Encrypt. Split. Vanish.
Your secrets hidden inside everyday images — and only you can bring them back.

See how StegoSafe works in 45 seconds:

🧪 Try the web version now: StegoSafe Web Demo →
🧰 Prefer the terminal? You're in the right place.

• AES-256 Encryption: Secrets are encrypted using military-grade AES-256 in CBC mode.
• Quantum-Resistant Security: AES-256 is considered resilient even against quantum computing attacks.
• Shamir's Secret Sharing: Your encryption key is split into 5 pieces — only 3 are needed to recover it.
• Steganography: Secrets are invisibly embedded inside ordinary PNG images, undetectable to the naked eye.
• Threshold Recovery: Lose 2 images? No problem. 3 shares are enough to recover your data.
• 100% Local: All operations happen on your device. No cloud, no leaks.

• Python 3.6 or higher
• Install dependencies:

  pip install -r requirements.txt

Dependencies:

• cryptography
• Pillow
• numpy

1. Clone this repository or download the source code.
2. Install the required Python packages:

   pip install -r requirements.txt

Encrypt and embed a secret into images:

python stegosafe_cli.py embed -i <input_folder> -s "<your_secret_text>" -o <output_folder>

Arguments:

• -i, --input_folder: Directory containing source images (must be PNG files).
• -s, --secret: The text you want to hide securely.
• -o, --output_folder: Directory where steganographic images will be saved.

Example:

python stegosafe_cli.py embed -i test_images -s "This is my hidden message" -o output_images

Recover the hidden secret from steganographic images:

python stegosafe_cli.py recover -i <stego_folder>

Arguments:

• -i, --stego_folder: Directory containing the steganographic images.

Example:

python stegosafe_cli.py recover -i output_images

1. Encrypt your secret with AES-256 and a randomly generated key.
2. Split the key into 5 shares using Shamir's Secret Sharing (threshold: 3 of 5).
3. Embed each share along with the full ciphertext into separate images using LSB steganography.
4. Recover the secret using any 3 valid stego images.

Even if attackers find some images, without the required threshold, your secret remains mathematically protected.

• Use PNGs: Only lossless PNG format is supported to preserve hidden data integrity.
• Threshold Protection: Fewer than 3 images reveal nothing about the secret.
• Imperceptibility: Only least significant bits are modified, making detection by visual inspection or naive analysis very difficult.
• Complete Locality: Your data never leaves your machine.

For updates and improvements, follow this repo — or contribute!

🔒 Protect your future secrets today — and be ready for tomorrow.

Stay tuned for updates. Follow the project for early access.

This tool is intended for lawful, ethical, and personal use only.
The author assumes no responsibility for misuse.

Released under the MIT License.