nanobanana

To generate AI images from text prompts using Gemini models, you provide detailed text descriptions to the Skill, which then creates visual art by processing your input via the nanobanana MCP server. It supports detailed prompt-based image creation.

Yes, you can edit existing images conversationally with Google Gemini. The Skill supports conversational image modification, allowing you to alter provided input images using natural language instructions.

Yes, you need a GEMINI_API_KEY to use this AI image generation Skill. The Skill requires this key to authenticate and access Google Gemini models for creating and editing visual content.

You control aspect ratios and quality when generating AI art by specifying desired dimensions and selecting different model tiers. The Skill supports aspect ratio control and model tier selection (flash, nb2, pro) for varying speed and quality needs.

The difference between the flash, nb2, and pro model tiers for image generation lies in speed and quality. These different model tiers allow users to select the appropriate balance of processing speed and output fidelity for their specific visual creation requirements.

You should use AI image editing instead of generating new images when you need to modify existing visuals based on natural language instructions. This approach is ideal for altering provided input images conversationally rather than creating entirely new artwork from text prompts.