A MULTI-FORMAT STEGANOGRAPHY TOOLKIT
- 🎵 Multi-format Support - Works with WAV, MP3, FLAC, M4A, AIFF
- 🔐 Strong Encryption - AES-256 with Argon2id key derivation for embedded files
- 📑 Multiple Files - Hide multiple files in a single carrier
- 🎛️ Quality Control - Choose between audio quality and storage capacity
- 💻 Python CLI & API - Use from command line or integrate into your code
- 🐳 Docker Ready - Containerized deployment available
- 🤖 LLM Integration - Built-in skills system for AI assistants
- 🖥️ Cross-platform Compatibility - MacOS, Linux, Windows
❗ Requirements
- Python 3.13+
- FFmpeg (for audio format conversion)
GitHub Release
Download the latest .whl file from Releases:
pip install git+https://github.com/kariemoorman/ghostbit.git@latestDevelopment Build
Install from source for development or to access the latest features:
git clone https://github.com/kariemoorman/ghostbit.git
cd ghostbit
pip install -e ".[dev]"GH0STB1T CLI provides quick encoding/decoding/analysis operations directly from the terminal.
Encode (Hide files)
ghostbit audio encode -i <audio_filepath> -s <secret_filepath> <secret_filepath> -q {low,normal,high} -o <output_filename>.<desired_format> -p
Calculate Carrier Capacity
ghostbit audio capacity -i <audio_filepath> -q {low,normal,high}
Decode (Extract Files)
ghostbit audio decode -i <audio_filepath> -p
Analyze File
ghostbit audio analyze -i <audio_filepath> -p
GH0STB1T provides a Python API for seamless integration into existing applications and workflows.
Encode (Hide files)
from ghostbit.audiostego import AudioMultiFormatCoder, EncodeMode
# Initialize coder
coder = AudioMultiFormatCoder()
# Encode files
coder.encode_files_multi_format(
carrier_file="music.wav",
secret_files=["document.pdf", "image.jpg"],
output_file="output.wav",
quality_mode=EncodeMode.NORMAL_QUALITY,
password="optional_password"
)# Encoding with Progress Callbacks
from ghostbit.audiostego import AudioMultiFormatCoder
coder = AudioMultiFormatCoder()
# Encoding progress
def on_encode_progress():
print(".", end="", flush=True)
coder.on_encoded_element = on_encode_progress
coder.encode_files_multi_format(
carrier_file="carrier.wav",
secret_files=["secret.pdf"],
output_file="output.wav"
)Calculate Carrier Capacity
from ghostbit.audiostego import AudioMultiFormatCoder, BaseFileInfoItem, EncodeMode
coder = AudioMultiFormatCoder()
wav_file = coder._convert_to_wav("carrier_file.flac")
def get_capacity(wav_file, encode_mode):
base_file = BaseFileInfoItem(
full_path=wav_file,
encode_mode=encode_mode,
wav_head_length=44,
)
return base_file.max_inner_files_size
capacity_bytes = get_capacity(wav_file, EncodeMode.NORMAL_QUALITY)
print(f"Maximum capacity: {capacity_bytes / (1024*1024):.2f} MB")from ghostbit.audiostego import AudioMultiFormatCoder, EncodeMode
import os
coder = AudioMultiFormatCoder()
# Check capacity with different quality modes
carrier = "long_audio.wav"
secret_file = "large_video.mp4"
secret_size = os.path.getsize(secret_file) / (1024 * 1024)
print(f"Secret file size: {secret_size:.2f} MB")
for mode in [EncodeMode.LOW_QUALITY, EncodeMode.NORMAL_QUALITY, EncodeMode.HIGH_QUALITY]:
capacity = get_capacity(carrier, mode) / (1024 * 1024)
fits = "✅ FITS" if capacity >= secret_size else "❌ TOO LARGE"
print(f"{mode.name}: {capacity:.2f} MB capacity - {fits}")Decode (Extract Files)
from ghostbit.audiostego import AudioMultiFormatCoder
# Initialize coder
coder = AudioMultiFormatCoder()
# Decode files
coder.decode_files_multi_format(
encoded_file="output.wav",
output_dir="extracted/",
password="optional_password"
)# Decode with Progress Callbacks
from ghostbit.audiostego import AudioMultiFormatCoder
coder = AudioMultiFormatCoder()
# Decoding progress
def on_decode_progress():
print(".", end="", flush=True)
coder.on_decoded_element = on_decode_progress
coder.decode_files_multi_format(
encoded_file="output.wav",
output_dir="extracted/"
)Password Protection
from ghostbit.audiostego import AudioMultiFormatCoder, KeyRequiredEventArgs
coder = AudioMultiFormatCoder()
# Handle password requests during decoding
def request_password(args: KeyRequiredEventArgs):
password = input(f"Enter password (version {args.h22_version}): ")
if password:
args.key = password
else:
args.cancel = True # Cancel operation
coder.on_key_required = request_password
coder.decode_files_multi_format(
encoded_file="encrypted_output.wav",
output_dir="extracted/"
)# Password-Protected Multiple Files
from ghostbit.audiostego import AudioMultiFormatCoder, EncodeMode
coder = AudioMultiFormatCoder()
# Encode with password
coder.encode_files_multi_format(
carrier_file="music.mp3",
secret_files=[
"report.pdf",
"spreadsheet.xlsx",
"presentation.pptx"
],
output_file="encoded_music.mp3",
quality_mode=EncodeMode.HIGH_QUALITY,
password="SuperSecure123!"
