Gated security proof-of-concept
This repository is publicly accessible, but you have to accept the conditions to access its files and content.
This repository contains a coordinated-disclosure security proof-of-concept: a crafted HDF5 model/weight file that demonstrates a memory-safety defect in the HDF5 scaleoffset filter reverse path (libhdf5). It contains no user data, no weaponized exploit, and no network activity. Access is restricted to the affected maintainers and the huntr / ProtectAI triage team for verification and remediation only.
Log in or Sign Up to review the conditions and access this model content.
HDF5 scaleoffset filter — crafted-file memory-safety PoC (gated)
This is a security research proof-of-concept submitted to the huntr Model File Formats
program (format hdf5). It contains crafted HDF5 files that trigger a heap out-of-bounds
read in the libhdf5 scaleoffset reverse (decompress) filter on a normal dataset open/read
path (e.g. stock h5py).
The payload only demonstrates the memory-safety defect. There is no exploit weaponization, no user data, and no network callback.
- Class: heap out-of-bounds read via untrusted filter parameters (CWE-125)
- Affected: HDF Group libhdf5 scaleoffset filter reverse path (reproduced on stock h5py 3.14 / libhdf5 1.14.x and ASAN HDF5 2.0.0)
- Disclosure: coordinated via huntr MFF
Full technical details, root cause, ASAN/Valgrind logs, crafted models, and the regenerator script are in the gated files. Please request access if you are triaging this report.
Contact: security researcher mrw0r57 (huntr: ssjcorpsec).