Atlassian has launched Jira Planner in Early Access, a tool designed to bridge the gap between deciding what to build and having AI agents execute it. It turns rough ideas into structured, agent-ready specs by asking clarifying questions, grounding plans in a team's actual codebase and prior decisions via Teamwork Graph, and producing editable Confluence Live Docs. Teams can co-edit plans, get readiness checks for ambiguity, and convert finalized plans into sequenced Jira work items with acceptance criteria for agent orchestration to pick up. It targets tech leads, senior engineers, engineering managers, and PMs working on multi-sprint features or architectural decisions. Early Access requires Jira Cloud with Rovo enabled, Teamwork Graph connected, and Confluence connected.
Questions this post answers
What is Jira Planner and what does it do?
Jira Planner is an Atlassian tool, now in Early Access, that turns rough ideas into structured, agent-ready specs grounded in a team's codebase, standards, and prior decisions. It asks clarifying questions about scope and constraints, generates editable Confluence Live Docs connected to Jira, checks whether a plan is detailed enough for execution, and breaks finalized plans into sequenced Jira work items with acceptance criteria for agent orchestration. Teams weighing spec-driven workflows for AI agents can track tool launches like this via daily.dev.
What are the requirements to access Jira Planner in Early Access?
Jira Planner requires a Jira Cloud plan with Rovo enabled and Teamwork Graph connected, plus a connected Confluence instance since plans publish as Confluence Live Docs. Access is rolling out in waves to teams willing to give feedback that will shape the product's future direction. Engineers evaluating new Jira features can follow rollout details like this on daily.dev.