Atlassian’s Rovo generative AI (genAI) assistant, which launched as a preview release in May, is now available for users.
The company bills Rovo as an “AI-powered” knowledge discovery tool that lets teams access information across their organization more efficiently. Thousands of Atlassian customers and partners have been testing Rovo in a closed beta for the past five months or so, the company said in a blog post Wednesday; early user feedback indicated it has helped save one to two hours each week.
Pricing for Rovo access starts at $24 per user each month when billed monthly.
There are three main elements to the genAI tool: Search, Chat, and Agents.
The search tool, based on an organization’s “teamwork graph,” helps users surface information from sources such as documents in Atlassian apps like Jira and Confluence. Rovo Search also can be connected to third-party apps such as Slack, Google Drive and Microsoft SharePoint.
Rovo Chat lets users ask questions about information held in these files via a natural language interface. This is available via a side panel in Atlassian apps, as well as via a browser extension.
Finally, there are genAI agents that function as virtual “teammates.”
The agents can be designed to generate and review marketing content, for instance, collate feedback from various sources, or streamline processes such as clearing up Jira backlogs and organizing Confluence pages. Users can also create their own Rovo Agents using a no-code text interface or Atlassian’s Forge app development platform. More than 20 pre-built agents are already available.
Atlassian also unveiled new developer-focused AI Agents that help perform a variety of tasks beyond writing code. These can generate code plans and recommendations, and pull requests in Jira based on task descriptions, requirements, and context across the organization.
Other features include the ability to “fast track” pull request reviews, with automatic code analysis and recommendations for improvements.
The AI Agents for developers are currently in preview via an early access program.