snapai  by betomoedano

AI-powered icon generation CLI for React Native & Expo

Created 2 months ago
1,087 stars

Top 35.0% on SourcePulse

GitHubView on GitHub
Project Summary

SnapAI is a command-line interface tool designed for React Native and Expo developers to generate app icons using OpenAI's image generation models. It streamlines the process of creating professional-quality icons quickly and efficiently, catering to developers who need custom icons without extensive design work.

How It Works

SnapAI leverages OpenAI's DALL-E 2, DALL-E 3, and GPT-Image-1 models to generate images based on user-provided text prompts. The CLI allows users to specify models, sizes, quality settings, and output formats, offering flexibility for various use cases. Its design prioritizes developer workflow, enabling integration into CI/CD pipelines and batch generation for multiple icon variations.

Quick Start & Requirements

  • Installation: npm install -g snapai or use npx snapai.
  • Prerequisites: An OpenAI API key is required.
  • Setup: Configure API key with snapai config --api-key sk-your-openai-api-key-here.
  • Generate: snapai icon --prompt "minimalist weather app with sun and cloud".
  • Docs: https://github.com/betomoedano/snapai

Highlighted Details

  • Supports multiple OpenAI models (GPT-Image-1, DALL-E 3, DALL-E 2) with distinct quality, speed, and cost trade-offs.
  • Offers advanced options including transparent backgrounds, various output formats (PNG, JPEG, WebP), and batch image generation.
  • Designed for CI/CD integration and professional workflows, with clear cost optimization tips.
  • Emphasizes privacy with zero data collection and local API key storage.

Maintenance & Community

The project is open-source under the MIT License. Development setup instructions are available for contributors.

Licensing & Compatibility

MIT License. Compatible with commercial and closed-source projects.

Limitations & Caveats

Some advanced features like transparent backgrounds and specific output formats are only available with the GPT-Image-1 model. DALL-E 3 is limited to one image generation per request.

Health Check
Last Commit

3 days ago

Responsiveness

Inactive

Pull Requests (30d)
3
Issues (30d)
5
Star History
179 stars in the last 30 days

Explore Similar Projects

Feedback? Help us improve.