AxonHash

AxonHash is a research and prototyping cryptographic hash function based on ARX (Add-Rotate-Xor) permutation and Merkle tree construction. It is designed for educational, experimental, and benchmarking purposes.

Features

• ARX Permutation: Modern, efficient mixing using add, rotate, and xor operations.
• Merkle Tree Aggregation: Parallelizable leaf hashing and secure digest aggregation.
• Configurable Block Size: User can set chunk size for flexibility and experimentation.
• Domain Separation: Supports domain tags for protocol separation.
• Extendable Output (XOF): Can produce digests of arbitrary length.
• Profiling and Testing: Includes scripts for benchmarking and avalanche/collision testing.

Usage

Hashing Example

from main import Axon

# Hash a message
h = Axon(b"hello world")
digest = h.digest()           # Raw bytes
digest_hex = h.hexdigest()    # Hex string
print(digest_hex)

Customization

• Chunk Size: Set via the chunks parameter (default: 64 bytes)
• Rounds: Set via the round parameter (default: 4)
• Domain Tag: Set via the domain parameter

h = Axon(b"data", chunks=32, round=8, domain=0x42)

Profiling

Run profiler.py to benchmark different stages of the hash pipeline.

Testing

Run axon_test.py for avalanche, collision, and speed tests:

python axon_test.py

File Structure

• main.py : Main AxonHash implementation (class-based)
• parallel.py : Standalone and parallelized hash utilities
• profiler.py : Profiling and benchmarking script
• axon_test.py : Test suite for correctness and performance
• axon.py : (If present) Shared core functions

Security Notice

AxonHash is experimental and has not undergone professional cryptanalysis. Do not use it for production or security-critical applications.

License

MIT License

Author

@rawbytedev