INCIDENT MAP

Browse / Unsafe Model Loading and Supply Chain / CVE-2026-80047

Hugging Face Transformers downloads custom generation code before trust consent

CVE-2026-80047 · 2026-09-01 high

Class
Unsafe Model Loading and Supply Chain (filed by a model that read the advisory)
Technique
Other weaknesses
Affected
transformers (PyPI)
Gets past
Approval or confirmation prompt
CVSS
CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
EPSS
0.1% chance of exploitation in 30 days

How it works, in short

Remote custom generation code is written into the local module cache before the trust prompt is shown, so declined code still persists on disk. (Our summary, written by a model. The source text is below.)

Details

A vulnerability in Hugging Face Transformers (versions 4.49.0, <= 5.8.1) allows remote Python files to be written to local disk without user consent when using GenerativePreTrainedModel.load_custom_generate(). The function fetches and caches a remote module file before performing the required trust_remote_code consent check, inverting the security model enforced by other code-loading paths (such as AutoConfig, AutoModel, and AutoTokenizer). As a result, attacker‑controlled Python code from custom_generate/generate.py is copied into the user’s ~/.cache/huggingface/modules directory even if the user declines the trust prompt. Although execution is correctly gated, the file write is not reversible and can persist across sessions. This can lead to persistent, unauthorized files on disk and stale cache collisions where cached attacker code may later be executed during trusted model loads. The issue stems from an unconditional file write in dynamic_module_utils.py prior to any trust verification.

References

Source: GitHub Advisory Database, used under CC BY 4.0.