Critical GitLab Flaw Exploited Shortly After Disclosure

Summary

Threat actors are actively exploiting a critical GitLab vulnerability, CVE-2026-19478, just days after its public disclosure and patching. The flaw allows unauthenticated attackers to modify or delete public projects and user data, with potential implications for supply chain attacks by forging merge records. WatchTowr warns that honeypots have already detected exploitation attempts, urging immediate patching and log monitoring.

Full text

Threat actors started exploiting a critical-severity GitLab vulnerability roughly two days after public disclosure, attack surface management company WatchTowr warns. Tracked as CVE-2026-19478 (CVSS score of 9.4), the code injection defect was patched on August 17, when GitLab warned that it could be exploited remotely without authentication. “GitLab has remediated an issue that under certain conditions could allow an unauthenticated user to remotely modify or delete public projects and user data via a GraphQL directive,” GitLab said. Fixes were rolled out in GitLab Community Edition (CE) and Enterprise Edition (EE) versions 19.2.4, 19.1.6, 19.0.8, and 18.11.11. On August 18, WatchTowr warned that the flaw could be easily reproduced, urging users to update their self-managed instances as soon as possible. As a mitigation, they should restrict unauthenticated access to the /api/graphql endpoint or remove public repository access entirely. “Whilst no public exploit code is available, WatchTowr was able to reproduce the vulnerability within minutes of its disclosure, armed only with the advisory details and patch. AI-enabled attackers are unlikely to be far behind,” the company told SecurityWeek at the time.Advertisement. Scroll to continue reading. On Wednesday, WatchTowr warned that its honeypot network has already caught the first in-the-wild exploitation attempts targeting CVE-2026-19478. “Organizations that haven’t patched yet should hunt through web logs for requests containing ‘@gl_introduced’, and look for signs of probes or attempted exploitation,” the company said. The narrow patch window, fueled by the bug’s severity and the use of AI, is the new reality of vulnerability reproduction and exploitation, WatchTowr principal security researcher Jake Knott points out. “The newly disclosed code injection vulnerability allows an unauthenticated attacker to delete publicly accessible GitLab projects and rewrite their state, deleting repositories entirely, forging merge records, or banning maintainers in a single HTTP request with no credentials, user interaction, or obscure configuration required,” Knott says. According to Mondoo co-founder and CSO Patrick Münch, the fresh GitLab defect could easily fuel the next wave of supply chain attacks, given that it can be exploited to forge any merge records. “Think about what that means. In every supply chain worm we’ve tracked this year, attackers had to work around code review, because faking an approval was harder than dodging it. This does it for free. An attacker can make a malicious change look reviewed and signed off by someone your team trusts, your pipeline builds it and ships it downstream, and your own audit log swears the whole thing was legitimate. Deleting a repo costs you a bad afternoon. Forging trust in it costs you every release that follows,” Münch said. Related: CISA Urges Immediate Patching of Exploited Microsoft, VMware, Apple Vulnerabilities Related: 943 Patches Rolled Out With Oracle’s August 2026 Security Update Related: Chrome, Firefox Updates Patch Dozens of Vulnerabilities Related: Trivy, Not LiteLLM Behind the 2,500 Org Compromise Written By Ionut Arghire Ionut Arghire is an international correspondent for SecurityWeek. Daily Briefing Newsletter Subscribe to the SecurityWeek Email Briefing for the latest cybersecurity threats, trends, and expert insights. More from Ionut Arghire Xpander Raises $7.5 Million for AI Management and Governance300,000 WordPress Sites Potentially Exposed to Hacking Due to Form Plugin FlawHeights Finance Data Breach Impacts at Least 1.2 Million IndividualsGitLab Patches Critical Code Injection VulnerabilityDozens of WebKit Vulnerabilities Patched With Fresh macOS, iOS Security Updates680,000 Impacted by French Tax Authority Data Breach40,000 Impacted by SafePal Data BreachRecent macOS Screen Sharing Vulnerability Exploited in Attacks Latest News Hackers Using AI to Target Siemens PLCs in Critical US SectorsVirtual Event Today: CodeSecCon – Secure Your Code and ApplicationsPrevalent AI Raises $22 Million to Expand Data Fabric PlatformUS Charges 17 Iranian Hackers, Offers $10 Million Rewards for 5 of ThemCl0p Ransomware Group Names Over 40 Victims of PTC Windchill CampaignCISA Urges Immediate Patching of Exploited Microsoft, VMware, Apple Vulnerabilities943 Patches Rolled Out With Oracle’s August 2026 Security UpdateChrome, Firefox Updates Patch Dozens of Vulnerabilities Trending Daily Briefing NewsletterSubscribe to the SecurityWeek Email Briefing to stay informed on the latest threats, trends, and technology, along with insightful columns from industry experts. Webinar: Rethinking Cyber Defense for AI-Speed Attacks August 18, 2026 Join this live webinar as we explore if detection-first security operations can keep pace with AI, or if it’s time to rethink prevention as the strongest default. Register Virtual Event: CodeSecCon 2026 August 19, 2026 CodeSecCon bridges the gap between dev and security. Discover best practices for secure coding, innovative risk-reduction tools, and safe AI integration to cultivate a true DevSecOps culture. Safely secure your apps! Register People on the MoveDali Rajic is joining OpenAI as Chief Revenue Officer.Erika Dean has been appointed Chief Information Security Officer at Tricentis.C1 has named Jeff St. Clair Chief Revenue Officer.More People On The MoveExpert Insights The AI Governance Gap Is a Leadership Problem: Waiting Won’t Close It Organizations are rushing to implement AI without fully grasping where its legal protections begin and end. (Steve Durbin) Rethinking AI Security: Why CASB and DLP Need an Interaction-Aware Layer Build your strategy around answering these questions to ensure employees use AI productively while keeping sensitive data, IP, and agent behavior within the boundaries set for safe AI use. (Etay Maor) Timeless Compliance: Why Better Questions Beat Bigger Frameworks The best compliance programs aren't the biggest ones. They're the ones built on a short list of questions that can actually be answered, and that still hold true when the models change. (Matt Honea) Is Patching Dead? Vulnerability Management in the Post-Mythos Era You cannot out-patch a machine that writes a working exploit from a vulnerability description in twenty hours. Stop trying to optimize a game you cannot win. (Danelle Au) When Identity Verification Fails: Lessons from a Real-World SIM Swap and Near Account Takeover Identity confidence changes throughout every interaction and should be reassessed continuously as new risk signals emerge. (Torsten George) Flipboard Reddit Whatsapp Whatsapp Email

Indicators of Compromise

• cve — CVE-2026-19478

Entities