GitLab Patches Critical 9.9 AI Gateway Flaw Allowing Command Execution on Self-Hosted Servers
GitLab patched CVE-2026-90970, a CVSS 9.9 command-execution flaw in self-hosted AI Gateway.
GitLab fixed CVE-2026-90970, a CVSS 9.9 flaw in its self-hosted AI Gateway that lets a logged-in Duo Agent Platform user escape a custom-flow prompt-template sandbox and run commands on the gateway. Fixes are in gateway versions 19.2.4, 19.3.2, and 19.4.1. GitLab.com, GitLab Dedicated, and customers using a GitLab-hosted gateway do not need to act. CISA's CVE assessment lists exploitation as none, and the bug is the same CWE-1336 template-engine class as February's CVE-2026-1868.
- CVE-2026-90970 scores CVSS 9.9 and allows gateway command execution.
- Only organizations hosting their own AI Gateway must patch.
- Fixed versions are 19.2.4, 19.3.2, and 19.4.1.
- CISA records exploitation as none and lists no public PoC.
- Issue is CWE-1336, same class as February CVE-2026-1868.
- Vendors
- GitLab
Vulnerabilities mentioned
All| CVE | Vulnerability | CVSS | EPSS | Signals | Published |
|---|---|---|---|---|---|
| CVE-2026-1868 | GitLab has remediated a vulnerability in the Duo Workflow Service component of GitLab AI Gateway affecting all versions of the AI Gateway from 18.1.6, 18.2.6… | 9.9 | <1% | ||
| CVE-2026-90970 | Prompt sandbox escape to RCE in GitLab AI Gateway | 9.9 | 17% | PoC |
From the original article
3 min read at thehackernews.com
A critical flaw in GitLab's AI Gateway could let a logged-in user with Duo Agent Platform access run commands on the gateway under certain conditions, GitLab said in an advisory. The gateway is the service that connects a GitLab instance to AI models, and only organizations that host their own gateway need to act. The flaw is fixed in gateway versions 19.
thehackernews.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.