GitLab Duo Agent Platform Beta Launches for Asynchronous Developer and AI Collaboration

GitLab Duo Agent Platform enables AI agents to collaborate asynchronously with developers, automating tasks and improving workflow. Beta supports IDEs like VS Code and JetBrains.

Published on: Jul 19, 2025
GitLab Duo Agent Platform Beta Launches for Asynchronous Developer and AI Collaboration

Introducing GitLab Duo Agent Platform: AI Meets DevSecOps Collaboration

GitLab has launched a public beta of the GitLab Duo Agent Platform, a new orchestration tool built to facilitate asynchronous collaboration between developers and AI agents. This platform creates an intelligent layer that lets AI agents work alongside development teams by understanding project context and aligning with organizational standards.

What GitLab Duo Agent Platform Offers

The Duo Agent Platform currently supports GitLab Premium and Ultimate customers, available both on GitLab.com and self-managed setups. The beta integrates directly with popular IDEs like Visual Studio Code and JetBrains IDEs, with plans to broaden IDE support soon. The general release is expected by the end of this year.

  • Chat Agent: Handles general development tasks like reading issues and code diffs.
  • Software Development Agent: Works on assigned tasks by creating code changes in virtual environments and opening merge requests for review.

Both agents are in beta, and GitLab is actively developing additional AI agents focused on areas such as product planning, software testing, code review, platform engineering, security analysis, deployment, and deep research.

Agent Flows: Coordinated Multi-Agent Workflows

Beyond individual agents, GitLab is building agent flows—complex, autonomous workflows that involve multiple AI agents working together. For example, the software development flow orchestrates agents to plan, implement, and test code changes by analyzing the project’s structure, code base, history, and relevant GitHub issues or merge requests.

Extending Reach with Model Context Protocol (MCP)

The platform supports connections to remote or local MCP (Model Context Protocol) servers. This allows AI agents to interact with external systems beyond GitLab, such as Jira, ServiceNow, and ZenDesk, to gather context or perform actions. GitLab plans to release its official MCP Server soon.

The GitLab Duo Agent Platform represents a practical step toward integrating AI into day-to-day software development workflows, helping teams automate routine tasks and maintain focus on higher-value work.

For developers and IT professionals interested in AI’s role in software development, exploring resources on AI automation and prompt engineering can provide a solid foundation. Check out comprehensive AI courses and training at Complete AI Training to stay ahead.


Get Daily AI News

Your membership also unlocks:

700+ AI Courses
700+ Certifications
Personalized AI Learning Plan
6500+ AI Tools (no Ads)
Daily AI News by job industry (no Ads)
Advertisement
Stream Watch Guide