manus-skills/.kortix/opencode/skills/media/album-art-director/prompt-examples.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

11 KiB

AI Art Prompt Examples

Complete prompt examples for different album types.


Prompt Anatomy

Each example shows prompts for Midjourney and Leonardo.ai — the two most common platforms with the biggest prompt format differences. For DALL-E, use the Leonardo.ai version in a more conversational tone. For Stable Diffusion, use the Midjourney version with added token weights and an expanded negative prompt. See Platform-Specific Tips below for adaptation guidance.

Platform Prompt Style Negative Prompt Key Difference
Midjourney Comma-separated tags + parameters Via --no flag Keyword-driven, concise
Leonardo.ai Natural language sentences Separate field Descriptive, conversational
DALL-E Conversational sentences Not supported State what you want, not what to avoid
Stable Diffusion Weighted tags (keyword:1.2) Separate field, extensive Maximum control

Example 1: Tech Documentary (Distros)

Concept: Linux distributions as evolving ecosystem

Midjourney:

Abstract digital tree made of circuit board traces, roots formed from command line code,
branches creating tux penguin silhouette, vibrant gradient from green terminal glow to
orange Ubuntu to red hat, dark background fading to black, isometric view, technical
illustration style, clean and modern, high detail, 4k, album cover art --ar 1:1 --v 6

Leonardo.ai:

Prompt: An abstract digital tree whose trunk and branches are made of glowing circuit
board traces. The roots dissolve into lines of command-line code. The canopy forms the
silhouette of a penguin. The color palette shifts from green terminal glow at the base
through warm Ubuntu orange to deep red at the crown. The background is pure black,
fading from the glow. Viewed from a slightly elevated isometric angle. Clean, modern
technical illustration style with high detail.

Negative Prompt: photorealistic, blurry, text, watermark, human figures, messy wires,
low quality, pixelated

Model: Leonardo Phoenix | Preset: Dynamic | Aspect Ratio: 1:1

Why it works:

  • Specific subject (circuit tree, penguin)
  • Clear style (technical illustration)
  • Defined colors (green, orange, red)
  • Composition detail (isometric, dark background)

Example 2: True Crime (Corporate Scandal)

Concept: Corporate corruption, hidden darkness

Midjourney:

Corporate boardroom from dramatic low angle, empty leather chairs around polished table,
single overhead light creating harsh shadows, document with redacted text in foreground,
film noir style, high contrast black and white with single red accent, cinematic lighting,
photorealistic, moody and ominous, 4k resolution, album cover art --ar 1:1 --v 6

Leonardo.ai:

Prompt: A corporate boardroom photographed from a dramatic low angle. Empty black
leather chairs surround a polished mahogany table. A single harsh overhead light casts
deep shadows across the scene. In the foreground, a document with redacted black bars
lies on the table surface. The image is rendered in high-contrast black and white with
a single red accent — perhaps the redaction ink or a pen. Film noir atmosphere, moody
and ominous. Photorealistic cinematic quality.

Negative Prompt: people, faces, colorful, bright, cheerful, cartoon, illustration,
blurry, low resolution, watermark

Model: Leonardo Kino XL | Preset: Cinematic | Aspect Ratio: 1:1

Why it works:

  • Specific scene (boardroom, chairs, document)
  • Clear mood (noir, ominous)
  • Defined lighting (overhead, harsh shadows)
  • Color strategy (B&W with red accent)

Example 3: Electronic/Ambient

Concept: Digital consciousness, abstract data

Midjourney:

Flowing data streams forming abstract humanoid silhouette, particles of light coalescing
and dispersing, deep blue and cyan color palette with electric purple accents, dark
void background, motion blur effect, digital art style, ethereal and mysterious,
centered composition, 3d render, high resolution, album cover art --ar 1:1 --v 6

Leonardo.ai:

Prompt: Streams of glowing data particles flow upward, coalescing into the vague
silhouette of a human figure. The particles are mid-motion — some forming, some
dispersing into the void. Deep blue and cyan dominate the palette with occasional
electric purple accents where streams cross. The background is a dark void. A subtle
motion blur gives the scene a sense of constant movement. The figure is centered in
the frame. Digital art style with a 3D rendered quality, ethereal and mysterious.

Negative Prompt: realistic face, detailed anatomy, text, bright colors, warm tones,
solid shapes, cartoon, sharp edges, cluttered

Model: Leonardo Phoenix | Preset: Dynamic | Aspect Ratio: 1:1

Why it works:

  • Abstract but clear subject (data streams, silhouette)
  • Specific colors (blue, cyan, purple)
  • Movement implied (flowing, motion blur)
  • Style defined (digital art, 3d render)

Example 4: Folk/Acoustic

Concept: Journey, solitude, natural beauty

Midjourney:

Lone wooden acoustic guitar leaning against weathered fence post, open countryside
at golden hour, wheat field stretching to horizon, soft focus background, warm amber
and gold tones, shallow depth of field, fine art photography style, intimate and
nostalgic, natural film grain, 4k, album cover art --ar 1:1 --v 6

Leonardo.ai:

Prompt: A lone wooden acoustic guitar leans against a weathered fence post in open
countryside. It is golden hour — the low sun bathes the scene in warm amber and gold
light. Behind the guitar, a wheat field stretches to the horizon with a soft-focus
background. Shallow depth of field keeps the guitar sharp while the landscape blurs
gently. Fine art photography style with a subtle natural film grain. The mood is
intimate and nostalgic, like a quiet moment on an empty road.

