CVE-2021-37678

mediumPriorität · beachten · 45EPSS 0.004 · 39. PerzentilExploit · PoCEUUS
grimoire priorityP1 · Urgent
Exploitable?
Public exploit
1 exploit · PoC
Likely / severe?
0.45% (39th) EPSS
CVSS 4.6 · medium
Is it mine?
22 products
Prüfe deine Angriffsfläche
Who says so?
4 sources
cnnvd · enisa_euvd · osv · nvd

EPSS · bewertet 2026-10-27

TensorFlow is an end-to-end open source platform for machine learning. In affected versions TensorFlow and Keras can be tricked to perform arbitrary code execution when deserializing a Keras model from YAML format. The [implementation](https://github.com/tensorflow/tensorflow/blob/460e000de3a83278fb00b61a16d161b1964f15f4/tensorflow/python/keras/saving/model_config.py#L66-L104) uses `yaml.unsafe_load` which can perform arbitrary code execution on the input. Given that YAML format support requires a significant amount of work, we have removed it for now. We have patched the issue in GitHub commit 23d6383eb6c14084a8fc3bdf164043b974818012. The fix will be included in TensorFlow 2.6.0. We will also cherrypick this commit on TensorFlow 2.5.1, TensorFlow 2.4.3, and TensorFlow 2.3.4, as these are also affected and still in supported range.

01

cvss

Technischer Schweregrad, 0–10. Eine Zeile je Quelle.

9.3
010

Basiswert 0–10 · höher = schwerwiegender · Vektor überfahren zum Auflösen⚠ Quellen uneinig · medium–critical · 4.6–9.3

v2.04.6mediumAV:L/AC:L/Au:N/C:P/I:P/A:Pprimary · NVD / NIST USAusnutzbarkeit 3.9 · Auswirkung 6.4
v3.18.8highCVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:Hprimary · NVD / NIST USAusnutzbarkeit 2 · Auswirkung 6
v3.19.3criticalCVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:Hsecondary · ENISA EU
v3.19.3criticalCVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:Hsecondary · security-advisories@github.comAusnutzbarkeit 2.5 · Auswirkung 6
v3.19.3criticalCVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:Hsecondary · OSV
v3.18.3highCVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H/E:Psecondary · CVSS-BT (t0sche)
02

schwächen

Die CWE-Schwächeklassen, zu denen dies gehört.

CWE-502USCWE-502US
03

Exploits (1)

Öffentlicher Proof-of-Concept- und Exploit-Code.

04

betroffene Produkte (22)

Betroffene Produkte und Versionsbereiche.

20 von 22 angezeigt
05

Advisories (5)

Hersteller- und Ökosystem-Meldungen dazu.

  • GHSA-r6jx-9g48-2r5rArbitrary code execution due to YAML deserialization2021-08-25ghsa · osvUS

    ### Impact TensorFlow and Keras can be tricked to perform arbitrary code execution when deserializing a Keras model from YAML format. ```python from tensorflow.keras import models payload = ''' !!python/object/new:type args: ['z', !!python/tuple [], {'extend': !!python/name:exec }] listitems: "__import__('os').system('cat /etc/passwd')" ''' models.model_from_yaml(payload) ``` The [implementation](https://github.com/tensorflow/tensorflow/blob/460e000de3a83278fb00b61a16d161b1964f15f4/tensorflow/python/keras/saving/model_config.py#L66-L104) uses `yaml.unsafe_load` which can perform arbitrary code execution on the input. ### Patches Given that YAML format support requires a significant amount of work, we have removed it for now. We have patched the issue in GitHub commit [23d6383eb6c14084a8fc3bdf164043b974818012](https://github.com/tensorflow/tensorflow/commit/23d6383eb6c14084a8fc3bdf164043b974818012). The fix will be included in TensorFlow 2.6.0. We will also cherrypick this commit on TensorFlow 2.5.1, TensorFlow 2.4.3, and TensorFlow 2.3.4, as these are also affected and still in supported range. ### For more information Please consult [our security guide](https://github.com/tensorflow/tensorflow/blob/master/SECURITY.md) for more information regarding the security model and how to contact us with issues and questions. ### Attribution This vulnerability has been reported by Arjun Shibu.

