CVE-2026-108958
Received Received - Intake

TensorLayer HDF5 Deserialization Vulnerability

Publication date: 2026-10-13

Last updated on: 2026-10-13

Assigner: VulDB

Description
A weakness has been identified in TensorLayer up to 2.2.4. This affects the function load_hdf5_graph of the file tensorlayer/files/utils.py of the component HDF5 Loading. Executing a manipulation of the argument model_config can lead to deserialization. The attack can be executed remotely. The exploit has been made available to the public and could be used for attacks. The project was informed of the problem early through an issue report but has not responded yet.

CVSS Scores

EPSS Scores

Probability:
Percentile:

Meta Information

Published
2026-10-13
Last Modified
2026-10-13
Generated
2026-10-13
AI Q&A
2026-10-13
EPSS Evaluated
N/A
NVD
CVE-2026-108958
EUVD
EUVD-2026-98267

Affected Vendors & Products

Showing 5 associated CPEs
Vendor Product Version / Range
n/a TensorLayer 2.2.0
n/a TensorLayer 2.2.1
n/a TensorLayer 2.2.2
n/a TensorLayer 2.2.3
n/a TensorLayer 2.2.4

Helpful Resources

Exploitability

CWE
CWE Icon
KEV
KEV Icon
CWE ID Description
CWE-20 The product receives input or data, but it does not validate or incorrectly validates that the input has the properties that are required to process the data safely and correctly.
CWE-502 The product deserializes untrusted data without sufficiently ensuring that the resulting data will be valid.

Attack-Flow Graph

AI Quick Actions

Instant insights powered by AI
Executive Summary

This vulnerability is an arbitrary code execution flaw in TensorLayer up to version 2.2.4. It occurs in the load_hdf5_graph function of tensorlayer/files/utils.py when loading HDF5 model files. The issue involves unsafe use of Python's eval() function on the model_config attribute from the HDF5 file without proper validation. An attacker can craft a malicious HDF5 file with a payload in model_config that executes during model loading.

Detection Guidance

Detecting this vulnerability requires checking for suspicious HDF5 model files or unusual activity during model loading. Monitor for files with model_config attributes containing executable code. Review logs for unexpected file writes or process executions during TensorLayer model loading. Use static analysis tools to scan Python code for eval(model_config_str) in tensorlayer/files/utils.py.

Impact Analysis

This vulnerability allows remote attackers to execute arbitrary code on your system with the privileges of the process loading the model. If you load an untrusted HDF5 model file, the attacker's payload will run silently while still restoring a valid model, making detection difficult. The exploit could lead to data theft, system compromise, or further network infiltration.

Compliance Impact

This vulnerability could lead to violations of GDPR and HIPAA by enabling unauthorized access to personal or health data. GDPR requires protection of personal data and timely breach notification, while HIPAA mandates safeguards for protected health information. A successful exploit could result in data breaches that trigger regulatory penalties and reputational damage.

Mitigation Strategies

Immediately update TensorLayer to a patched version replacing eval() with ast.literal_eval() in load_hdf5_graph(). Avoid loading untrusted HDF5 model files. If using the master branch, apply the suggested fix manually. Disable cloudpickle.loads() usage unless explicitly required. Monitor for any signs of exploitation in system logs.

Chat Assistant

Ask questions about this CVE
Hi! I’m here to help you understand CVE-2026-108958. Ask me anything about the vulnerability, its impact, or mitigation strategies.
0/70

EPSS Chart