https://vulnerability.circl.lu Contains only the most 10 recent entries. http://www.rssboard.org/rss-specification python-feedgen en Fri, 02 Oct 2026 09:10:27 +0000 https://vulnerability.circl.lu/vuln/cve-2025-71370 <p><strong>Affected:</strong> picklescan</p> <p>picklescan before 0.0.28 fails to detect malicious torch.jit.unsupported_tensor_ops.execWrapper function calls embedded in pickle files. Attackers can craft malicious pickle files that bypass picklescan detection and execute arbitrary code when loaded via pickle.load().</p> <p><strong>Affected:</strong> picklescan</p> <p>picklescan before 0.0.28 fails to detect malicious torch.jit.unsupported_tensor_ops.execWrapper function calls embedded in pickle files. Attackers can craft malicious pickle files that bypass picklescan detection and execute arbitrary code when loaded via pickle.load().</p> https://vulnerability.circl.lu/vuln/cve-2025-71370 https://vulnerability.circl.lu/vuln/ghsa-vr7h-p6mm-wpmh <p><strong>Affected:</strong> PyPI: picklescan</p> <p>### Summary</p> <p>Using torch.jit.unsupported_tensor_ops.execWrapper function, which is a pytorch library function to execute remote pickle file.</p> <p>### Details</p> <p>The attack payload executes in the following steps:</p> <p>First, the attacker craft the payload by calling to torch.jit.unsupported_tensor_ops.execWrapper function in reduce method Then when the victim after checking whether the pickle file is safe by using Picklescan library and this library doesn&#39;t dectect any dangerous functions, decide to pickle.load() this malicious pickle file, thus lead to remote code execution.</p> <p>### PoC</p> <p>```</p> <p>import torch.jit.unsupported_tensor_ops as unsupported_tensor_ops</p> <p>class EvilTorchJitUnsupportedTensorOpsExecWrapper: def __reduce__(self): code = &#39;__import__(&#34;os&#34;).system(&#34;whoami&#34;)&#39; glob = {} loc = {} return unsupported_tensor_ops.execWrapper, (code, glob, loc) ```</p> <p>### Impact</p> <p>Who is impacted? Any organization or individual relying on picklescan to detect malicious pickle files inside PyTorch models. What is the impact? Attackers can embed malicious code in pickle file that remains undetected but executes when the pickle file is loaded. Supply Chain Attack: Attackers can distribute infected pickle files across ML models, APIs, or saved Python objects.</p> <p>### Corresponding</p> <p>https://github.com/FredericDT https://github.com/Qhaoduoyu</p> <p><strong>Affected:</strong> PyPI: picklescan</p> <p>### Summary</p> <p>Using torch.jit.unsupported_tensor_ops.execWrapper function, which is a pytorch library function to execute remote pickle file.</p> <p>### Details</p> <p>The attack payload executes in the following steps:</p> <p>First, the attacker craft the payload by calling to torch.jit.unsupported_tensor_ops.execWrapper function in reduce method Then when the victim after checking whether the pickle file is safe by using Picklescan library and this library doesn&#39;t dectect any dangerous functions, decide to pickle.load() this malicious pickle file, thus lead to remote code execution.</p> <p>### PoC</p> <p>```</p> <p>import torch.jit.unsupported_tensor_ops as unsupported_tensor_ops</p> <p>class EvilTorchJitUnsupportedTensorOpsExecWrapper: def __reduce__(self): code = &#39;__import__(&#34;os&#34;).system(&#34;whoami&#34;)&#39; glob = {} loc = {} return unsupported_tensor_ops.execWrapper, (code, glob, loc) ```</p> <p>### Impact</p> <p>Who is impacted? Any organization or individual relying on picklescan to detect malicious pickle files inside PyTorch models. What is the impact? Attackers can embed malicious code in pickle file that remains undetected but executes when the pickle file is loaded. Supply Chain Attack: Attackers can distribute infected pickle files across ML models, APIs, or saved Python objects.</p> <p>### Corresponding</p> <p>https://github.com/FredericDT https://github.com/Qhaoduoyu</p> https://vulnerability.circl.lu/vuln/ghsa-vr7h-p6mm-wpmh