A comprehensive web application for creating, managing, and optimizing AI prompts to enhance assistant performance across various tasks.

Prompt Sanctuary combines the best features from three previous projects into a unified platform:

• Prompt Library - Curated collection of optimized prompts
• Prompt Generator - Dynamic prompt creation tool
• Stability Chat - Multi-modal AI assistant

• Prompt Library Access - Browse and copy proven prompts for specific use cases
• Custom Generation - Create tailored prompts using advanced generation tools
• Random Generation - Discover new approaches through experimental prompts
• Personal Library - Save, organize, and version your custom prompts
• Community Sharing - Contribute to and benefit from community-driven improvements
• Feedback System - Continuous improvement through user feedback

• Web Application: prompt sanctuary
• Streamlit Version: streamlit promptgen
• Self-Hosting: Follow the setup instructions

• Modern Design System - Tailwind CSS v3 with pastel theme and glassmorphism effects
• Enhanced Modals - Global popup system with keyboard navigation and accessibility features
• Rich Content Rendering - Markdown support with DOMPurify sanitization and Prism syntax highlighting
• Copy Functionality - One-click copying for generated content and code blocks

• Dual Language Interface - Complete English and Indonesian language support
• Instant Switching - Real-time language switching without page reloads
• Session Persistence - Language preferences saved across sessions

• Version Control - Automatic snapshots and version history for saved prompts
• Advanced Editing - Rollback functionality and prompt evolution tracking
• Community Integration - Enhanced sharing and feedback systems
• AI-Powered Titles - Smart title generation for saved prompts
• Prompt Refinement - Dedicated refinement interface with multiple sources

• Point System - Comprehensive economy with points, achievements, and rewards
• API Key Management - Personal API key integration with validation and compensation
• Usage Tracking - Detailed transaction history and point expiration management

• Enhanced Performance - SQLite WAL mode for improved concurrency
• Robust Architecture - Better error handling and retry mechanisms
• Database Migration - Production-ready migration utilities
• Latest AI Models - Updated to gemini-2.5-flash with easy model override

1. Clone the repository

   git clone https://github.com/1999AZZAR/prompt-sanctuary.git
   cd prompt-sanctuary

2. Create virtual environment

   python -m venv .venv
   source .venv/bin/activate  # On Windows: .venv\Scripts\activate

3. Install dependencies

   pip install -r requirements.txt

4. Configure environment

   # Create .env file with your Google AI Studio key(s)
   GENAI_API_KEY=your_api_key_here

5. Run the application

   python web/app.py

   The application will be available at http://127.0.0.1:5000

• Available Languages: English (default) and Indonesian
• Language Switching: Use EN/ID buttons in the sidebar navigation
• Session Persistence: Language preferences are saved across sessions
• Complete Coverage: All UI elements, forms, and templates are translated

Point-Based Economy

• Starting Balance: 80 points for new users
• Daily Bonuses: 5-12 random points per day
• Achievement Rewards: Unlock points through various activities
• Point Expiration: Different expiration periods based on source (17-95 days)

Prompt Generation Costs

Achievement System

• Automatic Unlocking: Based on user activity patterns
• Multiple Categories: Generation, sharing, consistency, and more
• Point Rewards: Each achievement grants additional points
• Progress Tracking: Monitor your statistics and achievements

Personal API Keys

• Integration: Use your own Gemini API keys for unlimited usage
• Validation: Real-time API key validation with immediate feedback
• Rewards: Earn 100 points for successful API key addition
• Cost-Free Generation: No point consumption when using personal keys
• Achievement: Unlock "API Key Provider" achievement

API Key Pool System

• Community Sharing: Validated keys are pooled for system-wide use
• Fair Distribution: LRU rotation ensures equitable usage
• Compensation: Earn 0.5 points per system usage of your key
• Usage Tracking: Detailed statistics and usage metrics
• Pool Management: Automatic refresh with new validated keys

Enhanced Point Tracking

• Transaction History: Detailed logs with source attribution
• Expiration Management: Different periods based on point source
• Point Cap: Maximum 500 points per user
• Interactive Dashboard: Click points display for detailed history

Migration System

• Safe Updates: Incremental migrations without data loss
• Automatic Backups: All operations backed up before changes
• Dry Run Mode: Preview changes before applying
• Smart Detection: Automatic database type detection
• Rollback Support: Robust error handling with recovery

