CVE-2026-108945
Received Received - Intake

Code Injection in Ultralytics YOLOv13 AutoBackend

Publication date: 2026-10-13

Last updated on: 2026-10-13

Assigner: VulDB

Description
A vulnerability was detected in iMoonLab yolov13 up to 73289949533efac82bb5f72ec19b746618656bd2. Affected by this issue is the function eval of the file ultralytics/nn/autobackend.py of the component AutoBackend Model Loading. Performing a manipulation of the argument imgsz/names/kpt_shape results in code injection. The attack must be initiated from a local position. The project was informed of the problem early through an issue report but has not responded yet.

CVSS Scores

EPSS Scores

Probability:
Percentile:

Meta Information

Published
2026-10-13
Last Modified
2026-10-13
Generated
2026-10-13
AI Q&A
2026-10-13
EPSS Evaluated
N/A
NVD
CVE-2026-108945
EUVD
EUVD-2026-98204

Affected Vendors & Products

Showing 1 associated CPE
Vendor Product Version / Range
iMoonLab yolov13 73289949533efac82bb5f72ec19b746618656bd2

Helpful Resources

Exploitability

CWE
CWE Icon
KEV
KEV Icon
CWE ID Description
CWE-94 The product constructs all or part of a code segment using externally-influenced input from an upstream component, but it does not neutralize or incorrectly neutralizes special elements that could modify the syntax or behavior of the intended code segment.
CWE-74 The product constructs all or part of a command, data structure, or record using externally-influenced input from an upstream component, but it does not neutralize or incorrectly neutralizes special elements that could modify how it is parsed or interpreted when it is sent to a downstream component.

Attack-Flow Graph

AI Quick Actions

Instant insights powered by AI
Executive Summary

This vulnerability is a code injection flaw in the iMoonLab yolov13 project. It occurs in the AutoBackend model loading component where the eval() function is used to process metadata strings like imgsz, names, and kpt_shape from ONNX model files. An attacker can craft a malicious .onnx file or metadata.yaml sidecar file to execute arbitrary code during model loading, even before inference begins. The issue stems from using eval() instead of safer alternatives like ast.literal_eval().

Detection Guidance

Check for vulnerable versions of iMoonLab yolov13 by inspecting installed packages or repository clones. Look for files like ultralytics/nn/autobackend.py and verify if eval() is used in the AutoBackend.__init__ method. Review .onnx model files and metadata.yaml for suspicious entries in imgsz, names, or kpt_shape fields.

Impact Analysis

This vulnerability allows local attackers to execute arbitrary code on systems running vulnerable versions of iMoonLab yolov13. Since the attack occurs during model loading, it could lead to system compromise, data theft, or unauthorized access. The impact depends on the privileges of the user running the application. If the application runs with elevated privileges, the attacker could gain full control over the affected system.

Compliance Impact

This vulnerability could lead to non-compliance with GDPR and HIPAA by enabling unauthorized access to sensitive data. GDPR requires protecting personal data, and HIPAA mandates safeguards for protected health information. A successful exploit could result in data breaches, violating these regulations and potentially leading to legal penalties, fines, or reputational damage.

Mitigation Strategies

Upgrade to a patched version of ultralytics (8.3.234 or later) if available. Replace eval() with ast.literal_eval() in the AutoBackend.__init__ method. Avoid loading untrusted .onnx files or metadata.yaml files. Implement input validation for metadata fields like imgsz, names, and kpt_shape.

Chat Assistant

Ask questions about this CVE
Hi! I’m here to help you understand CVE-2026-108945. Ask me anything about the vulnerability, its impact, or mitigation strategies.
0/70

EPSS Chart