Skip to content
BranchesTags

Repository files navigation

🍌 satware® AI Nano Banana Editor Pro

Professional-grade image generation and editing for TypingMind using Google Gemini 2.5 Flash Image

License: MIT Version Security

Features

🎨 Dual Mode Operation

  • Generate Mode - Create stunning images from text prompts
  • Edit Mode - Transform uploaded images with AI-powered editing
  • Tool Chaining - Edit previously generated images seamlessly

🚀 Professional Enhancements

  • 9 Preset Modes - Professional Photography, Promotional Products, Corporate Branding, Social Media Content, and more
  • Industry Optimization - 10 industry-specific enhancements (Automotive, Healthcare, Technology, Finance, etc.)
  • Brand Compliance - 6 brand style modes (Strict Corporate, Luxury Premium, Startup Modern, etc.)
  • Style Preferences - 7 visual styles (Photorealistic, Anime/Manga, Oil Painting, Watercolor, etc.)

⚙️ Quality Controls

  • 3 Quality Presets - Maximum Consistency, Balanced (default), Creative Exploration
  • Custom Parameters - Manual control over temperature, topP, topK
  • Multi-Image Support - Generate 1-4 images per request
  • Safety Filters - 4-tier safety system (Disabled, Minimal, Balanced, Strict)

Quick Start

1. Get Your API Key

  1. Visit Google AI Studio
  2. Create or select a project
  3. Generate an API key
  4. Keep it secure - never share publicly

2. Install Plugin

Method 1: GitHub Import (Recommended)

1. Copy this repository URL: https://github.com/satwareAG/plugin-gemini-flash-image
2. In TypingMind, go to Plugins → Import
3. Paste the GitHub URL
4. Click Import

Method 2: Manual Installation

1. Download plugin.json and implementation.js
2. In TypingMind, go to Plugins → Import from JSON
3. Paste the plugin.json content
4. Save

3. Configure Settings

  1. Open plugin settings in TypingMind
  2. Required: Enter your Gemini API Key
  3. Optional: Customize defaults:
    • Default Image Count (1-4)
    • Quality Preset (Balanced recommended)
    • Prompt Enhancement Mode (None default)
    • Safety Mode (Balanced recommended)
    • Industry Focus, Brand Compliance, Style Preference

4. Start Creating

Generate a new image:

"Generate a professional product photo of a sleek smartwatch"

Edit an uploaded image:

1. Upload an image to the chat
2. "Make this image brighter and more vibrant"

Tool chaining:

1. "Generate a sunset over mountains"
2. "Edit the previous image to add a lake in the foreground"

Usage Examples

Basic Generation

Prompt: "Create a modern office workspace"
Mode: Generate (no image uploaded)
Output: Professional image of modern office

Professional Photography

Settings: Enhancement = "Professional Photography"
Prompt: "A coffee cup on a wooden table"
Output: Studio-quality product photography

Industry-Specific

Settings: 
  - Enhancement = "Promotional Products"
  - Industry = "Technology"
  - Brand = "Startup Modern"
Prompt: "Laptop with company logo"
Output: Marketing-ready tech product image

Image Editing

1. Upload your product photo
2. Prompt: "Remove background and enhance lighting"
Mode: Edit (automatic with uploaded image)
Output: Enhanced product on clean background

Creative Exploration

Settings: Quality Preset = "Creative Exploration"
Prompt: "Abstract representation of artificial intelligence"
Output: Highly creative, artistic interpretation

Configuration Reference

User Settings

Setting Type Default Description
Gemini API Key Password Required Your Google Gemini API key
Default Image Count Number 1 Number of images to generate (1-4)
Quality Preset Enum Balanced Consistency/creativity balance
Prompt Enhancement Mode Enum None Professional preset mode
Custom Prompt Enhancement Text - Your own enhancement instructions
Industry Focus Enum - Industry-specific optimization
Brand Compliance Enum - Brand style enforcement
Default Style Enum - Visual style preference
Safety Mode Enum Balanced Content filter strictness
Temperature Number 0.7 Randomness (0.0-1.0)
Top P Number 0.9 Nucleus sampling
Top K Number 40 Top-K sampling

Enhancement Modes

Professional Presets:

  • Professional Photography - Studio lighting, commercial quality
  • Promotional Products - Clean backgrounds, marketing-ready
  • Artistic Creative - Unique perspectives, museum-worthy
  • Corporate Branding - Enterprise-grade, executive level
  • Social Media Content - Platform-optimized, viral quality
  • Technical Illustration - Precise documentation style
  • Lifestyle Photography - Natural, authentic feel
  • Product Mockup - Professional staging, showcase quality
  • Custom - Use your own enhancement instructions

