privacy-prompt-library 1.1.1

A privacy-preserving prompt transformation library that redacts disability mentions while maintaining functional context. Now includes key-based disability descriptions with improved documentation and usage examples.

Privacy-Preserving Prompt Librar## 🚀 Quick Start

Basic Usage - Transform Personal Prompts

from prompt_library import transform_prompt

# Protect your privacy while getting AI help
result = transform_prompt("I'm blind and need coding help")
print(result['output'])
# "I use screen readers and need coding help. Please ensure any visual content 
# includes text descriptions and is compatible with screen readers."

New Feature - Key-Based Descriptions

from prompt_library import transform_by_key, get_supported_keys

# See what accessibility categories are available
print("Available categories:", get_supported_keys())

# Get a detailed 60-word accessibility description
description = transform_by_key("visual-impairment")  
print(description['output'])
# Returns comprehensive visual accessibility needs without medical terms

# Use in your AI prompts
ai_prompt = f"{description['output']} Help me learn web development."

Library Information

from prompt_library import get_library_info
info = get_library_info()
print(f"Version: {info['version']}")
print(f"Features: {', '.join(info['features'])}")
```sive Python library that transforms user prompts to protect disability privacy while maintaining functional context for AI interactions.

## 🎯 Purpose

This library helps users interact with AI models without disclosing specific medical conditions by:
- **Detecting** disability mentions in prompts
- **Redacting** medical/diagnostic terms
- **Adding** functional context for better AI responses
- **Preserving** user privacy and intent

## 🔄 How It Works

Input: "I'm paralyzed and need help finding accessible restaurants" ↓ Output: "I use mobility equipment and need help finding accessible restaurants. I need step-free access to buildings and accessible parking close to entrances."

## 🏗️ Architecture

- **14 Disability Categories** with comprehensive subgroups
- **Pattern Detection Engine** for identifying disability mentions
- **Redaction Engine** for replacing medical terms
- **Context Enrichment** for adding functional needs
- **Privacy Validation** to ensure no medical data leaks

## � Installation

```bash
pip install privacy-prompt-library

�🚀 Quick Start

Python API

from prompt_library import transform_prompt

# Transform a single prompt
result = transform_prompt("I'm blind and need coding help")
print(result['output'])
# "I use screen readers and need coding help. Please ensure any visual content includes text descriptions and is compatible with screen readers."

# Get library information
from prompt_library import get_library_info
info = get_library_info()
print(f"Library version: {info['version']}")

🔑 Key-Based Descriptions (New in v1.1.0)

Generate detailed 60-word accessibility descriptions from simple category keys:

from prompt_library import transform_by_key, get_supported_keys

# Get all available category keys
supported_keys = get_supported_keys()
print("Available keys:", supported_keys)

# Generate a functional description from a key
description = transform_by_key("visual-impairment")
print(description['output'])
# Output: "I have specific visual accessibility needs requiring comprehensive 
# screen reader compatibility, high contrast display options with customizable 
# color schemes, detailed text descriptions for all visual content including 
# images, alternative format documents in accessible formats, large print 
# materials when needed, audio descriptions for multimedia content, tactile 
# feedback options, keyboard navigation support, and accessible navigation 
# systems that work seamlessly with assistive technology."

# Example usage for different needs:
mobility_desc = transform_by_key("physical-disability")
hearing_desc = transform_by_key("hearing-impairment")
speech_desc = transform_by_key("speech-language-communication-and-swallowing-disability")

print(f"Word count: {len(mobility_desc['output'].split())} words")  # Always 60 words

Available Keys:

• visual-impairment - Visual accessibility needs and assistive technology
• hearing-impairment - Hearing accessibility and communication needs
• physical-disability - Mobility and physical accessibility requirements
• speech-language-communication-and-swallowing-disability - Communication support needs
• speech-intellectual-autism-spectrum-disorders - Cognitive and sensory support
• maxillofacial-disabilities - Facial and oral function considerations
• progressive-chronic-disorders - Adaptive and flexible accommodation needs

💡 Usage Examples

Scenario 1: Getting a Functional Description for AI Prompts

from prompt_library import transform_by_key

# Instead of saying "I'm blind", use a functional description
visual_needs = transform_by_key("visual-impairment")
prompt = f"{visual_needs['output']} Can you help me learn Python programming?"

# This gives the AI detailed context about your accessibility needs
# without revealing medical information

Scenario 2: Building Accessibility Profiles

# Create comprehensive accessibility descriptions
keys = ["visual-impairment", "hearing-impairment", "physical-disability"]
accessibility_profile = []

for key in keys:
    description = transform_by_key(key)
    accessibility_profile.append(description['output'])

combined_profile = " ".join(accessibility_profile)
print(f"Complete accessibility profile: {combined_profile}")

Scenario 3: Privacy-Safe Prompt Enhancement

# Transform personal prompts to be privacy-safe
personal_prompt = "I have multiple sclerosis and need help with work accommodations"
safe_prompt = transform_prompt(personal_prompt)

print("Original:", personal_prompt)
print("Safe version:", safe_prompt['output'])
# Result protects medical info while preserving functional needs

Command Line Interface

# Transform a prompt via CLI
privacy-prompt "I have ADHD and need focus strategies"

# Get library information
privacy-prompt --info

# Custom privacy level
privacy-prompt --privacy-level medium "Your prompt here"

Async Support

from prompt_library import PromptLibrary

library = PromptLibrary()
await library.initialize()
result = await library.transform_prompt("I'm autistic and need help with social situations")
print(result['output'])

📋 Categories Supported

1. Physical Disabilities
2. Visual Impairments
3. Hearing Impairments
4. Speech & Language
5. Intellectual Disabilities
6. Learning Disabilities
7. Autism Spectrum
8. Developmental Disabilities
9. Mental Health
10. Emotional & Behavioral
11. Invisible Disabilities
12. Multiple Disabilities
13. Neurological
14. Genetic & Rare Disorders

🔒 Privacy Guarantee

• ✅ No medical terms in output
• ✅ No diagnostic language
• ✅ Functional descriptions only
• ✅ Complete user anonymity

🐍 Python Package Features

This repository now includes a fully-featured Python package with:

• Modern Packaging: Uses pyproject.toml and is available on PyPI
• Async Support: Full async/await compatibility
• CLI Tool: Command-line interface for easy integration
• Type Hints: Complete typing for better development experience
• Testing: Comprehensive test suite with pytest

Python Installation & Usage

# Install from PyPI
pip install privacy-prompt-library

# Basic usage
python -c "from prompt_library import transform_prompt; print(transform_prompt('I have autism and need help'))"

# CLI usage
privacy-prompt --info
privacy-prompt "I'm deaf and need communication help"

Development Setup

# Clone repository
git clone https://github.com/git-markkuria/kanuni-layer-sdk.git
cd kanuni-layer-sdk

# Install in development mode
pip install -e .

# Run tests
pytest tests/

📁 Repository Structure

├── prompt_library/          # Python package
│   ├── core/               # Core processing engines
│   ├── engines/            # Context and redaction engines
│   ├── data/               # JSON data files
│   └── cli.py              # Command-line interface
├── src/                    # JavaScript/Node.js version
├── tests/                  # Python tests
├── pyproject.toml          # Python packaging config
└── package.json            # Node.js config

🤝 Contributing

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.