CVE-2026-79785
Received Received - Intake

X-AnyLabeling Model Downloader TLS Certificate Verification Bypass

Publication date: 2026-08-25

Last updated on: 2026-08-25

Assigner: VulnCheck

Description
X-AnyLabeling's model downloader disabled TLS certificate verification. download_with_retry in anylabeling/services/auto_labeling/model.py built a context with ssl._create_unverified_context() and passed it to urllib.request.urlopen, so neither the certificate chain nor the hostname was checked on any model download, and models are fetched over HTTPS from the project's release host. Any party positioned to intercept that connection could therefore answer it with content of their own choosing. The response is written to a .part file and moved into place with os.replace, and the only post-download check, safe_check_model, validates the file's format rather than its provenance: no hash or signature is compared against an expected value. For an ONNX target the substituted file passes onnx.checker.check_model and is then used for inference, so the attacker chooses the model that produces the application's annotations. For a .pth or .pt target, which the shipped SAM2 video, YOLOE, UPN and open_vision configurations use, the check worker calls torch.load without weights_only, so a substituted file is unpickled and executes code of the attacker's choosing on PyTorch releases predating the weights_only default.

CVSS Scores

EPSS Scores

Probability:
Percentile:

Meta Information

Published
2026-08-25
Last Modified
2026-08-25
Generated
2026-08-25
AI Q&A
2026-08-25
EPSS Evaluated
N/A
NVD
CVE-2026-79785
EUVD
EUVD-2026-65595

Affected Vendors & Products

Showing 4 associated CPEs
Vendor Product Version / Range
cvhub520 x-anylabeling 4.0.0-beta.9
cvhub520 x-anylabeling to 4.0.0-beta.9 (exc)
anylabeling model_downloader *
pytorch pytorch *

Helpful Resources

Exploitability

CWE
CWE Icon
KEV
KEV Icon
CWE ID Description
CWE-295 The product does not validate, or incorrectly validates, a certificate.

Attack-Flow Graph

AI Quick Actions

Instant insights powered by AI
Executive Summary

This vulnerability involves X-AnyLabeling disabling TLS certificate verification when downloading AI models. The download_with_retry function in model.py used ssl._create_unverified_context(), bypassing certificate checks for HTTPS downloads. This allowed attackers to intercept connections and substitute malicious model files. The only post-download check validated file format, not authenticity, enabling attackers to control model behavior or execute arbitrary code.

Detection Guidance

Check if your X-AnyLabeling version is below 4.0.0-beta.9 by running: pip show x-anylabeling or checking the application version. Inspect model.py for use of ssl._create_unverified_context() in download_with_retry. Monitor network traffic for unencrypted or suspicious model downloads.

Impact Analysis

An attacker could intercept model downloads and replace them with malicious versions. For ONNX models, this allows controlling inference outputs. For PyTorch models, arbitrary code execution is possible on older PyTorch versions. This could lead to data manipulation, system compromise, or unauthorized access depending on how the models are used.

Compliance Impact

This vulnerability could violate compliance requirements for data integrity and security. GDPR requires appropriate technical measures to ensure data security and integrity. HIPAA mandates secure transmission of protected health information. The lack of TLS verification and potential for model substitution could compromise these requirements, potentially leading to regulatory violations.

Mitigation Strategies

Upgrade X-AnyLabeling to version 4.0.0-beta.9 or later. Ensure TLS verification is enforced in model downloads. Review downloaded model files for unexpected modifications or unusual behavior during inference.

Chat Assistant

Ask questions about this CVE
Hi! I’m here to help you understand CVE-2026-79785. Ask me anything about the vulnerability, its impact, or mitigation strategies.
0/70

EPSS Chart