Quality Presets

Preset Temperature Top P Top K Best For
Maximum Consistency 0.2 0.7 20 Brand materials, product shots
Balanced 0.7 0.9 40 General purpose (default)
Creative Exploration 0.9 0.95 80 Artistic projects, concepts
Custom User-defined User-defined User-defined Advanced users

Safety Modes

Mode Threshold Use Case
Disabled No filtering Private experimentation only
Minimal High-risk only Broad creative freedom
Balanced Medium+ (default) Recommended for most users
Strict Low+ Corporate, educational settings

Technical Details

API Information

  • Model: Gemini 2.5 Flash Image (Generally Available since October 7, 2025)
  • Endpoint: generativelanguage.googleapis.com/v1beta
  • Resolution: 1024px (automatic optimization)
  • Cost: ~$0.039 per image (~1,290 tokens)
  • Rate Limits: 300 requests/min (paid), 1,500/day (free tier)
  • Features: SynthID watermarking included

Permissions Required

  • read_last_user_message - Access user prompts
  • read_uploaded_files - Access uploaded images for editing

Implementation Type

  • Type: JavaScript (client-side)
  • Output: Cards (image cards)
  • Tool Chaining: Supported via previousRunOutput.cards

Security

  • ✅ No hardcoded secrets - User provides API key
  • ✅ HTTPS-only - Encrypted communication
  • ✅ No logging - API keys never logged
  • ✅ Local storage - Keys stored in TypingMind settings only

Troubleshooting

"Please provide your Gemini API Key"

Solution: You must enter your API key in plugin settings. Get one at Google AI Studio.

"Invalid Gemini API Key"

Solutions:

  1. Verify key is correct (copy/paste errors common)
  2. Ensure key has image generation permissions enabled
  3. Check if key is active in Google Cloud Console

"Content blocked by safety filters"

Solutions:

  1. Adjust Safety Mode to "Minimal" (settings)
  2. Modify your prompt to be less explicit
  3. Review blocked categories in error message
  4. Consider if content violates Google's policies

"No image generated"

Solutions:

  1. Check if prompt is specific enough
  2. Verify API quota not exceeded (Google Cloud Console)
  3. Try reducing image count (n parameter)
  4. Wait if rate-limited (300/min cap)

Image quality not as expected

Solutions:

  1. Try different Quality Preset (Creative Exploration for more variation)
  2. Use Prompt Enhancement Mode for professional results
  3. Adjust temperature/topP/topK parameters manually
  4. Add more detail to your prompt
  5. Use Style Preference to guide aesthetic

Rate Limits & Costs

Google Gemini API Limits

Free Tier:

  • 1,500 requests per day
  • Rate limit: 15 requests per minute

Paid Tier:

  • Rate limit: 300 requests per minute
  • Cost: $0.039 per image (~1,290 tokens)
  • No daily cap (usage-based billing)

Best Practices

  1. Start with free tier to test
  2. Monitor usage in Google Cloud Console
  3. Set up billing alerts
  4. Use single-image requests for testing
  5. Batch similar requests when possible

Contributing

We welcome contributions! Please see our Contributing Guidelines (coming soon).

Development Setup

git clone https://github.com/satwareAG/plugin-gemini-flash-image.git
cd plugin-gemini-flash-image
# Edit plugin.json or implementation.js
# Test in TypingMind via Import from JSON

Code Quality Standards

  • Follow satware AG coding standards
  • No hardcoded secrets
  • Comprehensive error handling
  • Professional JSDoc documentation
  • Clear, actionable error messages

Security

Reporting Vulnerabilities

Do NOT create public GitHub issues for security issues.

Email: ja@satware.com (Jane Alesi, Lead AI Architect)

See SECURITY.md for full security policy and previous vulnerability disclosures.

Key Security Notes

  • ✅ v1.0.0 has no hardcoded API keys (secure)
  • ⚠️ Development files in raw/ directory contained exposed key (fixed)
  • Action: Revoke key AIzaSyAcfq9qXDxw9dP6hZuRrRLgXfGuRoSv0WE if used

Changelog

See CHANGELOG.md for version history and release notes.

License

MIT License - see LICENSE file for details.

Copyright (c) 2025 satware AG

Links

Support

Community Support

  • GitHub Issues for bugs and feature requests
  • Discussions for questions and ideas

Professional Support

Made with 🍌 by satware AG

About

satware® AI Plugin for Gemini Image Generation

Resources

Readme
MIT license

Security policy

Security policy
Activity
Custom properties

Stars

0 stars

Watchers

0 watching

Forks

0 forks
Report repository

Releases

Packages

Contributors

Languages