GitLab released version 19.2 of its DevSecOps platform on 16 July 2026, introducing agentic automation features designed to address the growing backlog of security and review tasks. The update arrives as AI coding tools generate code at a pace that exceeds manual review capacity.
The release brings four features out of beta or into public beta, including Dependency Scanning Auto-Remediation, Security Review Flow, GitLab Duo CLI, and Custom Flows. These additions aim to automate security workflows and reduce the manual effort required to validate code changes.
Background
GitLab 19.2 focuses on automating repetitive security and review processes that have accumulated as developers produce more code with AI assistance. The company stated that the features target the gap between code generation and human review, a challenge that has intensified across the industry.
Dependency Scanning Auto-Remediation, now generally available, automatically proposes fixes for vulnerabilities identified in project dependencies. The feature scans dependency files and generates merge requests with suggested patches, reducing the time developers spend on manual remediation.
Security Review Flow
Security Review Flow, also exiting beta, streamlines the process of requesting and conducting security reviews. It integrates directly into the merge request workflow, allowing teams to enforce security review policies without leaving the platform. The feature includes automated assignment based on file ownership and review history.
GitLab Duo CLI
GitLab Duo CLI enters public beta with version 19.2. This command line interface provides access to GitLab Duo capabilities, including code explanation and vulnerability analysis, directly from the terminal. The tool aims to keep developers within their preferred environment without switching contexts.
Custom Flows, now in public beta, allow teams to define their own automation workflows using GitLab Duo agents. Users can configure triggers and actions based on events such as merge request creation or pipeline failures. The feature is designed to adapt to team specific security and review processes.
Matt Saunders, a GitLab representative, confirmed these updates in the announcement. He noted that the features were developed in response to feedback from organizations struggling to keep pace with AI accelerated code production. The release includes performance improvements to the platform’s scanning and review engines.
Industry analysts have observed that the integration of agentic automation into DevSecOps platforms reflects a broader trend. As AI coding tools increase code output, platforms are adding automation layers to maintain security and quality standards. GitLab’s approach focuses on embedding these capabilities directly into existing workflows rather than requiring separate tools.
Organizations using self managed GitLab instances can upgrade via the standard update channel. GitLab.com users will receive the features automatically. The company has published documentation for each new feature on its official site.
Looking ahead, GitLab plans to extend automation capabilities in future releases. The company has indicated that additional agentic features are under development, though specific timelines have not been announced. The focus remains on reducing manual toil in security and review processes as code generation volumes continue to rise.
