A Step-by-Step Guide for Teams Ready to Automate Smarter with Atlassian AI

Imagine having a dedicated team member available 24/7 – one that instantly answers your team’s questions, automates repetitive tasks, and connects the dots across all your Atlassian tools. That’s exactly what Atlassian Rovo Agents offer.

The best part? You don’t need to write a single line of code to build one.

In this guide, we’ll walk you through everything you need to know to create, configure, and deploy your first Rovo Agent – from understanding what it is to having it live and working for your team.

What Is a Rovo Agent?

A Rovo Agent is an AI-powered virtual assistant that lives inside your Atlassian environment. It can be configured (without coding) to perform specific tasks such as:

~ Answering team queries from your Confluence knowledge base

~ Creating, triaging, or updating Jira issues automatically

~ Summarizing project status or meeting notes

~ Onboarding new employees with guided workflows

~ Sending proactive notifications and reminders

Rovo Agents are built on top of Atlassian Intelligence and integrate natively with Jira, Confluence, and other Atlassian tools – making them a powerful addition to any team’s productivity stack.

Prerequisites: What You Need Before You Start

Before building your first Rovo Agent, make sure you have the following in place:

~ An Atlassian Cloud account (Standard, Premium, or Enterprise plan)

~ Rovo enabled on your site (available as an add-on or included in higher-tier plans)

~ Admin or product access to configure agents

~ A clear use case in mind for your first agent

Tip: If you’re unsure whether Rovo is enabled on your site, go to admin.atlassian.com and check under Billing & Products.

Step-by-Step: Building Your First Rovo Agent

Step 1: Access the Rovo Agent Studio

1. Log in to your Atlassian Cloud account.

2. Navigate to the Rovo section from the top navigation or the Apps menu.

3. Click on Agents in the left-hand sidebar.

4. Select Create Agent to open the no-code Agent Studio.

What you’ll see: A clean, form-based interface where you define your agent’s name, role, behavior, and data sources – no code editor in sight.

Step 2: Define Your Agent’s Identity and Purpose

Give your agent a clear identity so users know exactly what it does

1. Agent Name: Choose something descriptive, e.g., “IT Support Bot” or “HR Policy Assistant”.

2. Description: Write a short summary of what the agent does. This helps teammates understand its function.

Step 3: Set the Agent’s Instructions (Prompt Configuration)

This is the brain of your agent. In the Instructions field, describe in plain English what the agent should do:

~ What topics should it handle?

~ What should it avoid?

~ How should it respond when it doesn’t know something?

~ Should it always escalate complex issues to a human?

Sample instructions for an HR Policy Agent:

– “Answer questions about company leave policies, benefits, and HR processes using the HR Confluence space.”

–“If you cannot find the answer, direct the user to raise a request in Jira Service Management.”

– “Do not answer questions unrelated to HR policies.”

Step 4: Connect Your Knowledge Sources

Your agent is only as smart as the data you give it. In this step, connect relevant knowledge sources:

Inside Rovo Studio, you’ll see the Knowledge section with the following options:

~ All organizational knowledge – Gives the agent access to all available company knowledge sources

~ Custom knowledge – Lets you selectively choose specific sources such as Confluence spaces, pages, or Jira projects.

~ No organizational knowledge – The agent will not use any internal organizational data and will rely only on its built-in capabilities.

Pro Tip: Start with a single, well-structured Confluence space or a tightly scoped Jira project which will produce better results than a broad but messy one.

Step 5: Add Skills

Skills allow your Rovo Agent to go beyond answering questions and actually perform tasks across Atlassian tools.

~ Click Add Skills to view available capabilities.

~ Browse the list of supported Atlassian integrations (such as Jira, Confluence, and others).

~ Select the specific skills your agent should have based on its use case.

 Some common capabilities enabled through skills include:

– Create work item.

– Comment on work item.

– Raise a request.

– Suggest escalation.

Step 6: Add Conversation Starters

Conversation starters give users quick prompts to understand your agent’s capabilities and start interacting instantly without thinking what to ask.

~ Click Generate for me to automatically create suggested conversation starters.

~ You can edit the suggestions to better match your agent’s use case and tone.

Example Starters:

– Summarize the latest updates from our sprint

– Summarize this Confluence page.

– What should I work on next.

– Are any of my work items overdue.

Step 7: Enable Advanced Capabilities (Optional)

Rovo Studio offers additional features to enhance your agent’s capabilities:

~ Web Search – Allow the agent to fetch information from public websites.

~ Deep Research – Enable for more detailed and high-quality responses(Only enable for subagent).

Step 8: Create Subagents (Optional)

Subagents let you break complex tasks into smaller, specialized agents that support your main Rovo agent.

Configuration Fields:

~ Trigger – When the subagent should be activated

~ Instructions – What the subagent should do and how it should behave

~ Skills – Add actions like Jira, Confluence, etc.

~ Knowledge – Select data sources (All, Custom, No organizational knowledge).

How to Configure:

1. Click Add Subagent

2. Set Trigger and Instructions

3. Add Skills and Knowledge sources

4. Enable Web Search / Deep Research if needed

5. Click Activate to create and enable your Rovo agent.

Example:

– Main Agent: IT Helpdesk

– Subagent: Account Subagent

– Subagent: Network Subagent

– Subagent: Hardware Subagent

5 Quick-Win Use Cases to Get You Started

Not sure what to build first? Here are five high-impact agents teams are already using:

IT Helpdesk Agent: Answers common IT questions and raises Jira tickets for unresolved issues.

HR Policy Assistant: Guides employees through leave, benefits, and compliance policies.

Onboarding Buddy: Helps new joiners navigate tools, processes, and policies on Day 1.

Sprint Summary Bot: Generates sprint summaries and status updates for scrum teams.Customer Support Tier 1 Agent: Handles common customer queries and escalates complex ones to the right team.

Tips for Making Your Agent More Effective

~ Keep your Confluence content up to date. Your agent is only as accurate as its source material.

~ Be specific in your instructions. Vague instructions lead to vague answers. The more precise, the better.

~ Gather user feedback. Ask your team what questions the agent answered well and where it fell short.

~ Iterate regularly. Rovo Agents are not a set-and-forget tool. Schedule monthly reviews to update content and refine instructions.

~ Start small, scale fast. Begin with one focused use case, validate it, then replicate the model for other departments.

You’re Now Ready to Build Your First Rovo Agent

Building a Rovo Agent doesn’t require a developer, a budget for consultants, or weeks of planning. In under an hour, you can deploy an AI-powered assistant that genuinely improves how your team works.

Whether you’re in IT, HR, DevOps, or customer support – there’s a Rovo Agent waiting to save your team hours every week.

At Canarys Automations, we help organizations design and implement Rovo Agents tailored to your workflows, teams, and Atlassian ecosystem. If you’d like expert guidance on setting up your first agent – or building a full library of them — reach out to our team today.

About Canarys Automations

Canarys Automations Limited is a GitHub Strategic Partner and ranked #1 globally and #1 in Asia for GitHub Certifications. We help enterprises adopt, optimize, and scale their DevOps, ITSM, and Atlassian ecosystems with proven expertise and deep platform knowledge.