Which models does the media-gen skill support?

The skill supports Gemini 2.5 Flash (Nano Banana), Gemini 3.0 Pro (Nano Banana Pro), Imagen 3.0, and Veo 3.1 for video generation.

Can I control the output format of the images?

Yes, you can specify aspect ratios such as 1:1, 16:9, 9:16, and 4:3 using the --aspect-ratio parameter.

Is there support for negative prompts?

Negative prompts are supported specifically when using the Imagen 3.0 model to help you exclude unwanted elements from your images.

Does this skill support video generation?

Yes, it uses Google’s Veo 3.1 models to generate high-quality video previews from either text prompts or existing image references.

How do I handle authentication for these models?

The skill requires a GEMINI_API_KEY or GOOGLE_API_KEY environment variable to be set in your environment.

Media Generator for Google GenAI

Name: Media Generator for Google GenAI
Author: ninyawee

byninyawee

Design Tools

Generates high-fidelity images and videos directly within Claude using Google's Gemini, Imagen, and Veo models.

About

The media-gen skill empowers your Claude Code environment with professional-grade media creation capabilities by wrapping the Google GenAI SDK. It provides a streamlined interface for generating 4K images with Gemini 3.0 Pro, cost-effective visuals with 'Nano Banana' (Gemini 2.5 Flash), and cinematic video content via Veo 3.1. Whether you need rapid prototyping of UI assets, complex image-to-video animations, or fine-tuned control via negative prompts and aspect ratios, this skill automates the model selection and execution logic to simplify your creative development workflow.

Key Features

Video generation using Veo 3.1 with text-to-video and image-to-video capabilities
Smart model auto-selection based on quality requirements and task type
Integration with Imagen 3.0 for advanced prompt control and negative prompt support
Asynchronous video polling system to manage long-running generation tasks automatically
0 GitHub stars
Support for Gemini 2.5 Flash and 3.0 Pro for both cost-effective and 4K image generation

Use Cases

Automating the creation of marketing materials and social media content using GenAI models
Creating short cinematic video previews and animations from text or reference images
Rapidly generating visual assets and placeholders for web and mobile development projects

About

Key Features

Video generation using Veo 3.1 with text-to-video and image-to-video capabilities
Smart model auto-selection based on quality requirements and task type
Integration with Imagen 3.0 for advanced prompt control and negative prompt support
Asynchronous video polling system to manage long-running generation tasks automatically
0 GitHub stars
Support for Gemini 2.5 Flash and 3.0 Pro for both cost-effective and 4K image generation

Use Cases

Automating the creation of marketing materials and social media content using GenAI models
Creating short cinematic video previews and animations from text or reference images
Rapidly generating visual assets and placeholders for web and mobile development projects