# CVE-2025-71366: picklescan security bypass via torch.utils.bottleneck in pickle files

Source: https://junglewise.ai/threats/cve-2025-71366-picklescan-security-bypass-via-torch-utils-bottleneck-in-pickle  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-07-04. Updated: 2026-07-04  
License: CC BY 4.0 with a link back

CVE-2025-71366 · Severity: high · CVSS 8.1 · Published 2026-07-04

picklescan is a security tool used to scan Python pickle files and machine learning models for malicious code. A vulnerability exists where the tool fails to detect certain dangerous commands hidden within these files. If a user relies on an affected version of picklescan to verify a file and then opens it, an attacker could gain full control over the user's system.

Technologies: [picklescan (PyPI)](https://junglewise.ai/threats/technologies/picklescan). Vendors: [PyPI](https://junglewise.ai/threats/vendors/pypi).

## Quick answers

**Is CVE-2025-71366 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2025-71366 being exploited in the wild.

**How severe is CVE-2025-71366?** High severity, with a CVSS score of 8.1, published 4 July 2026.

**Which products and versions are affected?** picklescan (PyPI) ECOSYSTEM:introduced=0,fixed=0.0.28.

**Is there a fix for CVE-2025-71366?** Yes. picklescan (PyPI): upgrade to 0.0.28 or later. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.

**How is CVE-2025-71366 attacked?** It is attacked over the network, in a software library.

**Am I affected by CVE-2025-71366?** Check whether you run picklescan (PyPI) at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

## How to fix

- picklescan (PyPI): upgrade to 0.0.28 or later

  ```sh
  pip install "picklescan>=0.0.28"
  ```


## Technical details

A deserialization vulnerability (CWE-502) exists in picklescan versions prior to 0.0.28. The scanner's blocklist/allowlist logic fails to identify the 'torch.utils.bottleneck.__main__.run_cprofile' function as a dangerous global. An attacker can craft a malicious pickle file using this function in a '__reduce__' method to execute arbitrary Python code. When a victim scans the file, picklescan reports it as safe; subsequent loading of the file via 'pickle.load()' results in arbitrary code execution. This issue is resolved in version 0.0.28.

## Affected products

- picklescan picklescan < 0.0.28

## Identifiers

- CVE-2025-71366
- GHSA-4r9r-ch6f-vxmx

## Timeline

- 2025-08-22: advisory: GitHub Security Advisory published
- 2026-07-03: disclosed: CVE published to NVD

## References

- [https://github.com/mmaitre314/picklescan/security/advisories/GHSA-4r9r-ch6f-vxmx](https://github.com/mmaitre314/picklescan/security/advisories/GHSA-4r9r-ch6f-vxmx)
- [https://www.vulncheck.com/advisories/picklescan-arbitrary-code-execution-via-torch-utils-bottleneck-main-run-cprofile](https://www.vulncheck.com/advisories/picklescan-arbitrary-code-execution-via-torch-utils-bottleneck-main-run-cprofile)

## Related threats

- [picklescan detection bypass via _operator.methodcaller](https://junglewise.ai/threats/cve-2025-71375-picklescan-detection-bypass-via-operator-methodcaller)
- [picklescan security bypass via operator.methodcaller](https://junglewise.ai/threats/cve-2025-71373-picklescan-security-bypass-via-operator-methodcaller)
- [Picklescan detection bypass via numpy getlincoef gadget](https://junglewise.ai/threats/cve-2025-71372-picklescan-detection-bypass-via-numpy-getlincoef-gadget)
- [picklescan safety check bypass via PyTorch decoder basichandlers](https://junglewise.ai/threats/cve-2025-71369-picklescan-safety-check-bypass-via-pytorch-decoder-basichandlers)
- [picklescan security bypass via _operator.attrgetter function call](https://junglewise.ai/threats/cve-2025-71367-picklescan-security-bypass-via-operator-attrgetter-function-call)