GitLab Launches Newer Version to Boost AI Automation, Governance, and Developer Tools
The 18.7 update strengthens the foundation for agentic AI workflows that help streamline repetitive development tasks and maintain organizational standards.
GitLab has launched version 18.7 of its DevSecOps platform, bringing new AI-driven automation, governance controls, and developer enhancements as it prepares for the general availability of its GitLab Duo Agent Platform in early 2026.
The 18.7 update strengthens the foundation for agentic AI workflows that help streamline repetitive development tasks and maintain organizational standards.
"GitLab 18.7 delivers development, operations, and security capabilities that strengthen control, improve consistency, and build confidence as teams integrate AI further into their workflows," Bill Staples, CEO at GitLab, wrote in a blog post.
A highlight is Custom Flows, which lets teams define YAML-based sequences that automatically trigger multistep actions — such as fixing failed pipelines, updating dependencies, or running policy checks — in response to GitLab events.
Security and governance receive attention with AI-powered false positive detection for Static Application Security Testing (SAST). This feature uses AI to flag likely false positives early, helping developers and security teams cut through noise and focus on genuine vulnerabilities in real time.
GitLab also introduced Custom Agent Versioning, allowing projects to pin specific versions of AI agents and flows, offering stability and preventing disruption from unexpected updates. Administrators gain granular control over foundational agents, turning them on or off by instance or group to guide AI adoption responsibly.
Another addition is the Data Analyst Agent, which enables natural-language exploration of GitLab data by generating and running GitLab Query Language (GLQL) queries and surfacing insights without dashboards or manual coding.
Improvements to core DevOps workflows, including dynamic pipeline input selection, further enhance usability and efficiency.
The release is available now for Premium and Ultimate users on GitLab.com and self-managed environments, with expanded access planned for GitLab Dedicated customers.
Comments ()