Nano Banana Image Gen

What This Skill Does

Nano Banana Image Gen is a robust AI image generation tool integrated directly into the OpenClaw ecosystem. Powered by the Google Gemini 2.0 Flash model, it enables users to generate high-quality visual content, including YouTube thumbnails, blog headers, social media banners, and artistic graphics. A standout feature of this skill is its native support for Japanese text, making it an ideal choice for creators focusing on the Japanese market. By leveraging Gemini's advanced multimodal capabilities, users can bridge the gap between abstract ideas and polished visual assets within seconds, all through the command line or agent chat interface.

Installation

To begin using the skill, ensure you have Python 3.10 or higher installed. First, install the skill via the OpenClaw package manager using the following command: clawhub install openclaw/skills/skills/sa9saq/nano-banana-image-gen.

Once installed, you must secure your Google AI Studio API key. Open your configuration file at ~/.openclaw/secrets.env and append GOOGLE_AI_API_KEY=your_key_here. For security, ensure this file is protected with chmod 600. Finally, export the variable into your current session using: export GOOGLE_AI_API_KEY=$(grep GOOGLE_AI_API_KEY ~/.openclaw/secrets.env | cut -d= -f2). Verification of your setup can be performed by running a simple test generation command.

Use Cases

This skill is designed for versatility. Content creators can use it to produce eye-catching SNS thumbnails for platforms like Twitter or note.com. Bloggers can generate professional, thematic article covers to increase reader engagement. Marketers can design promotional banners for specific campaigns, and researchers or presenters can create visual assets to improve the impact of their slide decks and presentations.

Example Prompts

1. "Nano Banana, generate a 16:9 YouTube thumbnail showing a futuristic Tokyo skyline at sunset with neon lights and the text: AIの未来"
2. "Create a minimalist Instagram square image, 1:1, featuring a steaming cup of coffee on a wooden table, warm tones, photorealistic."
3. "Make a vertical banner for a story about digital art, using a vibrant and colorful oil painting style."

Tips & Limitations

To get the best results, specificity is key. Avoid vague requests; instead, describe the style, lighting, and mood in detail. If you need text in the image, clearly state the text within quotation marks. Be mindful that Gemini 2.0 has API rate limits; if you encounter errors, implement a brief delay between sequential image requests. Note that the API is programmed to reject NSFW or inappropriate content. Always review the final images, as generative AI can occasionally produce minor artifacts or layout inconsistencies that may require a second attempt.