imagegen

Generates design reference images for web sections, mobile screens, brand kits, and image-to-code builds.

1|Updated Aug 17, 2026
One-click install
npx skills add https://github.com/NalinDalal/skillset --skill imagegen-nalindalal
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: imagegen
Source: https://github.com/NalinDalal/skillset/tree/main/skills/ui/imagegen
Command: npx skills add https://github.com/NalinDalal/skillset --skill imagegen-nalindalal

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Designers and developers often need consistent, high-quality visual references before building a landing page, app, or brand identity, but ad-hoc image generation produces inconsistent palettes, unreadable text, and generic AI-looking results. This Skill enforces a disciplined generation workflow so every image set is coherent, readable, and directly implementable by a coder. ## Core Features & Use Cases - Four generation modes: web section references, mobile app screens in device mockups, 3x3 brand kit boards, and an image-to-code flow that implements what it generated. - Design consistency rules: one controlled palette per project, a design bible before any multi-image set, one image per section, and anti-slop checklists covering continuity, framing, and readability. - Use Case: Ask for a full marketing site reference and receive 8 labeled horizontal section images (hero, trust bar, features, showcase, benefits, testimonials, pricing, CTA) sharing one palette, type scale, and CTA family, ready to hand to a developer. ## Quick Start Generate a set of landing page design reference images for my product's marketing site, one horizontal image per section.

Frequently Asked Questions about imagegen

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

FAQPage Schema
How do I generate landing page design references with AI image models?▼

Use the web references mode, which produces one horizontal image per section in 16:9, 16:10, or 21:9 formats. A full marketing site defaults to 8 sections, each labeled and sharing one palette, type scale, and CTA family for continuity.

How to generate mobile app screen concepts with AI?▼

Use the mobile mode, which generates one screen per image framed inside a phone mockup with consistent device style and margins. Pick a platform mode first (iOS-native, Android-native, or cross-platform neutral) and never mix patterns across platforms.

Can I build a website from AI-generated design images?▼

Yes, the image-to-code mode generates section reference images first, deeply inspects them as the specification, then implements code matching the design faithfully. It never crops sections from larger images and regenerates unclear sections as fresh standalone images.

What makes AI-generated design images look generic and how to avoid it?▼

Generic results come from purple-blue gradients, default left-text-right-image layouts, fake copy, and template boards. This Skill rejects those patterns, requiring varied composition anchors, background modes, and a deliberate second-read moment per page.

What are the limitations of AI image generation for design references?▼

Generated images are references a coder recreates, not production assets, so text rendering and exact spacing require manual implementation. Multi-image sets need a design bible first, and small or crowded text means the image must be regenerated.