Remote Code Execution Vulnerability in InstructLab by Red Hat
CVE-2026-6859

8.8HIGH

Key Information:

Vendor

Red Hat
Status
Red Hat Enterprise Linux Ai (rhel Ai) 3
Vendor
CVE Published:
22 April 2026

What is CVE-2026-6859?

A vulnerability in InstructLab allows remote attackers to execute arbitrary Python code by exploiting the hardcoded setting of 'trust_remote_code=True' in the linux_train.py script. This flaw can be triggered when a user runs 'ilab train/download/generate' with a malicious model sourced from the HuggingFace Hub, potentially leading to a total compromise of the system.

References

https://access.redhat.com/security/cve/CVE-2026-6859
vdb-entryx_refsource_REDHAT
https://bugzilla.redhat.com/show_bug.cgi?id=2459998
issue-trackingx_refsource_REDHAT

CVSS V3.1

Score:
8.8
Severity:
HIGH
Confidentiality:
High
Integrity:
High
Availability:
High
Attack Vector:
Network
Attack Complexity:
Low
Privileges Required:
None
User Interaction:
Required
Scope:
Unchanged

Timeline

  • Vulnerability published

  • Vulnerability Reserved

Credit

Red Hat would like to thank Martin Brodeur (independent security researcher) for reporting this issue.
.