design

Generates logos, CIP mockups, SVG icons, banners, slides, and social images via Gemini AI and HTML rendering.

Updated Aug 31, 2026
One-click install
npx skills add https://github.com/MANR255926/GARAGE --skill design-manr255926
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: design
Source: https://github.com/MANR255926/GARAGE/tree/main/.agents/skills/design
Command: npx skills add https://github.com/MANR255926/GARAGE --skill design-manr255926

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires google-genai, pillow, and includes scripts (resource) and references (resource) components.

What problem does it solve? Producing brand assets like logos, corporate identity mockups, icons, banners, and social media images normally requires multiple design tools and manual resizing for each platform. This Skill unifies those tasks into scriptable workflows that generate assets with Gemini AI models and export pixel-exact images from HTML/CSS. ## Core Features & Use Cases - Logo & CIP Generation: Search 55+ logo styles and 50+ corporate identity deliverables with BM25 search scripts, then generate mockups using Gemini Flash or Pro models. - SVG Icon Generation: Create icons in 15 styles (outlined, filled, duotone, etc.) with batch variations and multi-size export via Gemini 3.1 Pro text output. - Banners, Slides & Social Photos: Design HTML/CSS banners and social images at exact platform dimensions (Instagram, Facebook, LinkedIn, YouTube, Google Ads) and export via headless Chrome screenshots; build Chart.js pitch decks from layout and copywriting references. - Use Case: A startup founder needs a complete brand package: generate a minimalist logo, produce business card and signage mockups with that logo, then render an HTML presentation of the full CIP set. ## Quick Start Ask the AI to design a minimalist logo for your brand named TechFlow and generate matching business card mockups.

Frequently Asked Questions about design

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate a logo with Gemini AI?▼

Run scripts/logo/generate.py with --brand, --style, and --industry flags after setting the GEMINI_API_KEY environment variable. The script supports 55+ styles such as minimalist, vintage, and luxury, and outputs logo images on a white background.

How to create corporate identity mockups with an existing logo?▼

Use scripts/cip/generate.py with --brand, --logo pointing to your logo file, --deliverable, and --industry. Add --set to generate a full CIP set or --model pro for 4K text rendering with gemini-3-pro-image-preview.

What banner sizes does it support for social media and ads?▼

The banner reference covers Facebook covers (820x312), YouTube channel art (2560x1440), Instagram stories (1080x1920), Google Ads formats like 300x250 and 728x90, website heroes, and print sizes with 300 DPI and bleed specifications.

Can it generate SVG icons without an image generation API?▼

Yes. The icon script uses gemini-3.1-pro-preview, which outputs SVG as XML text, so no image generation endpoint is needed. It supports 15 styles, batch variations, and multi-size export at 16, 24, 32, and 48 pixels.

What dependencies are required to run the design scripts?▼

The Python scripts require the google-genai and pillow packages, installed via pip, plus a GEMINI_API_KEY from Google AI Studio. Social photo export additionally uses headless Chrome, chrome-devtools, or Playwright for screenshots.

Why do exported social images show wrong fonts or broken layouts?▼

Screenshots taken before web fonts and images finish loading cause fallback fonts and layout shifts. Add a 3-5 second delay or use --virtual-time-budget with headless Chrome, then visually verify exports and re-export after fixing the HTML.