What's New in Rovo Agents: Your Bi-Monthly Update
Curious about what the Rovo Agents team has been up to? You’re in the right place! Every two months, we’ll highlight the latest features, key updates, and give you a preview of what’s coming next.
🚀 Major Releases
Deep Research in Rovo Agents: Tailored Insights, Your Way
Your custom agents can now go beyond quick answers and deliver comprehensive, well-structured research reports - tailored to your needs, your knowledge sources, and your workflow.
What's New:
-
Customizable Output: Format and present your research results whatever way makes sense for you. That includes battle cards, summaries, detailed reports, etc.
-
Source Control: Deep Research only pulls from the knowledge sources you specify in your scenario - no more noisy or irrelevant results.
-
Workflow Integration: Trigger Deep Research as part of automated workflows or scheduled processes.
-
Scenario-Driven Triggers: Enable Deep Research for specific custom scenarios, so you control exactly when and how research is performed.
How It Works:
-
In Rovo Studio, toggle Deep Research on for any custom scenario and configure your instructions and knowledge sources
Deep Research configuration in your custom Scenario
2. When triggered, your agent plans, searches, and synthesizes information using only the sources you've allowed
3. The agent applies your formatting instructions, delivering a research report that fits your workflow
Format report with your scenario instructions
Using only the confluence pages under the configured knowledge
Good to Know:
-
Available for custom scenarios only (not the default scenario)
-
Each user can make up to 30 Deep Research requests per day
-
Automation timeout is 15 minutes (we're actively working to improve this)
Read more here: https://community.atlassian.com/forums/Rovo-articles/Deep-Research-in-Rovo-Agents-Tailored-Insights-Your-Way/ba-p/3167044
Agent Jira Issue Create & Update Skill Improvements
Rovo Agents can now handle complex Jira workflows with support for custom fields, required fields, and parent/epic relationships.
What's New:
-
Custom Field Support: Agents can now populate custom fields unique to your projects - from team assignments to story point estimates
-
Required Field Handling: If a required field isn't set, the agent will prompt you to provide it or attempt to set an appropriate value itself
-
Epic & Parent Linking: Create issues that are automatically linked to their parent epic
-
Bulk Operations: Create or update multiple issues at once, with a clear UI showing what's being created or changed
Real-World Example:
Ask your agent to "“Use this page as context and create the tickets outlined here in the epic Agent Recommendations Microservice. For each ticket, set the appropriate priority based on the description and add the label rovo-agent." The agent will walk you through each step, handle the required fields, link everything up, and show you exactly what's being created or changed before you confirm.
Agent to Read Confluence Attachment
Agents can now read and summarize PDF attachments directly from Confluence pages.
What's New:
-
PDF Content Access: When you point an agent to a Confluence page with PDF attachments, it can now retrieve and read the content of those PDFs
-
Summarization: Ask your agent to summarize PDF attachments and get key information without having to download and read them yourself
Simply ask your agent to read or summarize the attachments on a Confluence page, and it'll handle the rest.
Web Search for Agents
You can now add web search as a knowledge source for your agent scenarios, allowing your agents to search the internet directly for real-time or external information.
How It Works:
- In your agent scenario settings, toggle on web search as a knowledge source alongside your other sources
- Your agent will go directly to the internet when it needs information not available in your configured knowledge sources
- Citations are included so you can see exactly where the information came from
Example Use Case:
Build a legal or compliance agent that can look up current regulations, tenancy rights, or other publicly available information - without needing to manually add and maintain that content in your knowledge base.
🔮 Coming Soon: Sneak Peeks
Agent Verification
Org admins will soon be able to mark agents as verified, displaying a badge that signals the agent has been reviewed and approved by your organization. This helps drive adoption of the right agents and gives users confidence they're interacting with trusted tools.
Follow along here: https://jira.atlassian.com/browse/ROVO-146
Agent Use Permissions
You'll be able to restrict an agent so that only specific users can find and chat with it - useful for agents that contain sensitive information or are designed for specific teams.
Follow along here: https://jira.atlassian.com/browse/ROVO-142
Agent Evaluations
We're developing evaluation tools that let you test your agents against a set of prompts and review how well they perform - before deploying changes to your users.
What's Coming:
-
Upload a dataset of test questions for your agent
-
Run evaluations and see responses
-
Drill into individual responses to understand how your agent performed
-
Export results for further analysis
Follow along here: https://jira.atlassian.com/browse/ROVO-296
Agent Drafts and Versions
We’re working on versioning and draft support for Rovo agents so you can ship changes with more confidence and control.
What's Coming:
-
Draft mode for agent changes – Safely edit an existing agent without immediately impacting end users.
-
Test in draft – Try out and refine your agent in draft before publishing any changes to your team.
-
Versioned agent releases – Each time you publish a draft, Rovo will create a new version of the agent.
-
Easy rollbacks – If a new version doesn’t perform as expected, you’ll be able to roll back to a previous version in just a few clicks.
Follow along here: https://jira.atlassian.com/browse/ROVO-93
Skill Invocation Latency in Debug Modal
The Debug Modal will soon show detailed timing information for each skill your agent invokes, helping you identify performance bottlenecks and pinpoint whether slowness is coming from specific skills or other parts of the agent workflow.
💬 We Want to Hear From You
These features are built based on your feedback. Have ideas for what's next? Questions about how to implement these new capabilities? Reach out to your Atlassian team or visit our documentation to learn more.
Stay tuned for our next bi-monthly update - we've got even more exciting features in the pipeline!
Was this helpful?
Thanks!
Rachelle Rathbone
2 comments