---
title: 'Atlassian Rovo AI Agents Tutorial for Beginners'
source: 'https://youtube.com/watch?v=JEyhRfkP0bY'
video_id: 'JEyhRfkP0bY'
date: 2026-08-12
duration_sec: 685
---

# Atlassian Rovo AI Agents Tutorial for Beginners

> Source: [Atlassian Rovo AI Agents Tutorial for Beginners](https://youtube.com/watch?v=JEyhRfkP0bY)

## Summary

This video is a step-by-step tutorial on building AI agents in Jira using Atlassian Rovo. It covers using pre-built agents, creating custom agents, enhancing them with company knowledge and skills, and automating their execution. The tutorial is sponsored by Atlassian and is presented by Kevin.

### Key Points

- **Introduction to Atlassian Rovo** [00:00] — Kevin introduces the video, explaining that he will show how to build AI agents in Jira using Atlassian Rovo. The video is sponsored by Atlassian.
- **Using Pre-Built Agents** [00:29] — Kevin demonstrates using the 'Ask Rovo' feature in Jira to access pre-built agents. He browses over 25 agents and selects the 'Jira Theme Analyzer' to identify themes in customer feedback.
- **Analyzing Feedback with Jira Theme Analyzer** [02:25] — The agent analyzes 31 customer feedback tickets and groups them into themes, showing frequency and providing links to original issues. Kevin notes that shipping and delivery is the most common theme.
- **Creating a Custom Agent** [03:06] — Kevin shows how to create a custom agent in Rovo Studio. He describes a customer support triage agent that categorizes requests, determines priority, recommends next steps, and drafts responses.
- **Enhancing Agent with Knowledge** [05:59] — Kevin adds a Confluence page (Customer Support Playbook) as custom knowledge to the agent. This allows the agent to reference company policies, such as offering a 15% discount for late deliveries.
- **Adding Skills for Actions** [07:55] — Kevin adds a Slack skill to the agent, enabling it to send a message to a Slack channel when a refund request exceeds $100. He tests this by prompting a refund of $150.
- **Automating Agent with Triggers** [09:37] — Kevin sets up an automation trigger so the agent runs automatically whenever a new support request is created in Jira. The agent analyzes the request and adds a comment to the issue.
- **Sharing Agents** [11:09] — Kevin explains how to share agents with others via the 'Users and Collaborators' section in Studio, allowing access to the entire organization or specific users.

### Conclusion

The video demonstrates how to build, customize, and automate AI agents in Jira using Atlassian Rovo, from using pre-built agents to creating custom ones with knowledge and skills. It shows how these agents can streamline workflows and improve efficiency in customer support.

