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 PoC — numpy.distutils pickle gadget (scanner denylist gap)

PoC for a pickle/joblib scanner denylist coverage gap, not a real model.

evil_numpy_distutils.joblib uses numpy.distutils.cpuinfo.getoutput (→ /bin/sh -c) as a __reduce__ gadget. It achieves arbitrary command execution on joblib.load() while picklescan, modelscan, and modelaudit (ProtectAI's current scanner) all fail to flag it — they denylist numpy.f2py.crackfortran but not the numpy.distutils.* command-exec family.

Harmless payload

Runs only id > /tmp/H004_PWNED (writes a marker). No damage / network / exfiltration. Reachability: numpy < 2.3 (numpy.distutils removed in numpy 2.3); confirmed on numpy 1.26.4.

This repo is deleted after triage.

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