Security researchers are raising concerns after GitLab on Monday issued an out-of-band patch for a critical code injection vulnerability. 

The vulnerability, tracked as CVE-2026-19478, could enable an attacker to remotely modify or delete a public project as well as user data through a Graph QL directive. The flaw has a severity score of 9.4 out of 10.

The flaw was reported through the HackerOne bug bounty program. 

Threat intelligence firm watchTowr warned Tuesday that it was able to reproduce the vulnerability within minutes of the public disclosure. Researchers said an attacker could do significant damage by exploiting this particular flaw. 

“The newly disclosed code injection vulnerability allows an unauthenticated attacker to delete publicly accessible GitLab projects and rewrite their state,” Jake Knott, principal security researcher at watchTowr, said. 

An attacker would be able to engage in a range of malicious activities, including deletion of entire repositories, forging merge records or banning maintainers through a single HTTP request without the need for credentials or user interaction. 

Organizers running internet-facing, self-hosted GitLab instances should immediately upgrade to a patched release, according to Knott. If they can’t upgrade right away, they can restrict unauthenticated access to “api/graphql.” Otherwise, as a final option, they can remove public repository access.  

The GitLab release also addresses a cross-site request forgery vulnerability, tracked as CVE-2026-19650. Under certain conditions, an attacker would be able to execute mutations via GET requests.