  • BIT-tensorflow-2021-37678Arbitrary code execution due to YAML deserialization2024-03-06vendor · osvUS

    TensorFlow is an end-to-end open source platform for machine learning. In affected versions TensorFlow and Keras can be tricked to perform arbitrary code execution when deserializing a Keras model from YAML format. The [implementation](https://github.com/tensorflow/tensorflow/blob/460e000de3a83278fb00b61a16d161b1964f15f4/tensorflow/python/keras/saving/model_config.py#L66-L104) uses `yaml.unsafe_load` which can perform arbitrary code execution on the input. Given that YAML format support requires a significant amount of work, we have removed it for now. We have patched the issue in GitHub commit 23d6383eb6c14084a8fc3bdf164043b974818012. The fix will be included in TensorFlow 2.6.0. We will also cherrypick this commit on TensorFlow 2.5.1, TensorFlow 2.4.3, and TensorFlow 2.3.4, as these are also affected and still in supported range.

  • PYSEC-2021-3002021-08-12vendor · osvUS

    TensorFlow is an end-to-end open source platform for machine learning. In affected versions TensorFlow and Keras can be tricked to perform arbitrary code execution when deserializing a Keras model from YAML format. The [implementation](https://github.com/tensorflow/tensorflow/blob/460e000de3a83278fb00b61a16d161b1964f15f4/tensorflow/python/keras/saving/model_config.py#L66-L104) uses `yaml.unsafe_load` which can perform arbitrary code execution on the input. Given that YAML format support requires a significant amount of work, we have removed it for now. We have patched the issue in GitHub commit 23d6383eb6c14084a8fc3bdf164043b974818012. The fix will be included in TensorFlow 2.6.0. We will also cherrypick this commit on TensorFlow 2.5.1, TensorFlow 2.4.3, and TensorFlow 2.3.4, as these are also affected and still in supported range.

  • PYSEC-2021-5912021-08-12vendor · osvUS

    TensorFlow is an end-to-end open source platform for machine learning. In affected versions TensorFlow and Keras can be tricked to perform arbitrary code execution when deserializing a Keras model from YAML format. The [implementation](https://github.com/tensorflow/tensorflow/blob/460e000de3a83278fb00b61a16d161b1964f15f4/tensorflow/python/keras/saving/model_config.py#L66-L104) uses `yaml.unsafe_load` which can perform arbitrary code execution on the input. Given that YAML format support requires a significant amount of work, we have removed it for now. We have patched the issue in GitHub commit 23d6383eb6c14084a8fc3bdf164043b974818012. The fix will be included in TensorFlow 2.6.0. We will also cherrypick this commit on TensorFlow 2.5.1, TensorFlow 2.4.3, and TensorFlow 2.3.4, as these are also affected and still in supported range.

  • PYSEC-2021-7892021-08-12vendor · osvUS

    TensorFlow is an end-to-end open source platform for machine learning. In affected versions TensorFlow and Keras can be tricked to perform arbitrary code execution when deserializing a Keras model from YAML format. The [implementation](https://github.com/tensorflow/tensorflow/blob/460e000de3a83278fb00b61a16d161b1964f15f4/tensorflow/python/keras/saving/model_config.py#L66-L104) uses `yaml.unsafe_load` which can perform arbitrary code execution on the input. Given that YAML format support requires a significant amount of work, we have removed it for now. We have patched the issue in GitHub commit 23d6383eb6c14084a8fc3bdf164043b974818012. The fix will be included in TensorFlow 2.6.0. We will also cherrypick this commit on TensorFlow 2.5.1, TensorFlow 2.4.3, and TensorFlow 2.3.4, as these are also affected and still in supported range.

06

auch bekannt als (11)

Weitere Namen und Kennungen dieser Entität.

07

quellen

Welche Quellen diesen Datensatz belegen und wann.

cnnvdenisa_euvdosvnvdNVDCVE.orgMITRE
veröffentlicht 2021-08-12geändert 2026-06-17
08

Referenzen (8)

Externe Belege — Patches, Meldungen, Analysen.

scry

Angreifer nutzen Schwachstellen wie diese gegen im Internet exponierte Dienste aus. Sehen Sie mit einem kostenlosen Passiv-Scan, was exponiert ist.

Kostenloser Scan