GitLab has introduced its 18.3 release, advancing its ambition to become the first AI-native platform for software engineering. The update builds on the company’s AI orchestration capabilities with enhanced Flows, stronger enterprise governance, and seamless tool integration designed to support collaboration between human developers and AI systems.
Enhancing collaboration between AI and developers
As debates continue over whether AI could replace developers, GitLab has taken a different approach. The company focuses on helping engineering teams work effectively with AI, positioning agents as partners rather than separate tools. By integrating AI into the core of development processes, GitLab aims to simplify the orchestration of complex workflows and improve productivity across teams.
Key features in the new release
The 18.3 release introduces several major updates. The Model Context Protocol (MCP) server allows AI systems to connect directly with GitLab projects in a secure and standardised way, removing the need for custom integrations. This enables AI tools, including Cursor, to operate natively within the platform.
Developers can now also access broader support for command line interface (CLI) agents such as Claude Code, Codex, Amazon Q, Google Gemini, and opencode with Bring Your Own Key functionality. Teams can delegate routine development tasks to these agents simply by mentioning them in issues or merge requests.
Transparency is another focus of the release. With Agent Insights, users gain visibility into AI agents’ decision-making processes, making it easier to track activity, optimise workflows, and ensure best practices are followed.
In addition, the platform’s Knowledge Graph now delivers real-time code indexing. This improves the speed and accuracy of code searches, offering contextual results that help developers identify the right information quickly.
Expanding GitLab AI flows
Building on the Software Development Flow introduced in version 18.2, GitLab 18.3 adds two new flows. The “issue to MR Flow” helps developers move from concept to working code in minutes, while the “convert CI File Flow” supports smooth migration and intelligence within continuous integration processes.
These additions highlight GitLab’s broader strategy of embedding AI throughout the development lifecycle, offering engineering teams more streamlined ways to plan, build, and deploy software.
