name: graphic-design
description: |-
This skill creates modern, aesthetic visual designs using Google Gemini image generation via MCP Playwright browser access. The skill generates high-quality design images, downloads them, and organizes them locally in the graphic-designs directory following modern aesthetic principles. This skill should be used when users need professional graphic designs for tech platforms, AI tools, learning products, or personal portfolios.

Graphic Design Skill

This skill creates modern, aesthetic visual designs using Google Gemini image generation via MCP Playwright browser access. The skill generates high-quality design images, downloads them, and organizes them locally following professional design standards.

Core Capabilities

1. Gemini Integration: Uses MCP Playwright to access Google Gemini for image generation
2. Modern Aesthetics: Creates designs following professional design principles
3. Image Download & Organization: Downloads generated images and saves them in organized directory structure
4. Prompt Construction: Converts user requests into detailed, structured image prompts
5. Professional Output: Generates images suitable for tech platforms, AI tools, and professional use

Design Aesthetic Requirements (VERY IMPORTANT)

All generated images must follow these principles:

• Modern: Contemporary design approach
• Clean: Uncluttered and purposeful
• Minimal: Focus on essential elements
• Professional: Suitable for business use
• High visual hierarchy: Clear information organization
• Balanced spacing: Proper visual breathing room
• Readable text areas: Consideration for overlay text
• Soft shadows: Subtle depth effects
• Subtle gradients: Gentle color transitions
• Muted or neutral color palettes: Sophisticated color choices

Avoid:
- Cartoon styles
- Loud or neon colors
- Overcrowded layouts
- Low-contrast text
- Generic clip-art visuals

Prompt Construction Rules (Critical)

Before sending a prompt to Gemini, the skill will:

1. Convert the user request into a detailed, structured image prompt
2. Explicitly define:
3. Layout style
4. Color tone
5. Mood
6. Lighting
7. Composition
8. Design purpose
9. Ensure the prompt is descriptive, precise, and visual

Image Generation Rules

• Generate 1–3 images per request
• Prefer high-resolution outputs
• Ensure designs leave space for text overlays
• Each image should feel intentional and usable

Download & File Handling Rules

After image generation:

1. Download each image from Gemini
2. Create directory if missing: graphic-designs/
3. Save images using descriptive filenames:
4. graphic-designs/linkedin-banner-modern-01.png
5. graphic-designs/ai-dashboard-hero-02.png
6. Do NOT overwrite files unless explicitly instructed

Input Parameters

• Design Purpose: The purpose of the design (e.g. LinkedIn post, banner, thumbnail)
• Style Keywords: Optional style keywords to influence the design
• Color Preferences: Optional color preferences for the design
• Size/Dimensions: Required dimensions or aspect ratio for the image (e.g., "1080x1080", "16:9", "1920x1080")

Output

A summary including:
- Number of images generated
- File names saved
- Design style used
- Location where files are saved

Before Implementation

Gather context to ensure successful implementation:

Ensure all required context is gathered before implementing.
Only ask user for THEIR specific requirements (domain expertise is in this skill).

Authentication Requirements

• The skill will open https://gemini.google.com
• If authentication is required, execution will pause
• User will be asked to log in manually
• Execution continues only after Gemini is accessible