Skill

image

Generate prompts for AI image generation tools (Midjourney, Flux, Stable Diffusion, Firefly, DALL·E, Nano Banana, etc.). Use when the user wants a single still image. Triggers when user says "image prompt", "make me an image", "Midjourney prompt", "generate a picture", or describes a static visual scene without motion.

npx claudepluginhub polarislt0710/claude-skills-hugo --plugin ai-prompts

Tool Access

This skill uses the workspace's default tool permissions.

Preview

Build a paste-ready prompt for any modern image generator.

SKILL.md

Similar Skills

cache-components

139.3k

Guides Next.js Cache Components and Partial Prerendering (PPR): 'use cache' directives, cacheLife(), cacheTag(), revalidateTag() for caching, invalidation, static/dynamic optimization. Auto-activates on cacheComponents: true.

cache-components

mcp-builder

124.2k

Guides building MCP servers enabling LLMs to interact with external services via tools. Covers best practices, TypeScript/Node (MCP SDK), Python (FastMCP).

9 files

anthropics-skills-13

Stats

Parent Repo Stars0

Parent Repo Forks0

Last CommitMay 7, 2026

Actions

View Source View Plugin View on GitHub View README

Help us improve

Share bugs, ideas, or general feedback.

image

Image Prompt Builder

Build a paste-ready prompt for any modern image generator.

Core formula

[Subject + Action] + [Location / Context] + [Composition] + [Lighting] + [Style / Aesthetic] + [Camera / Lens] + [Color grading]

Principles

Lead with the subject. Make the first 6 words count — it carries the most weight in CLIP-based models.

Positive framing. Describe what IS there, not what isn't. ("a clean kitchen" beats "no clutter")

Lighting is mood. Specify it explicitly: golden hour, harsh noon, neon glow, single key light, ambient overcast.

Camera + lens controls geometry. "Shot on 35mm f/1.4" sets depth, framing, character.

Materiality matters. Specify texture for objects / clothing / environments — "weathered linen", "polished obsidian".

Quoted text + font. "Open" in bold sans-serif for in-image typography.

Style signature at the end. Film stock (Kodak Portra 400), art movement (mid-century modern), render style (Studio Ghibli, photorealistic, isometric).

Output format

[Single clean prompt block, ready to paste into UI]

Optional follow-up:

1 sentence on what's iterable (aspect ratio, lighting variant, style swap)

If the request has multiple plausible interpretations, offer 2 prompt variants

Worked example

User: "I want a moody photo of an old bookstore at night"

A solitary owner shelves a leather-bound volume in a narrow second-hand bookshop late at night, warm tungsten desk lamp casting long shadows across stacked spines, deep window reflection of empty wet street outside, low-angle medium shot, shot on 50mm f/1.4 with shallow depth of field, moody chiaroscuro, Kodak Portra 800 film stock, slight grain.

Image Prompt Builder

Build a paste-ready prompt for any modern image generator.

Core formula

[Subject + Action] + [Location / Context] + [Composition] + [Lighting] + [Style / Aesthetic] + [Camera / Lens] + [Color grading]

Principles

Lead with the subject. Make the first 6 words count — it carries the most weight in CLIP-based models.
Positive framing. Describe what IS there, not what isn't. ("a clean kitchen" beats "no clutter")
Lighting is mood. Specify it explicitly: golden hour, harsh noon, neon glow, single key light, ambient overcast.
Camera + lens controls geometry. "Shot on 35mm f/1.4" sets depth, framing, character.
Materiality matters. Specify texture for objects / clothing / environments — "weathered linen", "polished obsidian".
Quoted text + font. "Open" in bold sans-serif for in-image typography.
Style signature at the end. Film stock (Kodak Portra 400), art movement (mid-century modern), render style (Studio Ghibli, photorealistic, isometric).

Output format

[Single clean prompt block, ready to paste into UI]

Optional follow-up:

1 sentence on what's iterable (aspect ratio, lighting variant, style swap)
If the request has multiple plausible interpretations, offer 2 prompt variants

Worked example

User: "I want a moody photo of an old bookstore at night"

A solitary owner shelves a leather-bound volume in a narrow second-hand bookshop late at night,
warm tungsten desk lamp casting long shadows across stacked spines, deep window reflection of
empty wet street outside, low-angle medium shot, shot on 50mm f/1.4 with shallow depth of field,
moody chiaroscuro, Kodak Portra 800 film stock, slight grain.