## Transcript

In this video, I’ll show you how to build AI  agents in Jira using Atlassian Rovo step by step.   We'll start with a pre-built agent, then create  our own custom AI agents, make it smarter with  
company knowledge and skills, and finally automate  it. Atlassian sponsored and supported this video.   I'm Kevin, and let's dive in. To get started,  let's look at some of the pre-built agents   available in Atlassian Rovo. Here I am in Jira.  Now we use Jira here at the Kevin Cookie Company  
to keep track of customer feedback. As you can  see, customers have plenty of opinions when it   comes to cookies. Let's see if an agent can help  us make sense of all of this feedback. In the top  
right-hand corner, let's click on Ask Rovo, and  this opens up Rovo, Atlassian's AI assistant.   Now I want to identify the key themes in this  feedback along with how often they show up.  
Now right down here, I could type in a prompt  directly into Rovo, but Atlassian already includes   an agent that's designed specifically for this  type of task. To switch an agent, let's click   on this dropdown right here, and you'll see  a list of available agents. Right down here,  
let's click on Browse Agents, and right up on top,  let's click on Browse All. Here, you'll find over   25 pre-built agents. Now, some help pull insights  from meetings. Right here, you can even generate   OKRs. If we scroll down a little bit more, here's  an example where you could write release notes,  
and you have many more. Now for this example, I  want help identifying trends in customer feedback,   and right here, there's the Jira Theme Analyzer.  Now as the description says, it'll find common  
themes from a group of issues in Jira. That's  exactly what I need, so let's click on this.   At the top, you'll now notice that I've switched  from Rovo to the Jira Theme Analyzer agent.   One of the benefits of an agent is that it's  built for a specific purpose. In this case,  
the Jira Theme Analyzer is designed to review  all these Jira issues, group them into themes,   and then summarize the results. That means I don't  have to figure out the perfect prompt every time I   want to run this analysis. Now, right down here at  the bottom, you'll notice a few suggested prompts  
that can help us get started. Let's try asking  the agent to identify all the different themes   in our Jira issues. Here, I'll type in my prompt,  and then let's run that. In just a few moments,   the agent analyzed all 31 customer feedback  tickets and grouped them into common themes. Now,  
let's switch into full screen mode so we can see  this better. Right up on top, I'll click on this,   and here, let's click on the full screen. Here, I get a table with all of the different   themes. I also get a sense of how often  those themes show up. Right over here,  
we have a summary, and on the right-hand side, we  get links back to the original customer feedback.   Now, looking at all these issues, it appears that  shipping and delivery comes up most often. Now,  
that's probably something I wouldn't have spotted  nearly as quickly by reviewing all these issues   one at a time. Now, this is a great way to quickly  identify trends without manually reviewing every   single piece of feedback. So far, we've looked  at how you can use existing agents, but what if  
you have a workflow that's unique to your team or  business? Let's now build our own custom agent.  Back here within Jira, in the top right-hand  corner, let's click on the Rovo icon,   and then click on the agent selector. Down here  at the bottom, you'll see the option to create  
an agent. Let's click on that. Now, alternatively,  you can also navigate to the website that you see   at the bottom of the screen. I'll click here. This  opens up Rovo Studio, where we can create our own   agents from scratch. For the Kevin Cookie Company,  let's create a customer support triage agent.  
When a new support request comes in, we'll  have the agent categorize the request,   determine the priority, recommend next steps, and  draft a response for the customer. So over here,   I'll type in my prompt, and then right over  here, let's run that. In just a few moments,  
here, Rovo generates an agent name. We  also get a description for this agent,   and right down below, we have all the instructions  for the agent. Now, honestly, that's a lot faster   than starting with a blank page. Of course, I  can always go through this, review it, and I  
could customize any of these instructions, but I  think this gives me a really good starting point.   Right down here, let's click on build agent. We've  now built the agent, and this drops us into the   agent details in studio. Now, this all looks  good to me. So, in the top right-hand corner,  
let's click on publish. Now that we've built our  agent, let's test it out. Back here within Jira,   in the top right-hand corner, when I click on  ask Rovo, let's click on the agent dropdown,   and right here, you'll notice that we now have  a new agent, the customer support triage agent,  
the one that we just created. Let's select that  agent, and right down here at the bottom, I can   now prompt my new agent. Now, I can reference  any one of these support tickets directly here,   but let's actually make this a little bit easier.  Over here, I'll open this ticket directly. This  
opens up the support request details, and I can  see that the customer selected expedited shipping,   so they were expecting it faster, but it actually  arrived late. That's completely unacceptable. So,   we need to figure out how to prioritize this,  and also draft a response to the customer.  
Before we do that though, let's expand this so  we can see things a little bit better. I'll click   here. Within the issue, right here, you'll notice  that we can run an agent. I'll click on this,   and there's that agent that we just created. I'll  select this, and that'll kick off the work. Here,  
the agent reviews the request, and first off, it  chooses the category. This is a delivery problem.   It also assigns a priority level. Here it's saying  medium, and it also gives me the reasoning for why   it shows that. Here we have next steps, and at the  bottom, it even drafts a response that we could  
send to the customer. Now, the beautiful thing  is, instead of manually reviewing every single   request, the support team now has a structured  starting point that they can work from. Now,   down below, let's now save this comment. Here,  I'll scroll down. This looks good, and let's  
