Critical severity · published Jun 30, 2021
CVE-2021-35958
TensorFlow through 2.5.0 allows attackers to overwrite arbitrary files via a crafted archive when tf.keras.utils.get_file is used with extract=True
9.1
CVSS
1.9%
EPSS
No
On KEV
Description
TensorFlow through 2.5.0 allows attackers to overwrite arbitrary files via a crafted archive when tf.keras.utils.get_file is used with extract=True. NOTE: the vendor's position is that tf.keras.utils.get_file is not intended for untrusted archives
CVSS vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:H/A:H
Attack vector
Network
Complexity
Low
Privileges
None
User interaction
None
Confidentiality
None
Integrity
High
Availability
High
Scope
Unchanged
Exploitation
Is it being used?
Exploit prediction (EPSS)1.9% · higher than 79% of CVEs
Exploited in the wildNo evidence yet
CISA KEVNot listed
Ransomware campaign useNot known
Affected products
Which versions
TE
Tensorflow · Google
through 2.5.0 · source: NVD
References
NVD
https://nvd.nist.gov/vuln/detail/CVE-2021-35958
CVE.org
https://www.cve.org/CVERecord?id=CVE-2021-35958
Third Party Advisory
https://docs.python.org/3/library/tarfile.html#tarfile.TarFile.extractall
Third Party Advisory
https://github.com/tensorflow/tensorflow/blob/b8cad4c631096a34461ff8a07840d5f4d123ce32/tensorflow/python/keras/README.md
Third Party Advisory
https://github.com/tensorflow/tensorflow/blob/b8cad4c631096a34461ff8a07840d5f4d123ce32/tensorflow/python/keras/utils/data_utils.py#L137
Third Party Advisory
https://keras.io/api/
Third Party Advisory
https://vuln.ryotak.me/advisories/52
More from Tensorflow
9.9
CVE-2020-15196
In Tensorflow version 2.3.0, the `SparseCountSparseOutput` and `RaggedCountSparseOutput` implementations don't validate that the `weights` tensor has the same shape as the data
9.8
CVE-2023-25668
TensorFlow is an open source platform for machine learning
9.8
CVE-2023-25664
TensorFlow is an open source platform for machine learning
9.8
CVE-2022-41900
TensorFlow is an open source platform for machine learning
9.8
CVE-2022-35939
TensorFlow is an open source platform for machine learning