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.
Default status is the baseline for the product, each version can override it (e.g. patched versions marked unaffected).
| Vendor | Product | Default status | Versions | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Hugging Face | Transformers | affected |
|
No data.
No data.
No data.
Project Subscriptions
No data.
Advisories
No advisories yet.
Fixes
Solution
No solution given by the vendor.
Workaround
No workaround given by the vendor.
References
History
Tue, 01 Sep 2026 13:45:00 +0000
| Type | Values Removed | Values Added |
|---|---|---|
| Description | 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. | |
| Title | Hugging Face Transformers library writes remote code to disk prior to consent check | |
| References |
|
Projects
Sign in to view the affected projects.
MITRE
Status: PUBLISHED
Assigner: certcc
Published:
Updated: 2026-09-01T14:08:14.152Z
Reserved: 2026-08-25T17:29:08.429Z
Link: CVE-2026-80047
Vulnrichment
No data.
NVD
Status : Received
Published: 2026-09-01T14:17:41.943
Modified: 2026-09-01T14:17:41.943
Link: CVE-2026-80047
Redhat
No data.
OpenCVE Enrichment
No data.
Weaknesses
No weakness.