Atlassian Rovo AI Agents Tutorial for Beginners

About this video

This sponsored Atlassian tutorial shows how to build AI agents in Jira using Atlassian Rovo, starting with pre-built agents, then creating, enhancing and automating a custom agent. In Jira for the fictional Kevin Cookie Company, Kevin opens Ask Rovo and browses over 25 pre-built agents, selecting the Jira Theme Analyzer. That agent analyzes all 31 customer feedback tickets, groups them into themes with frequency and links back to the original issues, finding shipping and delivery comes up most often. He then builds a custom customer support triage agent in Rovo Studio from a prompt describing categorizing requests, setting priority, recommending next steps and drafting a response; Rovo generates the name, description and instructions, which he publishes and tests on a late expedited-shipping ticket. To make it smarter he adds knowledge — a Confluence Customer Support Playbook covering refund, shipping and missing-item policies — and updates the instructions to reference it; in testing, a delivery three days late now correctly triggers the policy example of a 15% discount with code ‘SorryCookie15’. He adds a Slack skill so refund requests over $100 notify the Customer Escalation channel, tested with a $150 refund request, then adds an automation trigger so the agent runs whenever a new support request is created and posts its findings as a comment. Agents can be shared org-wide or limited to specific users.

Key takeaways

• Rovo includes over 25 pre-built agents; the Jira Theme Analyzer grouped 31 feedback tickets into themes with frequencies and source links
• Custom agents can be generated in Rovo Studio from a plain-language prompt, then reviewed, published and run directly inside a Jira issue
• Adding Confluence knowledge makes answers policy-specific — e.g. deliveries more than two days late get a 15% discount, demonstrated with code SorryCookie15
• Skills let agents take action: an instruction to escalate refund requests over $100 sends a summary message to a Slack escalation channel
• Triggers automate agents on events like a new issue being created, so triage runs automatically and posts findings as a comment

Episode details: published June 30, 2026 · 11:26 long. Summary written by Self Growth Videos from the episode itself. Business results and any income figures mentioned are Kevin Stratvert’s own claims, shared for education — they are not typical or guaranteed.

← Previous: Ep. 02 – Copilot Cowork and Apple’s Parental Controls: The Kevin Stratvert Podcast · Next: ChatGPT Tutorial for Beginners: How to Actually Get Work Done with AI →

All Kevin Stratvert videos · Kevin Stratvert profile · Kevin Stratvert on YouTube