HashTrace is a next-generation hash identification system that uses:
✔ Advanced cryptographic feature extraction
✔ Full MCF support ($argon2id$, $2a$, $5$, $6$, etc.)
✔ Machine Learning classifiers
✔ Automatic model comparison & selection
✔ Beautiful CLI with gradients
✔ Dataset generator for training custom ML models
The AI model is still in development mode. Current ML accuracy ranges between 60% to 65%. Accuracy is expected to increase with the expansion of the training dataset and refinement of generated synthetic data.
Extracts 100+ features from hash strings:
- Structural patterns
- Byte-level statistics
- Bit distribution
- Encoding detection
- Cryptographic randomness tests
- MCF parsing (Argon2, bcrypt, scrypt, SHA512-crypt, etc.)
- Algorithm-specific fingerprinting
- And others
Automatically trains and selects the best ML model:
- RandomForest
- Logistic Regression
- Decision Tree
Evaluates:
- Accuracy
- F1 score
- Cross-validation
- Training time
Includes:
- Neon gradient logo
- Animated progress bars
- Color-coded ML stats
- Error explanations
pip install -r requirements.txt
python hashtrace_ml.py --help
python hashtrace_ml.py --file <File name / path>
Supports:
$2a$,$2b$,$2y$→ bcrypt$argon2i$,$argon2d$,$argon2id$→ Argon2$1$→ MD5-crypt$5$→ SHA256-crypt$6$→ SHA512-crypt$scrypt$→ scrypt$y$→ yescrypt$8$,$9$→ Cisco$pbkdf2-sha256$→ PBKDF2-SHA256
Enter Hash:
$argon2id$v=19$m=65536,t=3,p=4$1YlQ...$G0PeE...
--------------------------------------------------
Possible Hash Types:
[+] Argon2id
AI Prediction:
Argon2id (99.3% confidence)
--------------------------------------------------
MIT License
⭐ If you find HashTrace 2.0 useful, please give it a star on GitHub! ⭐
"Identifying the unknown" 🔍