Which versions of picklescan are affected by CVE-2025-71367?

Picklescan, a tool used to validate untrusted machine learning model files, contains a detection bypass vulnerability in versions prior to 0.0.34 that allows attackers to slip malicious code past its…. A patch is available.

How to fix or mitigate CVE-2025-71367?

Within 24 hours: inventory all systems using picklescan and identify applications processing untrusted model sources. Within 7 days: upgrade all picklescan instances to version 0.0.34 or later; test thoroughly in staging environments first. Within 30 days: audit model files processed with vulnerable picklescan versions for signs of malicious artifacts or unexpected modifications.

picklescan CVE-2025-71367

Q: What is the CVSS score of CVE-2025-71367?

CVE-2025-71367 has a CVSS 4.0 base score of 7.6 (High). CVSS vector: CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:P/VC:H/VI:H/VA:N/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X. EPSS exploitation probability: 0.4%.

EUVD-2025-210421 HIGH

Deserialization of Untrusted Data (CWE-502)

2026-07-04 VulnCheck

GHSA-j2vq-r4j3-cj5m

Deserialization RCE Picklescan

7.6

CVSS 4.0 · Vendor: VulnCheck

Severity by source

Vendor (VulnCheck) PRIMARY

7.6 HIGH

CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:P/VC:H/VI:H/VA:N/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X

vuln.today AI

8.8 HIGH

Malicious file delivered over network (AV:N) with no auth (PR:N) but needs the victim to scan-then-load it (UI:R); successful pickle code execution yields full host compromise (C/I/A:H).

3.1 AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

4.0 AV:N/AC:L/AT:P/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N

Primary rating from Vendor (VulnCheck).

CVSS VectorVendor: VulnCheck

CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:P/VC:H/VI:H/VA:N/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X

Attack Vector

Network

Attack Complexity

Low

Privileges Required

None

User Interaction

Scope

Lifecycle Timeline

Patch available

Jul 04, 2026 - 03:01 EUVD

Analysis Generated

Jul 04, 2026 - 02:18 vuln.today

DescriptionCVE.org

picklescan before 0.0.34 fails to detect _operator.attrgetter function calls in pickle payloads, allowing attackers to bypass security checks. Remote attackers can craft malicious pickle files using _operator.attrgetter in reduce methods to execute arbitrary code when pickle.load() processes the file.

AnalysisAI

Security-scanner detection bypass in picklescan before 0.0.34 lets attackers slip malicious pickle files past its checks by invoking _operator.attrgetter inside a reduce method, so a file the scanner reports as clean still executes arbitrary code when pickle.load() deserializes it. The flaw affects ML/AI supply-chain pipelines that rely on picklescan to vet untrusted model files. No public exploit identified at time of analysis; the issue was reported by VulnCheck and fixed in 0.0.34.

Technical ContextAI

picklescan is a Python static-analysis tool that inspects pickle serialization streams for dangerous global/reduce opcodes before a model is loaded, and is used across the ML ecosystem (e.g., Hugging Face model vetting). The root cause is CWE-502 (Deserialization of Untrusted Data) expressed as an incomplete denylist: picklescan enumerates known dangerous callables but omitted _operator.attrgetter, a stdlib factory that returns a callable retrieving named attributes. Because attrgetter can be chained to resolve arbitrary attributes/callables at unpickling time, an attacker can reach code-execution primitives that the scanner never flagged, defeating the very control meant to make pickle loading safe. The single affected package is identified by cpe:2.3:a:picklescan:picklescan:*:*:*:*:*:*:*:* for all versions prior to 0.0.34.

RemediationAI

Vendor-released patch: upgrade picklescan to version 0.0.34 or later, which adds detection for _operator.attrgetter, then re-scan any model files previously cleared by an older version. Until upgraded, do not treat picklescan output from vulnerable versions as authoritative: avoid loading pickle-format models from untrusted sources, and where possible switch model serialization to a non-executable format such as safetensors, which eliminates the pickle code-execution class entirely (trade-off: requires converting existing artifacts and only covers tensor weights). As a compensating control, run any required pickle loading inside a sandboxed/network-isolated, least-privilege environment so that a bypassed payload cannot reach sensitive resources (trade-off: added pipeline complexity and latency). See GHSA-46h3-79wf-xr6c and the VulnCheck advisory (https://www.vulncheck.com/advisories/picklescan-remote-code-execution-via-operator-attrgetter-detection-bypass) for confirmation.