)
print("✅ Multiple files encrypted and hidden!")
# Decode
coder.decode_files_multi_format(
encoded_file="encoded_music.mp3",
output_dir="extracted_files/",
password="SuperSecure123!"
)
print("✅ Files extracted successfully!")GH0STB1T can be deployed using Docker for isolated, reproducible environments.
Initial Setup
- Clone the repository:
git clone https://github.com/kariemoorman/ghostbit.git
cd ghostbit- Create local
inputandoutputdirectories:
These local directories are mapped to Docker container directories, ensuring secure file access.
mkdir input output- Place your files in the
input/directory:
cp /path/to/carrier.wav input/
cp /path/to/secret.pdf input/Build & Run
# Build and start the container
docker-compose up -d ghostbit
# Encode files
docker-compose exec ghostbit ghostbit audio encode -i input/carrier.wav -f /input/secret.pdf -o encoded.wav -p
# Decode files
docker-compose exec ghostbit ghostbit audio decode -i output/encoded.wav -p
# Check capacity
docker-compose exec ghostbit ghostbit audio capacity input/carrier.wav -q high
# Analyze file
docker-compose exec ghostbit ghostbit audio analyze -i output/encoded.wav -pCleanup
# Stop the container
docker-compose stop
# Remove the container
docker-compose down
# Remove container and images
docker-compose down --rmi allGH0STB1T includes a Skills system designed for seamless integration with LLMs and AI assistants.
GH0STB1T provides three specialized skill documents:
- Steganography - Complete usage guide with examples
- Capacity - Capacity planning and optimization strategies
- Troubleshooting - Common issues and solutions
Retrieve Documentation
from ghostbit.audiostego import get_llm_context
# Get complete documentation formatted for LLMs
context = get_llm_context()
# Use in your LLM prompt
prompt = f"""
You are an expert in audio steganography using AudioStego.
{context}
User: How do I hide a 5MB PDF in a 10-minute WAV file with maximum security?
Please provide a complete Python example with security best practices.
"""
# Send prompt to your LLM
# response = your_llm_api(prompt)Load Specific Skills
from ghostbit.audiostego import load_skill
# Load a specific skill
stego_skill = load_skill("steganography")
# Get skill content
print(stego_skill.content)
# Get examples from skill
examples = stego_skill.get_examples()
for example in examples:
print(f"Language: {example['language']}")
print(f"Description: {example['description']}")
print(f"Code:\n{example['code']}\n")
# Get specific section
best_practices = stego_skill.get_section("Best Practices")
print(best_practices)Create a Prompt Template
from ghostbit.audiostego import get_llm_context
# Prepare context
skills_context = get_llm_context()
# Create detailed prompt
prompt = f"""
You are an expert Python developer specializing in audio steganography.
CONTEXT:
{skills_context}
TASK:
The user wants to create a secure file hiding system for sensitive documents.
Requirements:
- Hide multiple PDF files in a single audio carrier
- Use strong encryption with user-provided passwords
- Show progress during encoding/decoding
- Handle errors gracefully
- Verify file integrity after extraction
USER QUESTION: {user_question}
Please provide:
1. Complete working code
2. Security considerations
3. Error handling strategy
4. Usage example
Format your response as:
- Code blocks with explanations
- Security notes
- Example usage
"""
# Send to LLM API
# response = llm_api.generate(prompt)Integrate with Anthropic Claude API
import anthropic
from ghostbit.audiostego import get_llm_context
client = anthropic.Anthropic(api_key="your-api-key")
context = get_llm_context()
message = client.messages.create(
model="claude-3-opus-20240229",
max_tokens=2048,
system=f"You are an AudioStego expert.
{context}",
messages=[
{"role": "user", "content": "Show me how to use AudioStego with error handling"}
]
)
print(message.content[0].text)Integrate with OpenAI API
from openai import OpenAI
from ghostbit.audiostego import get_llm_context
client = OpenAI(api_key="your-api-key")
context = get_llm_context()
response = client.chat.completions.create(
model="gpt-4",
messages=[
{"role": "system", "content": f"You are an AudioStego expert.
{context}"},
{"role": "user", "content": "How do I encode files with maximum security?"}
]
)
print(response.choices[0].message.content)- GitHub Issues: Report bugs or request features
- Discussions: Ask questions and share ideas
- Documentation: Full API reference
Contributions are welcome!
Here's how to get started: CONTRIBUTING.md
GH0STB1T is a free and open source education and research tool. If you use GH0STB1T in your research or project, please cite it as:
@software{audiostego2026,
author = {Karie Moorman},
title = {GH0STB1T: A Multi-format Steganography Toolkit for Python},
year = {2026},
url = {https://github.com/kariemoorman/ghostbit},
version = {0.0.1}
}APA Format:
Moorman, Karie. (2026). GH0STB1T: A Multi-format Steganography Toolkit for Python (Version 0.0.1) [Computer software]. https://github.com/kariemoorman/ghostbit
This project is licensed under the Apache License 2.0 LICENSE.