You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Security Research PoC — ModelScan denylist bypass (joblib)

This repository contains a proof-of-concept model file for a responsible vulnerability disclosure submitted to huntr.com / Protect AI (Model Format Vulnerability program).

It demonstrates that a .joblib model can achieve arbitrary code execution at joblib.load() time while being reported as clean by ModelScan, because the payload uses a stdlib module (cProfile.run) that is not in ModelScan's unsafe_globals denylist.

The payload is benign (writes a marker file / prints a banner). Do not load this file outside an isolated test environment. For research purposes only.

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support