CVE-2026-105745

Description

A flaw was found in Docling. Docling imports registered plugin entry points before evaluating whether external plugins are permitted. A local attacker or compromised package installed in the environment can exploit this flaw to achieve arbitrary code execution during startup, even when external plugins are explicitly disabled.

Statement

A moderate-impact vulnerability was found in Docling, which is shipped in Red Hat OpenShift AI and Lightspeed Core products. Docling plugin factories in docling/models/factories/base_factory.py call load_setuptools_entrypoints() before applying the allow_external_plugins setting, so every module registered in the Docling entry-point group is imported during startup even when external plugins are explicitly disabled. A local attacker, or an attacker able to place a compromised third-party package into the Docling Python environment, can exploit this to execute arbitrary code at import time when Docling starts. Code execution is confined to the privileges of the process running Docling, and exploitation depends on an untrusted entry point already being present in the environment. Red Hat rates this flaw as Moderate severity (CVSS 6.7), lower than the NIST score of 7.8 (Important), because exploitation requires local access with low privileges to install a malicious entry point and depends on specific preconditions, resulting in higher attack complexity. Because the impact is Moderate with a CVSS score below 7.0, Red Hat has deferred the fix for the affected products rather than filing fix trackers at this time. The residual risk is limited to environments where an untrusted or compromised package can be installed into the same Python environment as Docling.

Mitigation

Mitigation for this issue is either not available or the currently available options do not meet the Red Hat Product Security criteria comprising ease of use and deployment, applicability to widespread installation base, or stability. The allow_external_plugins setting cannot be relied upon as a workaround in affected versions because Docling imports registered entry points before that setting is applied. The upstream project has addressed this issue in Docling 2.131.0; updates for affected Red Hat products should be applied when they become available.

Common Vulnerability Scoring System (CVSS) Score Details

Info alert:Important note

CVSS scores for open source components depend on vendor-specific factors (e.g. version or build chain). Therefore, Red Hat's score and impact rating can be different from NVD and other vendors. Red Hat remains the authoritative CVE Naming Authority (CNA) source for its products and services (see Red Hat classifications).

The following CVSS metrics and score provided are preliminary and subject to review.

CVSS v3 Score Breakdown

Red HatNVDcve.org
Base Score6.77.86.7
Attack VectorLocalLocalLocal
Attack ComplexityHighLowHigh
Privileges RequiredLowNoneLow
User InteractionRequiredRequiredRequired
ScopeUnchangedUnchangedUnchanged
ConfidentialityHighHighHigh
Integrity ImpactHighHighHigh
Availability ImpactHighHighHigh

Vector

Red Hat: CVSS:3.1/AV:L/AC:H/PR:L/UI:R/S:U/C:H/I:H/A:H

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

cve.org: CVSS:3.1/AV:L/AC:H/PR:L/UI:R/S:U/C:H/I:H/A:H

Understanding the Weakness (CWE)

Confidentiality,Integrity,Availability

Technical Impact: Execute Unauthorized Code or Commands

An attacker could insert malicious functionality into the program by causing the program to download code that the attacker has placed into the untrusted control sphere, such as a malicious web site. This could enable the injection of malware, information exposure by granting excessive privileges or permissions to the untrusted functionality, DOM-based XSS vulnerabilities, stealing user's cookies, open redirect to malware (CWE-601), etc.

Frequently Asked Questions

Want to get errata notifications? Sign up here.