Recognising the opportunity to transform the QA process, we asked: What if we could automate the entire workflow – from extracting acceptance criteria to creating and executing test cases, and even feeding results back into Jira for traceability?
Inspired by this vision, we created a custom Forge-built Rovo agent that provides a solution to automate and streamline the entire test suite lifecycle, fundamentally changing the way QA teams work.
How this custom Rovo Agent works
At its core, the Agent automatically pulls information from Jira tickets (acceptance criteria, comments and custom field information) and uses this data to build a comprehensive test suite in the Zephyr Scale Jira app.
The agent also taps into linked Confluence documentation, intelligently extracting additional feature information and other non-functional requirements to generate robust, meaningful test cases, regardless of ticket format or documentation style.
The agent reads and interprets information from both Jira and Confluence, extracts acceptance criteria and supporting information, and then creates the necessary test cases, cycles, and plans in Zephyr via API calls. It can even monitor test execution, and is designed with future integrations in mind, such as linking up with reporting tools like eazyBI, PowerBI, or Tableau for advanced analytics.
Intuitive user experience
The user experience was designed to be intuitive. Testers and managers can interact with the agent using natural language prompts, such as Generate a test suite for all stories in ‘Ready for Test’, or Create positive, negative, and edge case test cases for Work Item X.
The agent can also summarise testing coverage, show test plans and cycles for specific tickets, and update Jira with the latest test results from Zephyr.
This seamless integration between Jira, Confluence, and Zephyr is orchestrated entirely through APIs, laying the groundwork for even more advanced reporting and analytics in the future.