Migration Commands

# Standard migration
python safe_migration.py

# Preview changes
python safe_migration.py --dry-run

# Complete rebuild (destructive)
python safe_migration.py --rebuild

# Force migration
python safe_migration.py --force

Database Features

• Schema Updates: Automatic table and column creation
• Index Optimization: Performance index management
• Data Initialization: Default data seeding
• WAL Mode: Enhanced SQLite concurrency

Intelligent Formatting

• Multi-Layer Processing: AI prompts, backend, and frontend cleaning
• Pattern Recognition: Automatic detection and correction of formatting issues
• Smart Conversion: AI formatting quirks converted to proper markdown
• Security: DOMPurify sanitization for all rendered content
• Rich Display: Proper headers, lists, code blocks, and formatting

Supported Patterns

• Square brackets: [Header]* → ## Header
• Double hash brackets: # # [Header] → ## Header
• Standalone asterisks: text* → - text or **text**
• JSON artifacts: Clean removal of formatting characters
• Indentation: Proper markdown list formatting

• Landing Page (/) - Entry point with feature overview and login/signup options
• Home Dashboard (/home) - Main application interface and starting point
• Basic Generator (/generate) - Simple content generation with text input and options
• Advanced Generator (/advance) - Advanced content generation with detailed parameters
• Prompt Refinement (/refinement) - AI-powered prompt optimization from multiple sources
• Community Library (/library) - Browse shared prompts and templates
• Personal Library (/mylib) - Manage your saved prompts with version history

Prompt Management

• Save Prompts: Store generated content in your personal library
• Version Control: Access "History" button to view, preview, and restore previous versions
• Community Sharing: Share prompts to the community library for others to use

Economy System

• Point Balance: Check your points in the sidebar navigation
• Daily Bonuses: Log in daily to earn 5-12 random points
• Achievements: Unlock achievements through various activities for additional points

API Key Integration

• Personal Keys: Add your Gemini API key in profile settings
• Benefits: No point consumption + 100 bonus points + system compensation
• Compensation: Earn 0.5 points each time your key is used by the system

Language Support

• Switching: Use EN/ID buttons in sidebar for instant language changes
• Persistence: Language preferences saved across sessions

Core Application Routes

Generation Endpoints

Prompt Management

User Management

• Python 3.8 or higher
• Google AI Studio API key(s)

1. Clone and setup environment

   git clone https://github.com/1999AZZAR/prompt-sanctuary.git
   cd prompt-sanctuary
python -m venv .venv
   source .venv/bin/activate  # Windows: .venv\Scripts\activate
pip install -r requirements.txt

2. Configure environment variables

   # Create .env file
   GENAI_API_KEY=your_api_key_here,optional_second_key

   # Optional configurations
# GENAI_MODEL_NAME=gemini-2.5-pro
   # SECRET_KEY=your-secret-key

3. Start the application

python web/app.py

Application will be available at http://127.0.0.1:5000

SQLite databases are automatically created in web/database/:

• user.db - User accounts and authentication
• prompt_data.db - Personal prompt library
• community/query.db - System prompts
• community/shared.db - Community shared prompts
• feedback.db - User feedback and ratings

Reset Data: Stop the application and delete relevant .db files to reset data.

• CSRF Protection - All POST endpoints require CSRF tokens
• Content Sanitization - DOMPurify sanitization for all rendered content
• Session Management - Enhanced session handling with proper revocation
• Input Validation - Comprehensive input validation and error handling

Requirements

• Use a WSGI server (e.g., gunicorn) instead of Flask development server
• Set a strong SECRET_KEY environment variable
• Ensure HTTPS with Secure cookies
• Configure proper GENAI_API_KEY with outbound access to Google AI APIs

Recommended Setup

# Install production dependencies
pip install gunicorn

# Run with gunicorn
gunicorn -w 4 -b 0.0.0.0:5000 web.app:app

Main dashboard with quick access to all features including the new Prompt Refinement

Personal prompt library with version control, edit capabilities, and sharing options

Community library showcasing shared prompts from users with browsing and filtering options

Basic prompt generation interface with text input and simple generation options

Advanced generation interface with detailed parameters, image modes, and custom settings

AI-powered prompt refinement interface with multiple sources and quick actions

User profile page with account settings, API key management, and prompt statistics