Description (CNA)
A vulnerability in the `TFSMLayer` class of the `keras` package, version 3.13.0, allows attacker-controlled TensorFlow SavedModels to be loaded during deserialization of `.keras` models, even when `safe_mode=True`. This bypasses the security guarantees of `safe_mode` and enables arbitrary attacker-controlled code execution during model inference under the victim's privileges. The issue arises due to the unconditional loading of external SavedModels, serialization of attacker-controlled file paths, and the lack of validation in the `from_config()` method.
Affected (CNA)
| Vendor / product | Versions |
|---|---|
| keras-team / keras-team/keras | affected unspecified before 3.13.2 |
CWE
CWE-502
References
- https://huntr.com/bounties/7e78d6f1-6977-4300-b595-e81bdbda331c
- https://github.com/keras-team/keras/commit/b6773d3decaef1b05d8e794458e148cb362f163f
ADP enrichment
No English description supplied.
No references supplied.
A flaw was found in the `keras` package. This vulnerability allows an attacker to execute unauthorized code on a victim's system. It occurs when a victim loads a specially crafted `.keras` model, even if the `safe_mode` security feature is active. The issue arises because the `keras` package can unconditionally load external TensorFlow SavedModels without sufficient validation, thereby bypassing the intended security protections and leading to arbitrary code execution.