TLDR: GitLab has unveiled the public beta of its Duo Agent Platform, an advanced orchestration tool designed to integrate specialized AI agents into the DevSecOps lifecycle. This platform aims to transform traditional development workflows into dynamic, human-AI collaborations, allowing AI to manage routine tasks like refactoring, security scanning, and testing, while developers focus on complex problem-solving. Available to Premium and Ultimate customers, the platform leverages GitLab’s comprehensive system of record to provide AI agents with rich project context and supports interaction through an agentic chat interface in various IDEs and the GitLab Web UI.
GitLab Inc., a leading provider of DevSecOps solutions, has announced the public beta launch of its groundbreaking GitLab Duo Agent Platform. This innovative orchestration tool is set to redefine software development by enabling seamless, asynchronous collaboration between human developers and specialized AI agents across the entire DevSecOps lifecycle. The platform is currently available to GitLab.com Premium and Ultimate customers, as well as self-managed instances running version 18.2.
The Duo Agent Platform moves beyond traditional linear development processes, transforming them into dynamic, multi-agent systems. Its core objective is to offload routine and repetitive tasks—such as refactoring, security scanning, testing, code review, and research—to AI, thereby freeing developers to concentrate on more complex problem-solving and innovation.
Bill Staples, CEO at GitLab, emphasized this vision, stating, “We are reimagining the future of software engineering as a human and AI collaboration, where developers focus on solving technical, complex problems and driving innovation, while AI agents handle the routine, repetitive tasks that slow down progress.”
A key strength of the platform lies in GitLab’s position as a centralized system of record. By encompassing code, merge requests, CI/CD pipelines, test results, compliance checks, and project history, the Duo Agent Platform provides AI agents with comprehensive organizational context. This rich context empowers agents to make informed, autonomous contributions that align with established development standards and workflows.
The platform introduces several critical features, including ‘Duo Agentic Chat,’ a natural language interface that allows developers to interact with AI agents directly within their development environments, such as VS Code, JetBrains IDEs (IntelliJ, PyCharm, GoLand, WebStorm), and the GitLab web application. This chat experience supports iterative feedback, remembers context, tracks history, and facilitates streamlined delegation using slash commands like /explain and /tests.
Beyond GitLab’s ecosystem, the Duo Agent Platform integrates with the Model Context Protocol (MCP), enabling agents to interoperate with external systems and services. This positions the platform as an extensible, intelligent layer across the broader DevSecOps landscape.
Rachel Stephens, research director at RedMonk, commented on the significance of this release, noting that GitLab’s orchestration platform represents “a significant step forward in integrating AI agents into existing DevSecOps toolchains.”
GitLab plans to deliver monthly updates to the platform as part of its 18.x release train, with general availability anticipated later this year. Future enhancements include an AI Catalogue for sharing agents and Flows, a Knowledge Graph to boost context and accuracy, and support for custom agent creation, a CLI, and an official MCP server.
Also Read:
- CoreOps.AI Launches AgentCORE to Revolutionize Enterprise Operations with Intelligent AI Agents
- OpenAI Introduces ChatGPT Agent: A Leap Towards Artificial General Intelligence
The company acknowledges that it is not alone in this space, with rivals such as GitHub, JetBrains, and Atlassian already deploying their own agent-driven solutions, each taking a distinct approach to integrating AI into the development lifecycle.
