Generate images from prompts via Gemini Nano Banana models and save local artifacts.
npx skills add https://github.com/eaglhuang/cocos_new --skill nano-banana-gen
Please help me install this Agent Skill. Skill: nano-banana-gen Source: https://github.com/eaglhuang/cocos_new/tree/main/.github/skills/nano-banana-gen Command: npx skills add https://github.com/eaglhuang/cocos_new --skill nano-banana-gen
Nano Banana image generation requires wrapping Gemini models behind a stable CLI wrapper to streamline asset generation and exploration workflows.
Provide a prompt and an output path to generate an image using the nano-banana-gen skill.
High-intent search queries and answers about installing and using this skill.
To generate UI assets using Gemini image models, you can use a CLI wrapper that calls the Nano Banana family models. You provide a prompt and output path, and the tool saves the generated concept art or UI drafts locally while returning image metadata.
Yes, you can use prompt files for AI art generation by passing the --prompt-file argument to the CLI. This allows you to load complex prompts from a file rather than typing them inline via the --prompt argument.
The best way to securely handle API keys for Gemini image generation is to store your Google AI key in a .env file located in tools_mcp/nano-banana-mcp/. This ensures keys are kept out of your prompts and command line history.
Yes, the Nano Banana CLI supports different model versions for concept exploration. You can choose between nano-banana, nano-banana-2, and nano-banana-pro models to generate assets based on your specific design requirements.
You save generated images locally for asset pipelines by specifying an output path when running the generation command. The tool downloads the resulting image from the Gemini model to that local path and returns its metadata.
m2ai-portfolioGenerate images via the Gemini API with Nano Banana Pro prompting.
neo4j-labsGenerate Excalidraw JSON diagrams from textual descriptions.
locdinh209Construct AI image generation prompts for Gemini and GPT-4o models.
flight505Generate and edit images via OpenRouter AI models from Python.
