Summary

Banana Claude is an AI image generation skill for Claude Code, transforming it into a Creative Director. It interprets user intent, leverages domain expertise, and constructs optimized prompts using Google's 5-component formula to orchestrate Gemini models for advanced image creation and editing, offering superior control beyond simple API wrappers.

How It Works

The system employs Claude Code as an orchestrator, performing intent analysis to discern user needs. It selects a relevant 'domain mode' (e.g., Cinema, Product, UI) and constructs detailed prompts following a 5-component structure: Subject, Action, Location/Context, Composition, and Style. These prompts are adapted for Gemini models, with integrated post-processing and support for batch variations and multi-turn session consistency.

Quick Start & Requirements

Installation options include a recommended Claude Code plugin (/plugin marketplace add AgriciDaniel/banana-claude) or standalone scripts (git clone ... && bash install.sh). A one-liner curl command is also available. Requires Node.js 18+ for npx. A Google AI API key is mandatory; the free tier offers limited RPM/RPD (~5-15 RPM / ~20-500 RPD), with significant cuts expected in Dec 2025. ImageMagick is optional for advanced post-processing. Links: Google AI Studio for API key.

Highlighted Details

• Intent Analysis: Accurately interprets user requirements.
• Domain Modes: Specialized modes (Cinema, Product, Portrait, UI, Logo, etc.) tailor generation.
• 5-Component Prompt Formula: Enables detailed prompt construction (Subject, Action, Context, Composition, Style).
• High-Resolution Output: Supports up to 4096x4096 resolution.
• Aspect Ratio Control: Offers 14 aspect ratios, including 21:9.
• Session Consistency: Maintains character and style across multi-turn interactions.
• Prompt Database: Access to over 2,500 curated prompts.

Maintenance & Community

Developed by AI Workflow Architect @AgriciDaniel. Supported by the AI Marketing Hub community (2,800+ members) and educational content on YouTube.

Licensing & Compatibility

Released under the permissive MIT License, allowing for commercial use and integration into closed-source projects.

Limitations & Caveats

The primary limitation is the restrictive rate limit of the free Google AI API tier, which may hinder high-volume usage and is scheduled for significant reduction in late 2025. Advanced post-processing capabilities depend on the optional ImageMagick installation.