CriticalCVECVE-2026-90970
Critical GitLab AI Gateway Vulnerability Enables Remote Code Execution Attacks
Cyber Security News·Source published ·1 min read
What happened?
GitLab has released urgent security updates for a critical AI Gateway vulnerability that could allow authenticated attackers to execute commands remotely. Tracked as CVE-2026-90970, the flaw carries a CVSS score of 9.9 and affects self-hosted deployments used to support GitLab Duo AI features. The company released AI Gateway versions 19.2.4, 19.3.2, and 19.4.1 to address the issue. GitLab strongly recommends that customers running affected self-hosted gateways upgrade immediately. It also contac
Key details
- Authenticated attacker can execute arbitrary commands on AI Gateway via crafted flow prompt templates.
- Vulnerability tracked as CVE-2026-90970 with CVSS score 9.9, affects self-hosted AI Gateway deployments.
- Affected versions include 18.1.6 and earlier than 19.2.4, 19.3 branch before 19.3.2, and 19.4 branch before 19.4.1.
- Attack requires valid account with Duo Agent Platform access; no unauthenticated exploitation possible.
- GitLab hosted AI Gateways on GitLab.com, Dedicated, and Self-Managed are already patched and protected.
Affected
- Self-hosted AI Gateway versions 18.1.6 and earlier than 19.2.4, 19.3 branch before 19.3.2, and 19.4 branch before 19.4.1.
Fixes and mitigations
- For Docker: stop and remove existing container, pull and run new image with correct environment variables, verify image digest and run health checks.
- For Kubernetes/Helm: use image digests or appropriate pull policy to avoid cached images preventing update.
Summarised by SecOraLens from Cyber Security News. Always read the original source before acting. How SecOraLens verifies advisories.