Wisevoter
World

GitLab Adds Tools to Govern AI-Driven Development

GitLab introduced new capabilities for managing AI-generated code, including Artifact Central, Dependency Firewall, Secrets Manager and Impact Analytics.

Updated on Nov. 5, 2026 in Artificial Intelligence

Isometric editorial illustration of four identical sand-colored shipping containers in an orderly row on a concrete pier, one offset beside a steel crane arm.
GitLab on Tuesday introduced Artifact Central, Dependency Firewall, Secrets Manager and Impact Analytics to govern code generated by AI agents on its platform. AI Illustration. Upload story photo >

GitLab announced new capabilities for managing AI-driven software development, introducing Artifact Central, Dependency Firewall, Secrets Manager and Impact Analytics to govern AI-generated code. The company said active users of agentic software development on its platform grew 200% year over year over the last three months.

Why it matters

Organisations often rely on separate tools for coding, security, artifact management and deployment, which can make it difficult to keep identity and policy controls consistent across the software development lifecycle. The new features aim to bring those controls into one platform as AI-generated code becomes more common.

GitLab Dependency Firewall checks packages against organisational policies before they enter a build, and Secrets Manager centralises build-time secrets and limits access to the jobs that require them. GitLab claims Orbit reduced agent retries 45 times and token consumption 4.5 times across 280,000 coding-agent queries, figures the company has not independently verified.

The players

GitLab

GitLab is a software development platform that combines source code management, CI/CD and security tools in one application.

Anthropic

Anthropic is an artificial intelligence company that develops the Claude family of AI models.

The details

GitLab Artifact Central is in beta on GitLab.com, and GitLab Secrets Manager is generally available there. GitLab Security Standard is available now, GitLab Orbit entered beta in June and has been used by more than 3,500 organisations, and Duo Agent Platform Impact Analytics is in early access.

Timeline

  1. June: GitLab Orbit entered beta.

  2. Later this month: Artifact Central is planned for GitLab Self-Managed.

  3. Next month: Anthropic's Claude Mythos 5 and 5.1 are planned for new GitLab Duo Agent Platform security flows.

  4. Next month: Orbit is expected to reach general availability across deployment options.

  5. 19.5 release: Secrets Manager is scheduled for GitLab Self-Managed.

The Tech Race

GitLab's move follows the pattern of AI-coding platforms bundling third-party frontier models into developer workflows, as seen with Anthropic's Claude models now being embedded directly into security tooling. The competitive pressure is shifting from raw code generation toward governance, secrets handling and policy enforcement for AI-written software.

Developers using GitLab.com can now centralise build-time secrets through Secrets Manager, which limits access to only the jobs that need them. Teams on Self-Managed deployments will receive Artifact Central and Secrets Manager on later dates, so those features are not yet available to them.

The takeaway

As AI agents generate more of the code organisations ship, governance over packages, secrets and artifacts is becoming as important as the code generation itself. Teams evaluating these tools should check which features are available on their own deployment type and when.

What happens next

Artifact Central is planned for GitLab Self-Managed later this month, Claude Mythos 5 and 5.1 are planned for GitLab Duo Agent Platform security flows next month, and Secrets Manager is scheduled for Self-Managed with the 19.5 release.

Further reading

Explore more on Artificial Intelligence developments shaping how software is built and secured.