CVE Watch14 jul 2026

CVE-2026-57516

Ray prior to 2.56.0 contains an unsafe deserialization vulnerability in the WebDataset reader that allows attackers to achieve remote code e

CVSS

8.8

Alto

EPSS

0.5%

p39

KEV

—

Exploit Today

0-100

Publicado: 1 jul 2026 · Última mod.: 14 jul 2026 · CWE-502

EPSS · 30d

0.5%EPSS · 30 días0.5%

2026-07-022026-07-17

Descripción técnica

Ray prior to 2.56.0 contains an unsafe deserialization vulnerability in the WebDataset reader that allows attackers to achieve remote code execution by supplying a malicious tar archive to the read_webdataset() function. The _default_decoder() function in webdataset_datasource.py unconditionally calls pickle.loads() on tar entries with .pkl/.pickle extensions and torch.load() with weights_only=False on .pt/.pth entries, executing arbitrary code inside Ray remote workers on every worker that processes the malicious archive.

Referencias oficiales

CVEs relacionados

CVECVSSEPSSKEVRExplotTítuloVis.

CVE-2021-422379.8 CRÍ

99.9%

KEV—80Sitecore XP Remote Command Execution Vulnerability9d CVE-2026-456598.8 ALT

86.8%

KEV—76Microsoft SharePoint Server Deserialization of Untrusted Data Vulnerability16d CVE-2026-586449.8 CRÍ

70.7%

KEV—71Microsoft SharePoint Deserialization of Untrusted Data Vulnerability1d CVE-2026-125699.8 CRÍ

66.0%

KEV—70PTC Windchill and FlexPLM Improper Input Validation Vulnerability18d CVE-2026-505229.8 CRÍ

97.2%

——29Deserialization of untrusted data in Microsoft Office SharePoint allows an unauthorized attacker to execute code over a network.3d CVE-2025-560059.8 CRÍ

96.7%

——29An undocumented and unsafe feature in the PLY (Python Lex-Yacc) library 3.11 allows Remote Code Execution (RCE) via the `picklefile` parameter in the `yacc()` function. This parameter accepts a `.pkl` file that is deserialized with `pickle.load()` without validation. Because `pickle` allows execution of embedded code via `__reduce__()`, an attacker can achieve code execution by passing a malicious pickle file. The parameter is not mentioned in official documentation or the GitHub repository, yet it is active in the PyPI version. This introduces a stealthy backdoor and persistence risk. NOTE: A third-party states that this vulnerability should be rejected because the proof of concept does not demonstrate arbitrary code execution and fails to complete successfully.3d