Learning GitLab Duo Agent Platform - Tony Marsh

module-name: "GitLab Duo Agent Platform"
area: "Product Knowledge"
gitlab-group: "AI-powered:Agent Foundations"
maintainers:
  - TBD

Introduction

This training module is intended to provide Support Engineers with an understanding of the GitLab Duo Agent Platform (DAP), including its capabilities, architecture, configuration, and support scope.

Goals of this training module

At the end of this module, you should be able to:

  • Understand what the GitLab Duo Agent Platform is and how it differs from standard Duo Chat
  • Configure and deploy agents in self-managed and GitLab.com environments
  • Troubleshoot common Agent Platform issues
  • Answer Agent Platform-related support tickets
  • Understand the security and governance considerations for agents
  • Support peers with questions about the Agent Platform

Prerequisites

  • Completion of the GitLab Duo training module is recommended

General Timeline and Expectations

  • This issue should take you 4h to complete.

Reminders:

  • This is a Public issue; don't include anything confidential.
  • You should do stage 0 first and the final stage last. Outside of that, you can perform tasks from the stages in any order you like.

Stage 0: Create Your Module

  • Create an issue using this template by making the Issue Title: "Learning GitLab Duo Agent Platform - your name"
  • Add yourself as the assignee
  • Consider setting a milestone and/or a due date to help motivate yourself!
  • Update your Support Team yaml file to indicate that you've started learning this knowledge area:
knowledge_areas:
  - name: GitLab Duo Agent Platform
    level: 1

Note: If you already have GitLab Duo as a level 2 or 3 in your product_categories, you can skip the below change.

product_categories:
    - name: GitLab Duo
      level: 1

Consider using the Time Tracking functionality so that the estimated length for the module can be refined.

Stage 1: Foundational Concepts

  • Done with Stage 1

Goal Gain a basic understanding of what agents are, how they differ from traditional chatbots, and the role of the Agent Platform in GitLab's AI ecosystem.

  • Read What is Agentic AI? (approx. 5 minutes)
  • Watch Developer Onboarding with GitLab Duo Agent Platform (approx. 5 minutes)
  • Important terms:
    • Agentic Chat: Advanced Chat capability that enables autonomous agent behavior
    • Agent Platform: The broader framework enabling agents to perform complex tasks
    • Foundational Agents: Pre-built agents provided by GitLab (e.g., Issue to MR)
    • Custom Agents: User-created agents using flows
    • External Agents: Third-party agents integrated via MCP (Model Context Protocol)
  • Read about Agent Platform Architecture to understand:
    • How agents interact with the AI Gateway
    • The role of service accounts and composite identity
    • How agents access GitLab resources
    • Data flow and security considerations

Stage 2: Agent Platform Overview and Documentation

  • Done with Stage 2

Goal Understand the Agent Platform's features, capabilities, and support scope.

Stage 3: Technical Setup and Configuration

  • Done with Stage 3

Goal Learn how to configure and deploy the Agent Platform in different environments.

If you already have a working GitLab Self-Managed instance on GitLab 18.6 or higher with Duo Enterprise, you can skip Stage 3

Set up a self-managed instance with Agent Platform

  • Deploy a single node Omnibus instance with GitLab 18.6 or higher
  • Apply an Ultimate license with Duo Enterprise add-on 
gitlab_rails['env'] = {
'GITLAB_LICENSE_MODE' => 'test',
'CUSTOMER_PORTAL_URL' => 'https://customers.staging.gitlab.com',
'CLOUD_CONNECTOR_BASE_URL' => 'https://cloud.staging.gitlab.com'
}
  • Activate the subscription using an activation code

Configure Agent Platform

Stage 4: GitLab Duo Agent Platform in Practice

  • Done with Stage 4

Goal Gain hands-on experience with GitLab Duo Agent Platform features and understand how agents work in practice.

  • Read about Foundational Flows
  • Try the Developer flow:
    • Create an issue in your test project
    • Use the "Generate MR with GitLab Duo" feature
    • Observe the agent's reasoning and generated MR
    • Review the agent session and activity tracking
  • Try the Fix CI/CD Pipeline Flow:
    • Create a failing pipeline in your test project
    • Use the "Fix pipeline with Duo" feature
    • Observe how the agent analyzes and fixes the issue

Try GitLab Duo Chat (Agentic)

Explore Agent Capabilities and Limitations

  • Test agent behavior with:
    • Different project contexts
    • Various code languages and frameworks
    • Complex vs. simple tasks
    • Edge cases and error scenarios
  • Understand agent limitations:
    • What agents can and cannot do
    • Accuracy and hallucination risks
    • Performance and timeout considerations
    • Model-specific behaviors

Explore Custom Agents and Flows

Stage 5: Troubleshooting and Support

  • Done with Stage 5

Goal Learn how to troubleshoot Agent Platform issues and understand support scope.

Troubleshooting Agent Platform

Stage 6: Handling Agent Platform Tickets

  • Done with Stage 6

Goal Learn how to handle customer support tickets related to the Agent Platform.

  • Search Zendesk for closed Agent Platform tickets and review 5
    • Use search keywords: "Agent Platform", "Duo Agent", "Agentic", "Issue to MR", "Fix Pipeline"
  • Answer 3 Agent Platform-related support tickets:
    • Ticket 1: __
    • Ticket 2: __
    • Ticket 3: __

Stage 7: Advanced Topics (Optional)

  • Done with Stage 7

Goal Deepen your understanding of advanced Agent Platform topics.

Penultimate Stage: Review

You should be able to:

  • Understand Agent Platform capabilities and limitations
  • Configure Agent Platform in different environments
  • Troubleshoot common Agent Platform issues
  • Answer Agent Platform support tickets

Any updates or improvements needed? If there are any dead links, out of date or inaccurate content, missing content whether in this module or in other documentation, list it below as tasks for yourself! Once ready, have a maintainer or manager review.

  • Update ...

Final Stage

  • Have your trainer review your tickets and assessment. If you do not have a trainer, ask an expert to review.
  • Manager: schedule a call (or integrate into 1:1) to review how the module went.
  • Update your Support Team yaml file to indicate that you're ready to work on tickets in this knowledge area:
knowledge_areas:
  - name: GitLab Duo Agent Platform
    level: 2

Note: If you already have GitLab Duo as a level 2 or 3 in your product_categories, you can skip the below change.

product_categories:
    - name: GitLab Duo
      level: 1

You will now be listed as "Ready to work tickets" in GitLab Duo Agent Platform on Skills by Person page and Product Categories by Person page.

🎉 Congratulations! You are now prepared to assist customers and colleagues with Agent Platform-related tickets. 🎉

If you think of any improvements to this module, please submit an MR! The file is located in an issue template in the 'support-training` repository.

Edited by Tony Marsh