Atlassian has launched an open beta of AI agents in Jira, enhancing task assignment, deadlines, and progress tracking within the same interface as human users. This update introduces the Rovo Model Context Protocol (MCP) Server, which offers a secure connection for MCP-compatible AI clients to Jira and Confluence. Crucially, the governance design enables AI agents to work within existing permission structures, addressing compliance concerns from security and legal teams by ensuring all actions are logged alongside human work history. With enterprises accounting for nearly 50% of Rovo MCP Server usage, the new integration layer is now generally available, operating over secure HTTPS and utilizing OAuth 2.1 for authentication. Notable compatible AI clients include Claude by Anthropic and Google’s Gemini CLI. This strategic positioning differentiates Atlassian from competitors like Salesforce and ServiceNow, emphasizing its unique approach to integrating AI into project management while maintaining strict control over data and access.
Source link
