home / skills / partme-ai / full-stack-skills / midjourney

midjourney skill

safe

This skill provides comprehensive guidance for Midjourney image generation, including prompts, parameters, and best practices to improve AI artwork.

npx playbooks add skill partme-ai/full-stack-skills --skill midjourney

Review the files below or copy the command above to add this skill to your agents.

Files (2)

SKILL.md

679 B

---
name: midjourney
description: Provides comprehensive guidance for Midjourney AI image generation including prompt engineering, image generation, parameters, and best practices. Use when the user asks about Midjourney, needs to generate AI images, create prompts, or work with Midjourney features.
license: Complete terms in LICENSE.txt
---

## When to use this skill

Use this skill whenever the user wants to:
- [待完善：根据具体工具添加使用场景]

## How to use this skill

[待完善：根据具体工具添加使用指南]

## Best Practices

[待完善：根据具体工具添加最佳实践]

## Keywords

[待完善：根据具体工具添加关键词]

Overview

This skill provides comprehensive, practical guidance for using Midjourney AI image generation. It covers prompt engineering, parameter tuning, image remixing, and workflow tips to produce consistent, high-quality results. Use it to iterate prompts, understand settings, and apply best practices for production or creative exploration.

How this skill works

The skill inspects the desired visual outcome and recommends prompt structures, seed and style settings, and Midjourney parameters (like --v, --q, --stylize, --ar). It explains how to combine descriptive language, reference images, and parameter flags to influence composition, color, and detail. The guidance includes step-by-step examples for building prompts and troubleshooting common artifacts.

When to use it

You want to create a detailed, reliable Midjourney prompt for a specific visual concept.
You need to optimize image quality, aspect ratio, or generation speed using Midjourney parameters.
You want to refine outputs by remixing, upscaling, or using image references.
You are troubleshooting unexpected artifacts, composition problems, or style inconsistencies.
You are preparing assets for concept art, marketing visuals, or social media content.

Best practices

Start with a clear subject and style, then add specific modifiers for lighting, mood, and color.
Use short, prioritized phrases separated by commas; place the most important descriptors first.
Introduce reference images to anchor composition, and use seeds to reproduce successful results.
Iterate quickly at lower quality (--q 0.25 or draft mode) then upscale once the composition is right.
Test one parameter change at a time to isolate effects from --stylize, --v, or aspect ratio adjustments.

Example use cases

Generate character concept art: prompt structure for silhouette, costume details, and cinematic lighting.
Create social media visuals: optimized aspect ratios and stylistic keywords for platform consistency.
Product mockups: use reference photos plus descriptive modifiers for materials and reflections.
Mood boards: produce variations of a theme with controlled color palettes and consistent seeds.
Marketing hero images: iterate composition and camera angles, then upscale and touch up in post.

FAQ

How do I make Midjourney produce consistent characters across images?

Use a reference image, a fixed seed, and a concise set of character descriptors; iterate with small changes and save successful seeds.

What parameters control stylization and detail?

Use --stylize for artistic freedom and --v (version) or model-specific flags for detail; lower stylize values yield more literal renders.

When should I use image prompting?

Use image prompts to anchor composition, color, or subject appearance; combine with textual modifiers to guide style and mood.

How can I reduce unwanted artifacts?

Lower complexity in the prompt, reduce stylize, try different seeds, and use upscaling only after composition is correct.