nano-banana-image-generator

原文涵盖的主题

• Nano Banana Image Generator
• Workflow Overview
• Step 1: Brainstorm Concepts
• When to Ask Clarifying Questions
• Generating Concepts
• Step 2: Source Reference Photos (Person-Based Images)
• Why This Matters
• Process

Full Production Studio note: this skill comes from another creator's own setup. Brand names, people and folder paths in it (for example OpenEd, Skill Stack, Howseen, /Desktop/...) are their examples. Replace them with the user's brand, voice and folders (use the arabic-voice Voice Card when available) and never write in that other brand's voice. Generate professional, non-generic images using Nano Banana Pro (Gemini API). 1. Brainstorm Concepts - Generate 4-6 high-level visual ideas 2. Select Direction - User picks the concept they like 3. Source Reference Photos - Find high-res images of real people/places (if applicable) 4. Optimize Prompt - Refine into a strong, detailed prompt 5. Style Variations - Adapt to 2-3 different visual styles 6. Generate Images - Run via Gemini API Before brainstorming, assess if you have enough information. Ask 2-4 focused questions if: Subject is unclear or too generic Purpose/Context is missing (what's this for?) Style preferences are unspecified Text requirements are ambiguous Skip questions if: The user provides a detailed brief or says "just generate it." Don't create friction when the request is already clear. Example: User: "Can you write me a

这里只展示节选，请以完整原文和当前要求为准。