GitLab 19.2 Brings Governed Agentic Automation to Clear the Backlog AI Coding Creates
GitLab 19.2 introduces Security Review Flow, Dependency Scanning Auto-Remediation, GitLab Duo CLI and Custom Flows, to take the heavy lifting of clearing the code backlog off of deverlopers.
The Security Review Flow can catch security flaws such as race conditions and business logic errors that are not seen by pattern-based scanners, according to the company announcement. This capability is in public beta, and is a core piece of GItLab’s Duo Agent Platform. What it does is figure out what the code is meant to do instead of trying to match known patters, and its findings will include the severity of the rist and suggests fixes where it can. Humans still need to give final approval, as the flow will not act on its own.
The dependency scanning piece, also in beta, was created to address the amount of time and effort developers need to find security risks by scanning vulnerable packages and suggesting a fix via a merge requesat. And, if dependency updates breaks a build, agents can fix those in the same request. Those changes the agent makes leaves a full audit trail, and still need to be approved before implementation — another safety feature.
Duo CLI brings the Duo Agent Platform’s agents to any terminal where developers work, and provides full project context.
Custom Flows, created by developments teams, can run automatically on GitLab events, and according to the announcement, “Custom Flows now authenticate to external services with short-lived, job-scoped tokens, so automation reaching cloud providers or internal APIs uses the same keyless pattern GitLab CI/CD pipelines already trust.”
Coming next from GitLab is the Flow Creating Agent, which turns a natural language description into a custom flow, the company said. An improved Fix CI/CD Pipeline Flow orders failures before taking action and makes inline suggestions for fixes or via a new merge request. And Duo Agentic Chat hands off much multi-step work to agents.
“Coding agents made it possible to generate far more code and moved the bottleneck downstream to reviews and security,” said Manav Khurana, chief product and marketing officer at GitLab, said in the announcement. “GitLab 19.2 puts agents to work on that bottleneck: fixing vulnerable dependencies, catching the flaws scanners miss, and automating the steps in between with a person still approving what ships.”
The post GitLab 19.2 Brings Governed Agentic Automation to Clear the Backlog AI Coding Creates appeared first on SD Times.
Tech Developers
No comments