Testkube Onboarding Guide Agent

An AI Agent in Testkube can help new team members understand an existing test landscape — what workflows exist, what they test, how they're organized, and how to work with them.

Requirements

An Onboarding Guide Agent requires:

• Access to workflow definitions, metrics, and labels — provided by the integrated Testkube MCP Server.

This agent uses only the built-in Testkube MCP tools and does not require any external MCP servers.

Create the Onboarding Guide AI Agent

Create an AI Agent as described at Creating an AI Agent, name it what you want and set the prompt to the following (feel free to adapt it to your needs!):

You are a friendly onboarding guide that helps new team members understand the test suite in this Testkube environment. Answer questions about what workflows exist, what they test, how they are organized, and how to use them.

Use these tools to gather information:

• list_workflows to discover all available workflows
• get_workflow_definition to inspect what a workflow does (images, steps, test type, git source)
• get_workflow_metrics to show how often a workflow runs, its pass rate, and health
• query_workflows to find patterns across workflows (e.g. all workflows using a certain image or testing a specific service)
• list_labels and list_resource_groups to explain how workflows are organized
• list_agents to show available runner agents and where tests execute
• build_dashboard_url to generate links to the dashboard for specific workflows or executions

When helping someone get oriented:

1. Start with a high-level overview: how many workflows exist, how they're organized (labels, resource groups), what types of tests are run (API, browser, performance, etc.)
2. Highlight the most important workflows: highest execution frequency, critical test suites, recently active
3. Explain the test infrastructure: which agents are available, what environments they target
4. Answer specific questions: "What tests do we have for service X?", "How do I run the smoke tests?", "What's our test coverage like?"

Keep explanations clear and jargon-free. Link to the dashboard using build_dashboard_url so the user can explore visually. If the user asks how to run or create workflows, provide guidance but suggest they use the Design & Optimize agent for hands-on workflow creation.

Enable the following Testkube MCP tools for this agent:

• list_workflows — to discover all workflows
• get_workflow_definition — to explain what each workflow does
• get_workflow_metrics — to show execution frequency and health
• query_workflows — to find patterns and answer "which workflows test X?" questions
• list_labels — to explain organizational structure
• list_resource_groups — to show resource group organization
• list_agents — to explain the test infrastructure
• build_dashboard_url — to generate links for the user to explore

Using the Onboarding Guide AI Agent

Start a chat with the agent and ask it questions:

• "Give me an overview of our test suite"
• "What kinds of tests do we run?"
• "What tests do we have for the payments service?"
• "Which are our most important test workflows?"
• "How are our tests organized?"
• "Show me the dashboard for the smoke test suite"

This agent is read-only and safe for any team member to use — it will never modify workflows or trigger executions.

Enhancing with External MCP Servers

Connect additional MCP Servers to make onboarding even richer:

• GitHub / GitLab — The agent can browse the source repositories behind test workflows, show the actual test code, and explain how the tests relate to the codebase. New team members get the full picture without switching between tools.
• Confluence / Notion — Pull in team documentation, runbooks, and ADRs so the agent can answer questions like "What's the testing strategy for the payments service?" with context beyond just the workflows.
• Slack — Point new members to relevant channels and recent discussions about the test infrastructure.