Rewterz

UNC6783 Targets Organizations Through Social Engineering and Data Extortion – Active IOCs

October 12, 2026
Rewterz

Microsoft Fixes 1,413 Vulnerabilities Across Windows and Enterprise Products

October 14, 2026

Critical GitLab Flaw Could Enable Unauthenticated Code Execution

Severity

High

Analysis Summary

GitLab has released security patches to address a maximum-severity vulnerability, CVE-2026-97229 (CVSS 10.0), affecting its Community Edition (CE) and Enterprise Edition (EE). The flaw, described as an integer overflow vulnerability, could potentially allow an unauthenticated attacker to execute arbitrary code on a GitLab server by exploiting insufficient regular expression validation in a specially crafted CI/CD configuration.

The vulnerability affects the following versions:

  • GitLab 19.2: All versions before 19.2.8
  • GitLab 19.3: All versions before 19.3.5
  • GitLab 19.4: All versions before 19.4.2

GitLab has addressed the issue in versions 19.2.8, 19.3.5, and 19.4.2. Organizations using affected versions are advised to upgrade to the appropriate patched release to mitigate potential exploitation.

The latest security update also resolves three high-severity vulnerabilities:

  • CVE-2026-92883 (CVSS 8.7): A cross-site scripting (XSS) vulnerability that could enable an authenticated attacker to execute arbitrary JavaScript in other users' browser sessions through a specially crafted repository.
  • CVE-2026-16301 (CVSS 8.5): An improper authorization vulnerability that could allow an authenticated user with the Developer role to execute CI/CD pipelines on protected branches.
  • CVE-2026-14814 (CVSS 8.2): A regular expression vulnerability that could enable cache poisoning of raw repository file responses, potentially causing targeted CI/CD pipelines to execute malicious content and exposing protected environment variables.

This disclosure follows recent security incidents involving other critical GitLab vulnerabilities. Earlier in October 2026, GitLab patched CVE-2026-90970 (CVSS 9.9), which could allow an authenticated user with Duo Agent Platform access to escape the AI Gateway prompt template sandbox and execute arbitrary commands. Additionally, CVE-2026-85706 (CVSS 10.0), disclosed in September 2026, reportedly faced in-the-wild probing within hours of its public disclosure.

Organizations should prioritize patching affected GitLab installations, review CI/CD configurations and pipeline permissions, and monitor server logs for suspicious activity.

Impact

  • Buffer Overflow
  • Cross-Site Scripting
  • Arbitrary Code Execution
  • Unauthorized Gain Access

Indicators of Compromise

CVE

  • CVE-2026-97229
  • CVE-2026-92883
  • CVE-2026-16301
  • CVE-2026-14814
  • CVE-2026-90970
  • CVE-2026-85706

Remediation

  • Upgrade GitLab CE/EE to the patched versions 19.2.8, 19.3.5, or 19.4.2, as applicable, to address the identified vulnerabilities.
  • Review CI/CD configurations and validate regular expressions to prevent exploitation through specially crafted patterns.
  • Restrict access to CI/CD pipeline execution and enforce role-based access controls, particularly for protected branches.
  • Review repository permissions and inspect untrusted or suspicious repository content for potential cross-site scripting attempts.
  • Validate cached responses for raw repository files and investigate potential cache poisoning that could affect CI/CD pipeline execution.
  • Protect sensitive CI/CD variables by restricting access to protected environments and limiting variable exposure in pipeline jobs.
  • Monitor GitLab server logs, pipeline execution history, and authentication events for suspicious activity or potential exploitation attempts.
  • Conduct vulnerability assessments of GitLab instances to identify outdated versions and verify that security patches have been successfully applied.
  • Follow official GitLab security advisories for additional updates and apply future security patches promptly.

Reading this advisory was a good start.

Make it a habit.

Rewterz publishes threat advisories ahead of mainstream cybersecurity media, informed by an AI-Native Autonomous SOC that sees regional threat actor activity in real time. Subscribe to receive each new advisory as it publishes, plus a monthly Middle East threat landscape brief drawn from our own SOC telemetry. For teams evaluating their detection coverage, a 30-minute consultation with a senior analyst is also available, at your pace, when you're ready.