Threat Pulse 62High
GitLab AI Gateway RCE (CVE-2026-90970, CVSS 9.9) via prompt template sandbox escape
Severity 4PatchedconfirmedAI and software supply chainCloud and SaaSAI-related

A logged-in user with Duo Agent Platform access could craft a custom flow configuration that escapes the prompt template sandbox and runs commands on a self-hosted GitLab AI Gateway. Fixed in 19.2.4, 19.3.2 and 19.4.1; GitLab.com was not affected. Reported through HackerOne; no exploitation was known at disclosure.

What to do: Fixed in AI Gateway 19.2.4, 19.3.2 and 19.4.1.

AI angle: The vulnerable component is the prompt template engine of an AI coding agent gateway.

· 34 days ago · self-hosted GitLab AI Gateway deployments · The Hacker News
SourcesThe Hacker NewseSecurity PlanetSentinelOne
Why this score

Thousands of self-hosted gateways (impact 1), no known exploitation (exploitability 1), general sector (criticality 0), plus the near-maximum CVSS.

Back to the board

Updated Nov 5, 01:45 UTC. Refreshes on its own.

Add a threat

Paste a link to an advisory, an article or an X post. A person reads every submission before it shows; nothing posts on its own.

Share

The picture people see when the link is posted
Post on XSave picture

This picture is what X, iMessage, Slack and Facebook show under the link. It is drawn fresh from the live board: the board and chart pictures update every 10 minutes, an event picture follows that event.