We will be undergoing planned maintenance on January 16th, 2026 at 1:00pm UTC. Please make sure to save your work.

CipherMuse

“Whisper to the shadows, let AI speak.” CipherMuse hides messages inside AI art and music secure, invisible, and untraceable.

Comment

CipherMuse - Kiroween Hackathon Submission

Inspiration

The idea for CipherMuse came from discovering the Kiroween hackathon's "Frankenstein" category and realizing I could combine my interests in AI, cryptography, and Halloween aesthetics. The concept of hiding secrets in plain sight - steganography - has always fascinated me.

The inspiration was to create something that stitches together modern AI technology with classical cryptography, wrapped in a Halloween theme that makes these complex concepts more approachable and engaging.

What it does

CipherMuse is an AI-powered steganography platform with Halloween theming that allows users to:

Generate Halloween Content: Create spooky images using Stable Diffusion v1.5 with four distinct themes (Spooky, Gothic, Witch, Zombie) and atmospheric audio using MusicGen-Small

Hide Secret Messages: Encrypt messages with AES-256 and embed them invisibly into generated content using LSB steganography techniques

Secure Authentication: Register and login with encrypted password storage and session management

Personal Galleries: Store and manage encrypted files in user-specific galleries

Multi-format Support: Work with various file formats including PNG, JPG, BMP for images and WAV, MP3, M4A for audio

Extract Hidden Messages: Decrypt and retrieve secret messages from steganographic files

How we built it

Planning with Kiro

I used Kiro's spec-driven development approach to plan the project architecture. I created comprehensive specifications for UI consistency, defining requirements, design patterns, and 8 correctness properties for maintaining consistent styling across the application.

Project Structure

Backend: Python Flask application in the backend/ directory

Frontend: React application in the src/ directory with components, pages, and styling

AI Models: Integration with Stable Diffusion v1.5, MusicGen-Small, and SwinIR for content generation and enhancement

Styling System: CSS variables with dark orange theme colors (#D97642, #C96A3D, #B85E38)

Kiro Integration

Created comprehensive UI specifications with detailed requirements and design architecture. Defined property-based testing strategy with 8 correctness properties. Used Kiro's conversational coding for UI component development. Implemented centralized CSS variable system for maintainable theming across all components.

Challenges we ran into

UI Consistency Challenge

The main challenge I focused on was maintaining consistent button colors and styling across all application pages. This led me to explore Kiro's spec-driven development approach and create detailed specifications for UI consistency.

Learning Kiro's Methodology

Adopting Kiro's spec-driven development was a learning curve. Understanding how to define correctness properties for UI elements and structure comprehensive specifications required research and experimentation.

Halloween Theme Design

Balancing the Halloween aesthetic with functional usability was challenging. The dark orange color scheme needed to be atmospheric while maintaining good contrast and accessibility standards.

Architecture Planning

Designing a system that could theoretically integrate AI generation, steganography, and encryption while maintaining clean separation of concerns required careful architectural thinking.

Technical Complexity

Understanding the theoretical integration of multiple technologies - AI models, cryptographic libraries, and steganography algorithms - presented a significant learning challenge.

Accomplishments that we're proud of

Kiro Spec-Driven Development: Successfully created comprehensive UI specifications with 8 correctness properties for button consistency

Project Architecture: Designed a cohesive system that combines AI, cryptography, and web technologies in the "Frankenstein" spirit

CSS Variable System: Implemented centralized theming with dark orange Halloween colors for maintainable styling

Comprehensive Planning: Created detailed specifications covering requirements, design patterns, and testing strategies

Halloween Theme Integration: Built an atmospheric Halloween aesthetic that enhances the "hidden secrets" concept

Learning Achievement: Successfully adopted Kiro's methodology for systematic development and UI consistency

What we learned

Kiro Development Methodology

Spec-Driven Development: Learned to create comprehensive specifications before implementation, including detailed requirements and design architecture

Property-Based Testing: Discovered how to define correctness properties for UI consistency and systematic validation

CSS Architecture: Understood centralized theming with CSS variables for maintainable design systems

Technical Implementation

AI Model Integration: Successfully integrated multiple AI models (Stable Diffusion, MusicGen, SwinIR) into the web application

Steganography Implementation: Built LSB (Least Significant Bit) steganography with pseudo-random positioning for secure message embedding

Cryptography Integration: Implemented AES-256 encryption with secure key derivation for message protection

Full-Stack Development: Built complete Flask backend with React frontend integration

Project Planning

Systematic Approach: The value of upfront specification and architectural planning

Testing Strategy: How property-based testing can create more robust applications

User Experience: Balancing complex technical concepts with intuitive design

Documentation: The importance of comprehensive specifications for maintainable projects

What's next for CipherMuse

Enhancement Phase

Performance Optimization: Improve AI model loading times and memory usage

Advanced Features: Add batch processing for multiple files and enhanced encryption options

Testing Expansion: Implement the full property-based testing strategy defined in the specifications

User Experience: Refine the interface based on user feedback and usability testing

Feature Expansion

Additional AI Models: Explore integration with other AI generation models

Enhanced Encryption: Research other powerful cryptography algorithms for future-proofing

Mobile Experience: Optimize the interface for mobile devices

Performance Optimization: Implement efficient model loading and caching strategies

The systematic approach learned through Kiro's methodology provides a solid foundation for these future developments.

Share this project:

Updates