Token导航 LogoToken导航TokenDH.com
AI 工具需要联网github未标认证来源可访问clear审计通过

midjourney-card-news-backgrounds中途卡新闻背景

Agent Skill

midjourney-card-news-backgrounds 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

643

周安装

26

GitHub Stars

836

下载量

202
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

3

许可证

MIT

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:midjourney-card-news-backgrounds(中途卡新闻背景)
来源仓库:https://github.com/bear2u/my-skills
仓库路径:skills/midjourney-card-news-backgrounds
安装命令:
npx skills add https://github.com/bear2u/my-skills --skill midjourney-card-news-backgrounds
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/bear2u/my-skills --skill midjourney-card-news-backgrounds

简介

用于 Midjourney 新闻卡片背景素材的处理与管理。

  • 适合围绕视觉素材生成或协作事项进行整理。
  • 通过 GitHub 安装,建议查看原始 README 了解使用方式。
  • 涉及图片生成时应注意版权与输出格式限制。
  • midjourney-card-news-backgrounds 属于AI 工具类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Midjourney Card News Background Generator

Generate optimized Midjourney prompts for 600x600px card news backgrounds that work well with text overlays.

When to Use This Skill

Use this skill when the user requests:

  • Card news background images
  • Instagram post backgrounds
  • Social media graphic backgrounds
  • 600x600 or 1:1 aspect ratio backgrounds

Core Workflow

  1. Identify the topic: Extract the main subject from user's request
  2. Determine style category: Match to predefined style patterns
  3. Generate prompt: Create Midjourney prompt with optimal parameters
  4. Provide variations: Offer 3-5 different style options

Prompt Structure

All prompts follow this structure:

[subject/scene description], [style keywords], [color palette], [texture/atmosphere], [technical specs] --ar 1:1 --v 6

Essential Components

Subject description (5-10 words):

  • Describe the visual elements clearly
  • Use specific, vivid adjectives
  • Avoid abstract concepts

Style keywords (3-5 keywords):

  • minimal, clean, modern, professional
  • gradient, abstract, geometric
  • organic, fluid, soft, dreamy

Color palette (specific colors):

  • Name exact colors: "soft blue and purple" not "cool colors"
  • Include intensity: "vibrant", "pastel", "muted", "bright"
  • Maximum 3-4 colors for cohesion

Texture/atmosphere (2-3 keywords):

  • smooth, flowing, textured, grainy
  • light and airy, bold and dramatic
  • subtle, prominent, delicate

Technical specifications:

  • Always include: --ar 1:1 (for 600x600 square format)
  • Version: --v 6 (current Midjourney version)
  • Optional: --s 50-200 for stylization control

Style Categories

For quick reference, match topics to these categories:

Business/Tech

  • Clean gradients with geometric elements
  • Blue, purple, teal color schemes
  • Professional and modern feel

Health/Wellness

  • Soft pastels with organic shapes
  • Green, peach, soft pink tones
  • Calming and natural atmosphere

Finance/Investment

  • Bold gradients with sharp lines
  • Navy, gold, green color schemes
  • Confident and premium feel

Education/Learning

  • Friendly colors with simple shapes
  • Yellow, orange, light blue tones
  • Approachable and energetic mood

Food/Lifestyle

  • Warm tones with natural textures
  • Earth tones, warm oranges, browns
  • Inviting and authentic feel

Creative/Art

  • Bold abstract patterns
  • Vibrant multi-color schemes
  • Expressive and dynamic energy

Text Overlay Optimization

All backgrounds must be text-friendly:

Contrast zones: Include areas with:

  • Subtle gradients (not busy patterns)
  • Consistent brightness in center
  • Darker or lighter edges for visual interest

Space planning:

  • Keep center 60% relatively uniform
  • Place complex elements in corners
  • Avoid horizontal lines through center

Avoid:

  • High-contrast busy patterns
  • Text-competing elements in center
  • Overly detailed textures

Example Prompts by Topic

Tech/AI:

abstract neural network patterns, modern tech aesthetic, gradient blue and cyan tones, smooth digital waves, clean negative space for text, futuristic minimalism --ar 1:1 --v 6

Fitness/Exercise:

soft flowing energy waves, dynamic movement feel, gradient coral and peach colors, light and motivating atmosphere, space for text overlay --ar 1:1 --v 6

Finance/Money:

elegant geometric patterns, premium professional style, navy and gold gradient, subtle texture with depth, sophisticated minimal design --ar 1:1 --v 6

Food/Cooking:

organic food texture background, warm earthy tones, rustic natural aesthetic, soft focus with gentle shadows, appetizing color palette --ar 1:1 --v 6

Mental Health:

calming abstract clouds, serene peaceful atmosphere, soft lavender and mint gradients, dreamy gentle textures, meditative minimal space --ar 1:1 --v 6

Response Format

When user provides a topic, respond with:

  1. Primary prompt: The most suitable style
  2. Alternative 1: Different mood/style for same topic
  3. Alternative 2: Another variation
  4. Brief explanation: Why these styles work for the topic

Example response:

주제: 재테크 팁

1. 메인 추천:
[prompt here]
→ 전문적이고 신뢰감 있는 금융 분위기

2. 대안 1:
[prompt here]
→ 친근하고 접근하기 쉬운 느낌

3. 대안 2:
[prompt here]
→ 프리미엄하고 고급스러운 느낌

Tips for Better Results

Do:

  • Use specific color names
  • Include texture descriptors
  • Specify mood/atmosphere
  • Keep prompts under 60 words
  • Test with simple subjects first

Avoid:

  • Generic terms like "nice" or "good"
  • Too many competing elements
  • Complex scene descriptions
  • Overly long prompts
  • Requesting specific brands/logos

Advanced Parameters

For fine-tuning (optional):

  • --s 50: More realistic, less stylized
  • --s 150: Balanced stylization (default)
  • --s 250: More artistic interpretation
  • --chaos 0-100: Variation control (0 = consistent)

Korean Language Support

When user asks in Korean:

  • Understand Korean topic descriptions
  • Translate concepts to English prompts
  • Provide explanations in Korean
  • Keep prompts in English (Midjourney requirement)

Common Topics Reference

Read topics_reference.md for comprehensive list of:

  • Industry-specific color schemes
  • Seasonal variations
  • Trend-based styles
  • Cultural considerations

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

04

需要参考平台分布和安装热度时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Claude Code

31.06%
按下载量换算63

OpenCode

24.58%
按下载量换算50

Cursor

15.62%
按下载量换算32

Antigravity

13.2%
按下载量换算27

Gemini CLI

7.6%
按下载量换算15

windsurf

3.53%
按下载量换算7

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。

来源信息

继续浏览同类 Skills