Negative Prompt: people, modern objects, electric guitar, urban, bright neon, digital,
sharp background, cluttered, text, watermark

Model: Leonardo Kino XL | Preset: Photography | Aspect Ratio: 1:1

Why it works:

  • Tangible subject (guitar, fence, field)
  • Specific time/lighting (golden hour)
  • Clear mood (intimate, nostalgic)
  • Photography style (fine art, film grain)

Example 5: Historical Disaster (Dark Tide)

Concept: Industrial negligence looming over immigrant neighborhood

Midjourney:

Massive rusted industrial tank towering over cramped tenement buildings,
dark sepia and brown tones, 1919 Boston atmosphere, ominous calm before disaster,
vintage photograph aesthetic meets punk rock, gritty, high contrast,
dramatic low angle, album cover art --ar 1:1 --v 6

Leonardo.ai:

Prompt: A massive rusted industrial storage tank looms over a row of cramped tenement
buildings, shot from a dramatic low angle. The palette is dark sepia and brown,
evoking a 1919 Boston atmosphere. The scene has an ominous calm — the moment before
disaster. The aesthetic blends vintage photography with punk rock grit. High contrast,
grainy texture, oppressive composition with the tank dominating the upper frame.

Negative Prompt: modern buildings, cars, bright colors, cheerful, clean, people smiling,
digital look, smooth rendering, text

Model: Leonardo Kino XL | Preset: Cinematic | Aspect Ratio: 1:1

Example 6: Concept Album (Surrender)

Concept: Letting go, release

Midjourney:

Open hands releasing glowing particles into sky, soft sunset colors,
particles dissolving into light, peaceful but dynamic, close-up on hands,
background bokeh blur, cinematic photography, warm tones, gentle movement,
album cover art --ar 1:1 --v 6

Leonardo.ai:

Prompt: A close-up of open hands releasing glowing golden particles upward into a
sunset sky. The particles dissolve into soft light as they rise. The background is a
warm bokeh blur of sunset colors — peach, amber, soft violet. The mood is peaceful
but dynamic, capturing the exact moment of release. Cinematic photography style with
warm tones and a sense of gentle upward movement.

Negative Prompt: closed fists, dark mood, cold colors, horror, sharp background,
cluttered, extra fingers, deformed hands, text

Model: Leonardo Phoenix | Preset: Cinematic | Aspect Ratio: 1:1

Refinement Keywords

To simplify:

Add: "minimalist," "clean composition," "negative space," "single focal point"

To add drama:

Add: "dramatic lighting," "high contrast," "cinematic," "bold"

To increase realism:

Add: "photorealistic," "high detail," "8k," "octane render"

To increase style:

Add: "stylized," "artistic interpretation," "painterly," specific art movement

To adjust color:

Specify: exact hex codes, "muted color palette," "vibrant," "monochrome with accent"


Platform-Specific Tips

Midjourney Tips

  • Use --ar 1:1 for square album covers
  • --v 6 for latest model, --style raw for less stylized output
  • --no [element] for negative prompts (e.g., --no text watermark)
  • Keep prompts under ~60 words — Midjourney ignores excess
  • --s 250 increases stylization, --s 50 decreases it

Leonardo.ai Tips

  • Write sentences, not tags — Leonardo responds better to natural descriptions
  • Use the negative prompt field — don't embed exclusions in the main prompt
  • Model selection matters: Phoenix for versatility, Kino XL for cinematic, SDXL for control
  • Presets: Cinematic, Dynamic, Photography — pick one matching your concept
  • Guidance scale: 7-9 for balanced results, higher for more prompt adherence
  • Alchemy mode: Enable for higher quality at the cost of generation credits
  • Always include "album cover artwork" or "square format artwork" in the prompt

DALL-E Tips

  • Write as a clear instruction: "Create an image of..."
  • No negative prompts — focus only on what you want to see
  • Be explicit about style (e.g., "in the style of fine art photography")
  • Handles text in images better than most generators

Stable Diffusion Tips

  • Use weighted tokens: (important element:1.3) to emphasize
  • Extensive negative prompts improve quality significantly
  • Common negative: (worst quality:1.4), (low quality:1.4), blurry, watermark, text
  • CFG scale 7-9 balances creativity and prompt adherence
  • Sampler: DPM++ 2M Karras or Euler a for most use cases

Common Mistakes

1. Too Busy

Problem: Too many elements, cluttered composition Fix: Focus on one strong element, use negative space

2. Cliché

Problem: Generic genre imagery (mic for hip-hop, guitar for rock) Fix: Find unique angle on familiar symbols, use metaphor

3. Doesn't Scale

Problem: Looks great full-size, illegible at thumbnail Fix: Test at small size early, simplify if needed

4. Wrong Mood

Problem: Serious album has cheerful art (or vice versa) Fix: Align visual mood with musical mood via color, lighting, style

5. Too Literal

Problem: Album about cars shows car (boring) Fix: Symbolic representation, unusual angle, artistic interpretation

6. Text Overload

Problem: Album title, artist name, tracklist, year all on cover Fix: Minimal text, let image speak, text can be added digitally later