Add "AI Security Tools" section and Veritensor

[arsbr]

Hi team!

As the Hugging Face ecosystem grows, Supply Chain Security is becoming a critical part of the MLOps workflow. I would like to propose adding a new section: "AI Security Tools" (or similar), and submitting Veritensor as an entry.

Veritensor is an open-source CLI specifically designed to secure the consumption of Hugging Face models and datasets.

Why it fits:

• Identity Verification: It calculates local file hashes and queries the HF API to ensure the downloaded file matches the official upstream version bit-for-bit (detecting MITM or corruption).
• LFS Validation: Automatically detects broken Git LFS pointers (a common issue when downloading models).
• Malware & Poisoning Detection: Performs static analysis on Pickle/PyTorch models and streams Parquet datasets downloaded from the Hub to detect RCE and malicious URLs.

Repo: https://github.com/arsbr/Veritensor
License: Apache 2.0

Proposed entry for the README.md:

• Veritensor - Security scanner for Hugging Face artifacts. Verifies model hash integrity against the Hub API and detects malware in Pickle/PyTorch files and Parquet datasets.

If you prefer not to create a new section, please feel free to merge this under "Utilities" or "Tools". Thanks for maintaining this awesome list!