning.