Nano Banana Slides Prompter

Generate optimized prompts for Nano Banana Pro Slides - Generate slides using AI.

🆕 v1.0.6 Update - Dynamic Character & Slide Generation!

Major improvements to character generation and slide type intelligence!

What's New:

• 🎭 Dynamic Character Generation - Characters are now generated based on your content, topic, and audience instead of using generic defaults
• 🧠 Content-Aware Slide Types - LLM analyzes your content and selects optimal slide types from 50+ templates
• 🎯 Hybrid Slide System - Smart fallback to category-based templates if analysis fails
• 🦊 Non-Human Characters - Support for animals, robots, mascots, animated objects based on content context
• ⚡ Parallel Processing - Character and content analysis run simultaneously for no added latency

Slide Type Categories (50+ templates):

• Opening: dramatic titles, problem statements, hooks
• Concept: diagrams, metaphors, breakdowns, ecosystems
• Data: charts, dashboards, trends, correlations
• Process: timelines, flowcharts, flywheels, journeys
• Technical: architecture, stacks, APIs, infrastructure
• Business: strategy, ROI, market analysis, roadmaps
• Comparison: side-by-side, before/after, feature matrices
• Examples: case studies, testimonials, demos
• Educational: definitions, examples, quizzes
• Closing: summaries, CTAs, Q&A

New Sample Slides (v1.0.6):

---

🆕 v1.0.5 Update - Internationalization & Session Management!

Major update with internationalization support and session management features!

What's New:

• 🌐 Internationalization (i18n) - Full support for English and Chinese (中文) with react-i18next
• 💾 Session Management - Save, load, and manage multiple presentation sessions
• ⚙️ LLM Settings Dialog - Configure API keys, base URL, and model per user in browser
• 🔄 State Management - Zustand-based store for better state handling
• 🔒 Security Updates - All npm vulnerabilities fixed (7 → 0)
• ⚡ Vite 7 - Upgraded to latest Vite for better performance
• 📦 Dependency Updates - 120+ packages updated for security and stability

Technical Improvements:

• Hybrid LLM configuration: server defaults + optional user overrides
• Session persistence with localStorage
• RESTful API endpoints for sessions and settings
• Improved error handling and user feedback
• Enhanced UI with session sidebar

Credits: Special thanks to @movclantian for contributing the internationalization and session management features!

---

🆕 v1.0.4 Update - Character Presenter!

Introducing the Character Presenter feature! Add a consistent animated presenter across all your slides with extensive customization options.

What's New:

• 🎭 Character Presenter - Add an animated character to guide your presentation
• 🎨 8 Render Styles - Pixar, Real, Anime, Cartoon, Sketch, Chibi, Low-Poly, Mascot
• ⚧️ Gender Options - Male, Female, or let AI decide
• 🎯 24 Character Combinations - Mix and match styles with gender preferences
• 🔄 Adaptive Styling - Characters automatically adapt to match your slide style

New Sample Slides (v1.0.4):

## 🆕 v1.0.3 Update - Even More Amazing Results!

We've made significant improvements to the prompt generation engine. The results are now more cinematic, feature-rich, and visually stunning than ever before!

What's New:

• 🎭 Style Personas - Each style now has a unique creative identity that guides the AI
• 🎨 Expanded Visual Vocabulary - 200+ visual terms across 24 categories
• 📝 Chain-of-Thought Prompting - AI now thinks through each visual decision
• 🖼️ Richer Backgrounds - More texture, pattern, and atmospheric options
• 📄 CSV File Support - Upload CSV data for data-driven presentations

New Sample Slides (v1.0.3):

Sample Slides

Slides generated using prompts from this tool:

Features

• Multiple visual styles (Professional, Technical, Creative, Educational, and more)
• Character Presenter with 8 render styles and 24 customization combinations
• Configurable color palettes and layouts
• Collapsible slide previews with individual copy buttons
• URL content extraction for automatic content generation
• CSV file upload for data-driven presentations
• Configurable LLM backend (OpenAI, OpenRouter, Ollama, etc.)

Quick Start

Prerequisites

• Node.js 18+
• Bun runtime (for backend)
• OpenAI-compatible API key

Installation

# Install frontend dependencies
npm install

# Install backend dependencies
cd server && bun install && cd ..

Configuration

Create /server/.env from the example:

cp server/.env.example server/.env

Edit the .env file with your LLM configuration:

OPENAI_API_BASE=https://api.openai.com/v1
OPENAI_API_KEY=your-api-key
OPENAI_MODEL=gpt-4o
PORT=3001

Development

# Run both frontend and backend
npm run dev

• Frontend: http://localhost:8080
• Backend: http://localhost:3001

Docker Deployment

You can run the entire stack locally using Docker Compose.

Prerequisites

• Docker
• Docker Compose

Running with Docker

1. Build and Run:

   docker-compose up --build

2. Access the Application:

   • Open http://localhost:8080 in your browser.

The docker-compose.yml orchestrates:

• Frontend: Served via Nginx on port 8080.
• Backend: Hono server running on internal port 3001.

Deploy from Images (Production)

To deploy using the pre-built images from GitHub Container Registry (GHCR), create a docker-compose.yml file on your server:

services:
  frontend:
    image: ghcr.io/nomie7/nano-banana-slides-prompter-frontend:latest
    ports:
      - "80:80"
    depends_on:
      - backend
    restart: always

  backend:
    image: ghcr.io/nomie7/nano-banana-slides-prompter-backend:latest
    environment:
      - PORT=3001
      - OPENAI_API_BASE=${OPENAI_API_BASE:-https://api.openai.com/v1}
      - OPENAI_API_KEY=your-api-key-here
      - OPENAI_MODEL=gpt-4o
    restart: always

Then run:

docker-compose up -d

Tech Stack

• Frontend: React, TypeScript, Vite, Tailwind CSS, shadcn/ui
• Backend: Bun, Hono, OpenAI SDK
• UI Components: Radix UI primitives

Version

v1.0.6