click on save. This works really well already,  and we could share it out with the team. However,   we can also continue improving the agent by giving  it additional capabilities. Back here in Atlassian  
Studio, over on the left-hand side, let's click  on agents, and right here, let's click on View   My Agents. And here, I can see the customer  support triage agent that we created earlier.  
Let's open it. Right here at the top, you'll see  instructions that guide the agent's behavior.   And if I scroll down, here, you'll notice that we  could add additional capabilities. We have skills,   knowledge. Here, we have conversation starters,  and then more. Now, let's start with knowledge.  
To do that, in the top right-hand corner, let's  click on Edit. Then, on this screen, let's scroll   down, and here we see knowledge. I'll expand this,  and let's add custom knowledge. I'll check this,   and then let's click on Add Knowledge. One  of the easiest ways to improve an agent is by  
giving it access to company-specific information.  So right up in here, let's add a Confluence page.  I'll check this box, and let's select Content  Under, and right here, I see the Kevin Cookie   Company Customer Support Playbook. This includes  our refund, shipping, and missing item policies.  
Just as an example, let's imagine that an order  arrives, let's say, more than two days later   than the estimated delivery date. In the policy,  we provide customers with a 15% discount code.   Without access to this information, the  agent would have no idea about this policy.  
Now that I've added this page, down below, let's  click on Add. Right up on top, let's also update   the instructions with one additional line to  make sure that the agent references the custom   knowledge that we just provided. Once you do  that, in the top right, let's click on Publish.  
At the top, we can now test the updated agent. Over here, let's click on Test. At the bottom   here in the prompt field, let's type in, "My  cookies arrived three days later than the   estimated delivery date," and then let's run that.  Earlier, the agent would have given us just a  
generic recommendation, but right up here, we can  see that it's referencing official Kevin Cookie   Company policy. Here, it offers the customer a 15%  discount, and it's also providing the coupon code,   "SorryCookie15," just like I expected. Let's now  make our agent even more capable by giving it the  
ability to take action, and one way to do that  is by adding a skill. Back here within Studio,   I have the Customer Support triage agent open,  and up on top, let's click on Edit. And on the  
Edit screen, let's scroll down, and we'll see the  section for Skills. I'll expand that. Here, we can   see we currently have no skills. Let's click on  Add. Whenever the agent sees a refund request over   $100, I want it to notify the customer support  team in Slack so they can review it right away.  
Right up on top, we can search for Skills. Here,  I'll type in Slack, and I get all of the different   actions that I could take. Now, here, I see the  option to send a message. That's what I'm looking   for, so let's select that, and down below, I'll  click on Add. And right here, we can see that we  
added the first skill. However, we still need to  tell the agent when to use this skill, just like   we did with knowledge. So right over here, let's  go up to Instructions, and I'll add a sentence.   If a customer requests a refund greater than $100,  send a message to the Customer Escalation Slack  
channel that includes a summary of their request,  the reason for the escalation, and the recommended   next action. Now that I've entered that, right  up on top, let's click on Publish. Let's now test   that. Right over here, let's click on Test. Right  down below, I'll type in the following prompt,  
my order never arrived. I'd like a refund of 150.  Let's send that. Right up on top, let's send the   message. Right here, it detects that the refund  request is over 100, so it escalated it to Slack.   Let's have a look in Slack. Here in Slack, we  can see the notification. That way, the customer  
support team knows to look at this request right  away. Running the agent manually is helpful, but   we can make this even more powerful by automating  the process. Whenever a new customer support   request is created in Jira, I want this agent to  automatically review it. Here in Atlassian Studio,  
over on the left-hand side, let's click on  Triggers. A trigger determines when an agent   should run. For example, I could run an agent  on a schedule, when a work item changes status,   or when a new issue is created. Right over here,  let's add an automation trigger. For this example,  
I want the trigger to run whenever a new customer  support request is submitted. So over here, I'll   type in the trigger. Then for the instructions,  let's enter in, analyze a new support request,   and add your findings as a comment on the issue.  Then down here, let's add the trigger. Here,  
Atlassian automatically builds out the automation  for me. Right here, we have step number one,   and let's select the Jira space that it should  monitor. So over here, I'll select My Team,   and next, we have the agent. And this will  review any new support requests that come in.  
This will use the instructions, the knowledge, and  the skills that we added earlier. And then lastly,   it'll add its findings directly to the issue as a  comment. Now, this all looks good to me. So right   up on top, let's click on Save and Enable. Now,  let's say it in action. To do that, let's go back  
to Jira. Down below, let's now add a new support  request. Discount code won't apply at checkout,   and let's create that. Right up on top, let's  click into it. Now, over on the right-hand side,   we can see that the agent has added its analysis  to the support ticket. Now, instead of waiting for  
someone to manually review every request, instead,  the agent works as part of the support workflow   from the moment a ticket is submitted. Before  we wrap up, let's take a quick look at how you   can share your agents with others. Within Studio,  over on the left-hand side, let's click on Users  
and Collaborators. Here, I can make the agent  available to everyone in my organization, or I   can limit access to specific users. If you'd like  to learn more about Atlassian Rovo, check out the   links right down below in the description. Thanks  for watching and I'll see you in the next video.