CVE-2025-71356: picklescan - Arbitrary Code Execution via torch.fx.experimental.symbolic_shapes.ShapeEnv.evaluate_guards_expression
Published Jul 4, 2026
·Updated
picklescan before 0.0.28 fails to detect malicious torch.fx.experimental.symbolicshapes.ShapeEnv.evaluateguardsexpression function calls in pickle files. Attackers can embed undetected code in pickle files that executes remote code when loaded by victims.
Affected Software
1 affected component
picklescan<0.0.28
Event History
Jul 4, 2026
CVE Published
via MITRE·01:23 AM
Data Sourced
via MITRE·01:23 AM
DescriptionSeverityWeakness
Data Sourced
via NVD·02:16 AM
DescriptionSeverityWeakness
Frequently Asked Questions
1
What is the severity of CVE-2025-71356?
CVE-2025-71356 has a severity rating of 8.1, indicating it is classified as high risk.
2
How do I fix CVE-2025-71356?
To mitigate CVE-2025-71356, upgrade to picklescan version 0.0.28 or later.
3
What type of vulnerability is CVE-2025-71356?
CVE-2025-71356 is an arbitrary code execution vulnerability found in picklescan.
4
What systems are affected by CVE-2025-71356?
CVE-2025-71356 affects picklescan versions prior to 0.0.28.
5
How does CVE-2025-71356 allow exploitation?
CVE-2025-71356 allows attackers to embed malicious code in pickle files that is executed when these files are loaded.